Journal of the Society of Motion Picture and Television Engineers

From the collection of the

j d

7t "^ *M^

m

o Prejinger h
v Jjibrary
t p

San Francisco, California
2007

JOURNAL OF THE
SOCIETY OF

MOTION PICTURE

AND

TELEVISION

ENGINEERS

THIS ISSUE IN TWO PARTS
Part I, June 1952 Journal • Part II, Index to Vol. 58

VOLUME 58
January — June 1952

SOCIETY OF MOTION PICTURE
AND TELEVISION ENGINEERS

40 West 40th St., New York 18

CONTENTS —Journal

Society of Motion Picture and Television Engineers

Volume 58 : January — June 1952

Listed below are only the papers and major reports from the six issues. See the
Volume Index for those items which generally appear on the last few pages of each
issue: Standards, Society announcements (awards, Board meetings, committee
reports, conventions, engineering activities, membership, nominations, section
activities), book reviews, current literature, letters to the Editor, new products and
obituaries.

January

Continuous Motion Picture Projector for Use in Television Film Scan-
ning A. G. JENSEN, R. E. GRAHAM and C. F. MATTKE

Color Television Reproducers HARRY R. LUBCKE

Film-Spool Drive With Torque Motors A. L. HOLGOMB

Heat-Transmitting Mirror ... G. L. DIMMIGK and M. E. WIDDOP

Recent Improvements in Silencing Engine-Driven Generators . . .

L. D. GRIGNON 43

Cinecolor Multilayer Color Developing Machine

J. W. KAYLOR and A. V. PESEK 53

New Magnetic-Recording Head ........ MARVIN CAMRAS 61

Push-Pull Direct-Positive Recording — An Auxiliary to Magnetic
Recording L. I. CAREY and FRANK MORAN 67

February

Factors Affecting the Quality of Kinerecording

P. J. HERBST, R. O. DREW and J. M. BRUMBAUGH 85

Multichannel Magnetic Film Recording and Reproducing Unit . .

C. C. DAVIS, J. G. FRAYNE and E. W. TEMPLIN 105

Magnetic Sound Track Placement

LOREN L. RYDER and BRUCE H. DENNEY 119

ii Contents: Journal of the SMPTE Vol. 58

New Principle for Electronic Volume Compression

HAROLD E. HAYNES 137

Prints from 1 6mm Originals

R. L. SUTTON, K. B. CURTIS and LLOYD THOMPSON 145

High-Constant-Speed Rotating Mirror

J. W. BEAMS, E. C. SMITH and J. M. WATKINS 159

Report of SMPTE Standards Committee . . . FRANK E. CARLSON 169

March

Image Gradation, Graininess and Sharpness in Television and Motion
Picture Systems — Part II: The Grain Structure of Motion Picture
Images — An Analysis of Deviations and Fluctuations of the Sample
Number OTTO H. SCHADE 181

Color Negative and Color Positive Film for Motion Picture Use . .

W. T. HANSON, JR. 223

Printer Control in Color Printing C. A. HORTON 239

Desirable Characteristics of 16mm Entertainment Film for Naval Use .

LOWELL O. ORR and PHILIP M. COWETT 245

High-Speed Motion Picture Cameras From France

PAUL M. GUNZBOURG 259

April

The Nature and Evaluation of the Sharpness of Photographic Images .

G. C. HIGGINS and L. A. JONES 277

Progress in Three-Dimensional Films at the Festival of Britain . . .

RAYMOND SPOTTISWOODE 291

The Cash Customers at the Festival of Britain Telecinema

NORMAN JENKINS 304

Optical-Magnetic Sound 16mm Projector

G. A. DEL VALLE and F. L. PUTZRATH 312

Twin-Drum Film-Drive Filter System for Magnetic Recorder-Re-
producer CARL E. HITTLE 323

A Technical Solution of Magnetic Recording Cost Reduction . . .

KURT SINGER and H. CONNELL WARD 329

Contents: Journal of the SMPTE Vol. 58 iii

May

Techniques for Effective High-Speed Photography and Analysis . .

RICHARD O. PAINTER 373

A Direct-Projection System for Theater Television . F. N. GILLETTE 385
Progress Committee Report C. W. HANDLE Y 397

Magnetic Print-Through — Its Measurement and Reduction ....

LYMAN J. WIGGIN 410

A Scientific Approach to Informational-Instructional Film Production
and Utilization .... C. R. CARPENTER and L. P. GREENHILL 415

Film Production Principles — The Subject of Research

KEN KENDALL 428

Audio- Visual Instruction Conference . D. F. LYMAN 445

Television Studio Lighting Committee Report . . RICHARD BLOUNT 450

June

The Ansco Color Negative-Positive Process . . HERMAN H. DUERR 465

Multiple-Image Silhouette Photography for the NOTS Aeroballistics
Laboratory ERNEST C. BARKOFSKY 480

Optical Problems in High-Speed Camera Design . . JOHN C. KUDAR 487

Effective Sum of Multiple Echoes in Television

A. D. FOWLER and H. N. CHRISTOPHER 491

The Image Iconoscope — A Camera Tube for Television (Abstracted
by Pierre Mertz) . . P. SCHAGEN, H. BRUINING and J. C. FRANCKEN 501

TelePrompter — New Production Tool

FRED BARTON and H. J. SCHLAFLY 515

The Synchro-screen as a Stage Setting for Motion Picture Presentation

B. SCHLANGER, W. A. HoFFBERG and C. R. UNDERHILL, JR. 522

Resolution Test Chart of the Motion Picture Research Council . . .

ARMIN J. HILL 529

Laboratory Practice Committee Report JOHN G. STOTT 531

iv Contents: Journal of the SMPTE Vol. 58

Continuous Motion Picture Projector for Use
in Television Film Scanning

By A. G. JENSEN, R. E. GRAHAM and C. F. MATTKE

The projector used for this equipment drives a 35mm motion picture film
at the standard (nonintermittent) speed of 24 frame/sec and produces a
television signal of 525 lines and 30 frames interlaced 2 to 1. The projector
utilizes a system of movable plane mirrors mounted on a rotating drum and
controlled by a single stationary cam. Vertical jitter in the television image
is minimized by means of an electronic servo system operating on the film
sprocket holes, resulting in a residual vertical motion of about 1/2000 of a
picture height. A second electronic servo system is incorporated to suppress
flicker. The combination of this scanner and a high-grade monitor is capable
of producing a television picture with a resolution corresponding to about 8
me and with good tone rendition over a range up to 200 to 1.

-L HE PROBLEM of designing a motion
picture projector, in which the film
motion is continuous, has occupied
inventors and designers almost since
motion pictures first made their appear-
ance. In the early days of motion
pictures the need for a continuous pro-
jector stemmed largely from a desire to
decrease the wear and tear suffered by
the film in the intermittent projector.
Later on, with the advent of sound
pictures, it was felt that a continuous
projector could fit in better with a
machine in which the film had to move
continuously through the soundhead.
Many different types have been pro-

Presented on October 15, 1951, at the
Society's Convention at Hollywood, by
A. G. Jensen, R. E. Graham and G. F.
Mattke, Bell Telephone Laboratories,
Murray Hill, NJ.

posed and patented but very few of them
have gone beyond the experimental
stage. A measure of the interest in this
problem may be obtained from the bib-
liography at the end of this paper in
which are listed the more important
papers published on the subject during
the years 1920-1945.

One particular type of continuous
projector, the Mechau projector, did
reach the commercial stage and was used
in a limited number of German motion
picture theaters in the 1930's.1 This
projector used eight movable mirrors,
the motion of each mirror being con-

*L. Burmester und E. Mechau, "Unter-
suchung der mechanischen und optischen
Grundlagen des Mechau-Projektors," Die
Kinotechnik, 70: 395-401, 423-426 and
447-451, Aug. 5, Aug. 20 and Sept. 5,
1928.

January 1952 Journal of the SMPTE Vol. 58

trolled by its own individual cam of
rather intricate design. The mechanical
portion of this machine is rather com-
plicated and expensive and is difficult to
keep in good running order. However,
the machine has high light efficiency
and, when properly serviced, does
produce high-quality pictures.

With the coming of television the need
for a satisfactory continuous projector
again became apparent. Such a pro-
jector would lend itself admirably to
translating the. 24-frame/sec film picture
into a 30-frame/sec television picture
as called for by the present television
broadcasting standards. In a continu-
ous projector the motion of the film
frames is in effect frozen in some image
plane, and in this image plane it is then
possible to scan the picture 30 times per
second, or at any other desired rate,
synchronous or nonsynchronous, for
that matter. The British Broadcasting
Company realized this many years ago
and installed a German Mechau pro-
jector as a film scanner in their Alexandra
Palace studio. One or more of these
Mechaus are still being used for that
purpose in London.

In the U.S.A. present commercial
film scanners use a 24-frame/sec inter-
mittent drive in combination with a
storage-type camera tube such as an
iconoscope. Very short, intense light
pulses are flashed through the film
frame onto the storage mosaic, which
is then scanned between light flashes
at 60 fields per second, interlaced, mak-
ing 30 complete television pictures per
second. Thus every other film frame is
scanned twice, and the remaining frames,
three times.

Unfortunately the iconoscope does
not have a very good contrast range
and inherent in the storage action are
certain spurious signals which must be
eliminated by introducing so-called shad-
ing or compensating signals, a fact which
further tends to degrade the contrast.
The result is that presently produced
television signals from motion picture

film are generally not as satisfactory as
good direct pickup pictures.

In the Bell Telephone Laboratories
there has been a need for high-grade
television signals ever since the first
development of wide-band television
transmission facilities around 1935.
Such signals are needed for test purposes
and for determining the fundamental
transmission requirements for compo-
nents of wide-band circuits such as the
coaxial cable and the microwave link.
For this purpose several film scanners
have been developed in the past.

The first of these was used to demon-
strate the transmission of 240-line, 24-
frame television signals over the early
New York-Philadelphia coaxial cable
in 1937.2 It was a mechanical scanner
using for the scanning unit a 6-ft disk
with 240 lenses mounted along the
periphery and rotating at 1440 rpm.

As the requirements for good definition
went up, mechanical scanners became
impractical, and a new electronic film
scanner was developed and first used in
the transmission of 441 -line, 30-frame
(60 fields interlaced) television signals
over the later New York-Philadelphia
coaxial cable circuit in 1941.3 This
film scanner employs specially prepared
60-frame/sec motion picture film, con-
tinuous film motion, and a Farnsworth
dissector tube, which is a nonstorage
device. The continuous motion of the
film furnishes the vertical scan for the
pickup so that only a horizontal scan
is required of the dissector tube. The
equipment has since been redesigned to
produce 525-line, 30-frame pictures as
presently standardized, and was used
for transmitting television signals over
the New York-Boston microwave relay
in 1947. The pictures from this scanner
show very good detail and a wide range

2 M. E. Strieby, "Coaxial cable system for
television transmission," Bell System Tech.
F., 17: 438-457, July 1938.

3 A. G. Jensen, "Film scanner for use in
television transmission tests," Proc. IRE,
29: 243-249, May 1941.

January 1952 Journal of the SMPTE Vol. 58

Fig. 1. Basic principle of operation of continuous projector.

of contrast, and the signals are still being
^used for test purposes in the laboratory.
The chief shortcoming of this film scan-
ner is the inconvenience and high cost
of preparing the special 60-frame/sec
film.

In order to obtain a wider range of
picture material for test purposes, it was
decided therefore to develop a continuous
projector film scanner capable of using
standard 24-frame/sec motion picture
film. The design of such a scanner was
made more feasible by the development
of cathode-ray tube spot scanners with
very short decay phosphors. Tubes of
this type had been used with photo-
multipliers to produce television signals
from still slides, and the resulting pic-
tures showed very good resolution and a
wide range of contrast.

In designing the optical system of this

projector it was decided, as in the
Mechau projector, to use a moving
mirror system, since systems involving
such mirror optics appear to have the
best light efficiency, and freedom from
certain refractive optics limitations. The
design as evolved greatly simplifies the
mechanical construction and operation
by controlling all the mirrors from one
simple stationary cam. During the
development of the machine, further
features were incorporated, such as an
electrooptical servo system to eliminate
picture jitter due to nonuniform film
motion, and a second servo system to
eliminate flicker due to nonuniform
light efficiency through the frame cycle.
The result is a laboratory model of a
film scanner which is now being used for
producing test signals and which is
described in detail below.

Jensen, Graham and Mattke: Television Film Projector

CENTER LINE
OF FILM FRAME

APPARENT POSITION
OF FILM FRAME

Fig. 2. Geometric relationship between film and moving mirror
in continuous projector.

Fundamental Principles

The basic principle of operation of the
machine used as an optical projector is
shown on Fig. 1. Film 3 is moved at
a uniform rate by sprocket 4 down over
curved gate 9. Light from lamp 1
passing through condensing lens 2 and
film 3 is reflected by compensating mirror
5 through objective lens 6 and reflected
from fixed mirror 7 to screen 8. As
sprocket 4 is rotated to move film 3,
mirror 5 is caused to rotate about axis
1 0 by cam 1 1 . The amount of rotation
of mirror 5 is such that the image of the
film on the screen produced by lens 6
remains stationary.

The geometric relation between film 3
and the mirror 5 is shown in Fig. 2.
Consider the horizontal line CO passing
through the center of the aperture in
gate 9 and the center of curvature O
of gate 9, as a fixed horizontal optical
axis; also the radial lines aO and bO
passing through the centers of two ad-
jacent film frames a and b of film 3 and
point O to form an angle a. Also
consider line dO as a fixed optical axis
passing through point O and the nodal
point d of objective lens 6. Finally
consider the reflecting surface 5 of a
mirror pivoted about point O.

The requirement for optical compensa-
tion of the moving film is as follows:

January 1952 Journal of the SMPTE Vol. 58

CONDENSING
LENS '

FIXED
, MIRROR

FILM / PROJECTION

LENS

VIEWING
SCREEN

PROJECTION
LAMP

¥

FILM

FILM DRIVE

Fig. 3. Schematic diagram of mirror drum arrangement.

When a ray of light aO moves through
the angle a to bO, the direction of the
reflected ray Od must remain stationary.

This is accomplished by the rotation
of mirror 5 about point O through an
angle a/2 while the film moves through
the angle a.

Optically speaking, during the motion
of the film through the gate, frame a
appears stationary to lens 6, i.e., the
apparent position of the film frame has
not changed, as indicated on the figure.

The ratchet action suggested in Fig. 1
is obviously not suitable for any practical
working mechanism. Therefore, for
continuity of projection, the action of
mirror 5 is made repetitive by using a
suitable number of axially mounted
mirrors equally spaced in a circle to
form a sort of drum, the axes of the
mirrors lying in the plane of the mirror
reflecting surfaces and all parallel to
the axis of the drum. As the drum ro-
tates, the mirrors are then made to
rotate at the required rate about their
axes, by means of a suitable cam action.
Figure 3 shows a schematic diagram of
the mechanism. The mirror drum is
geared directly to the film-drive sprocket,
and as the drum rotates the individual

mirrors are rotated through the required
angle by means of cam followers rolling
on a common stationary cam.

The continuity of the action of the
mirrors is shown in Figs. 4a and 4b,
where 4a shows one mirror at the middle
of its compensation cycle and 4b shows
two adjacent mirrors at the extremities
of their compensating cycle. The axes
e, h and j (perpendicular to the plane of
the diagram) of the three mirrors shown
are located on the arc of a circle with
center of rotation at point d, the axis
of the drum. Line Of is a diameter of
the circle and lines de, dh and dj are
radii to the mirror axes e, h and j.
If the angles hde and hdj are both equal
to |8 then the geometry of the system
makes the angles efh and jfh both equal
to |8/2. Also lines drawn through e,
h or j, perpendicular, respectively, to ef,
hf and jf, will all pass through point O.

Referring to Fig. 2 it is seen that the
angle a of the arc subtended by a film
frame is equal to the angle ft of the
corresponding rotation of the mirror
drum. It follows, therefore, that for
proper compensation of film motion, the
reflecting planes of the mirrors must at
all times contain the respective per-

Jensen, Graham and Mattke: Television Film Projector

(a)

Fig. 4. Diagram illustrating proper motion of mirror through active com-
pensation cycle, (a) One mirror at the middle of cycle; (b) two adjacent
mirrors at overlapping part of cycle.

pendicular through e, h or j. In other
words, during the active cycle the re-
flecting plane of a mirror must at all
times contain the point O.

As the drum rotates and the mirrors
obey the geometric principles just out-
lined, the reflected ray Od remains
stationary while the film moves from
a to b through angle «. Figure 4a
shows the mirror in a position when
frame b is at the center of the aperture
in the gate, while Fig. 4b shows the
mirror positions when the frame is at
the limit of its travel in the gate.

Film Shrinkage

So far in this discussion, the considera-
tion of the principles involved in the
mechanism has been theoretical. It
may be assumed that the mechanical
parts can be made and assembled with
the degree of accuracy necessary to
satisfy the geometrical requirements for
successful performance. On the other
hand, the film is a plastic and therefore
is mechanically unstable throughout its
useful life. This instability must be
considered in the design and suitable
adjustment provided.

Standard 35mm motion picture film
has a nominal frame pitch dimension of
0.748 in.; normal shrinkage, however,
changes this value. In this projector
design the longitudinal shrinkage re-
quires consideration since its effect
manifests itself at the curved gate by
altering the angle subtended by a frame.

Referring to Fig. 2, it will be re-
membered that one picture frame (or
really one frame pitch) in the curved
gate should subtend an angle a = /3
for proper operation. As the film
shrinks this angle a decreases and no
longer corresponds to the associated
mirror drum angular rotation of /3,
thus resulting in a frame-to-frame jitter
of the projected picture. Exact com-
pensation for shrinkage would require
that the curvature of the film gate be
increased, but for normal shrinkage it
has been found adequate simply to
move the gate a little closer to the
mirror drum until a film frame again
subtends the proper angle. Such an
adjustment is provided for in the
machine, together with a corresponding
focusing adjustment of the projection
lens.

January 1952 Journal of the SMPTE Vol. 58

Mechanical Specifications

So far the machine has been described
in terms of a conventional optical pro-
jector and as such is shown in the
schematic diagram of Fig. 3. In order
to convert it into a film scanner all that
is necessary is to replace the viewing
screen in Fig. 3 by a cathode-ray
spot scanning tube and to replace the
projection lamp by a photomultiplier
tube. As long as the projection lens
is such as to provide the proper reduction
from spot scanner raster size to film
frame size, the remaining components of
the machine are unchanged.

The geometric relations of these
components are established by the choice
of 18 compensating mirrors in the drum
and an 8-frame drive sprocket as follows:

1. Angle a equal to frame

pitch on the curved gate . 20 °

2. Angular separation ft be-
tween adjacent mirror axes

on the drum 20 °

3. Radius R of curved gate . 2.142 in.

4. Angle of rotation of the
mirrors about their axes
while traversing the active

arc 10°

5. Rotational speed of drum . 80 rpm

6. Number of mirrors in
drum 18

7. Number of teeth on drive
sprocket . 32

8. Number of frames per
revolution of drive sprocket 8

Constructional Details

Film drive. The drive is of more or less
conventional design incorporating the
usual feed sprockets, idlers and drive
sprocket. A friction-controlled film-
tensioning sprocket immediately above
the gate serves to keep the film taut
during its passage through the gate,
in order to insure proper radius of
curvature while it is being scanned.

Mirror assembly. The outer diameter
of the mirror drum is about 14 in. and

it has 18 equally spaced mirror units
mounted along the periphery on a
diameter of 11^ in. The general ar-
rangement may be seen from the photo-
graphs in Fig. 5 and Fig. 6.

An individual mirror unit as shown
in Fig. 7 consists of bearing housing 1,
mirror support casting 2, shaft 3, mirror
4, bearings 5, bearing spacer 6, lock
nut 7, mounting screws 8 and mirror
setscrews 9.

The bearing housing is accurately
machined to fit the holes in the drum.
In order to maintain the necessary loca-
tional accuracy and allow easy as-
sembly, the finished bearing cylinders are
etched except for four contact areas as
shown in Fig. 7. These areas are
located about the threaded clamp screw
hole, and when in place are the only
surfaces that are in contact with the
drum bore.

The shaft 3 is mounted in the shell
on two ball bearings, preloaded by the
proper adjustment of the spacer 6 to
eliminate any radial motion. The mir-
ror support casting 2 is mounted on the
flange of the shaft. The contact face of
the flange is machined after assembly
in the shell to insure an accurate align-
ment of the casting 2. The correspond-
ing face of the casting 2 is also machined
on a special fixture to insure an accurate
90° angle between the mirror face and
the flange face.

The casting 2 is designed with ma-
chined pads with steel ball inserts,
which support and locate the mirror
at the reflecting surface, the mirrors
being of the front-surface type.

The mirrors are made of glass 2 X
3 X i in. thick, and the surface is flat
to one wavelength in visible light.

Cam Follower. The cam follower and
adjusting detail are mounted on the
cylindrical end of the flanged mirror
shaft as shown on Fig. 7. This mech-
anism consists of roller 10, roller
shaft and nut 11, follower casting 12,
shaft adjusting casting 13, tension

Jensen, Graham and Mattke: Television Film Projector

Fig. 5. Photograph of mechanism showing moving mirror drum.
January 1952 Journal of the SMPTE Vol. 58

Fig. 6. Photograph of mechanism showing stationary cam and cam followers.

Fig. 7. Mechanical details of drum mirror assembly.
Jensen, Graham and Mattke: Television Film Projector

spring 14, spring pin 15, spring stud 16,
adjusting pressure spring 17, adjusting
screw 18, and setscrew 19.

Spring 14 maintains contact of roller
10 on the cam. Stud 16, to which one
end of spring 14 is fastened, is fixed to
the body of the drum; the other end of
the spring is anchored to the roller
casting by pin 15. Spring 17 main-
tains loading on the adjusting screw 18,
which, when turned, changes the angular
setting between the follower casting 12
and the mirror shaft 3, thereby allowing
the relative mirror angle to be changed.

A photograph of a mirror unit and cam
follower is shown in Fig. 8.

Cam. The design of the stationary
cam was considered from two require-
ments: (a) the optical performance of
the mirrors and (b) the dynamic balance
of the drum. For the optical perform-
ance a 40° segment of the cam is all
that is necessary. It is this segment that
produces the compensation; the re-
mainder is used simply to return the
cam follower roller to the beginning of
the segment. Dynamically, however,
in order to eliminate any unbalance in
the rotating system (drum), duplicate
cam segments must be located diametri-
cally opposite each other. In this manner
the radial distance of opposite mirror
mounting castings and associated cam
follower mechanisms will always be
the same and thereby provide the
necessary counterbalance. The final
shape of the cam is indicated in Fig. 3
and a portion of the cam may be seen
in Fig. 6. A feature of the cam curve is
that it can be generated quite easily
with a grinding fixture constructed for
the purpose. A schematic diagram of
the fixture is shown in Fig. 9. In this
figure point B represents the axis of
rotation of the mirror drum and the
circle D represents the location of the
axes of rotation of the individual mirrors.
Point A corresponds to the point O in
Figs. 2 and 4, and point C corresponds
to point f in Fig. 4. The fixture consists

of two movable arms 1 and 2, inter-
linked at point E. Arm 1 rotates about
point B and arm 2 rotates and slides
about point C. The grinder is fixed
on arm 2 in such a manner that the
center of the grinding wheel is located
at point F, where the distance EF is
equal to the center-to-center length of
the cam follower arm. The diameter
of the grinding wheel is equal to the
diameter of the cam roller. A photo-
graph of the grinding fixture is shown in
Fig. 10.

Position Control

•The fundamental principles of opera-
tion of this machine were discussed on
page 5. If these are fulfilled, if the
gears are perfect, without backlash and
with correct teeth profile, if the cam has
the correct shape and the cam followers
are correctly aligned, and above all,
if the friction is constant so the film moves
at an absolutely uniform rate, then the
images on the screen of succeeding film
frames will fall exactly on top of each
other. The image of each frame will
lap dissolve into the image of the
previous one without blurring and with-
out loss of registration. In other words,
the picture on the screen will be steady
without any vertical jitter. Conversely,
if the machine is used for film scanning,
the image of the scanning raster on the
film surface will move in such a manner
as to remain stationary with respect to
the film frames and no vertical motion
or jitter will be observed in the resulting
picture on a television receiving tube.*

The present machine does not perform
in this perfect manner. It is assumed
that the friction in the film drive is
not constant, but whatever the cause,
the fact is that without any further
control, the image moves up and down
erratically with a maximum excursion of

*The discussion above relates only to
vertical motion of the image. It is as-
sumed that adequate guides in the film
drive prevent any sideways weave of the
image.

10

January 1952 Journal of the SMPTE Vol. 58

Fig. 8. Photograph of drum mirror assembly.

Fig. 9. Diagram of

cam grinding jig. ~ /\

Fig. 10. Photograph of cam grinding jig.

Jensen, Graham and Mattke: Television Film Projector

11

PICTURE TUBE USED
AS LIGHT SOURCE

PHOTOTUBES

TO PHOTO

MULTIPLIER

TUBE PICKUP

Fig. 11. Schematic diagram of jitter correcting servo system.

about 1/100 of the picture height.
This is, of course, disturbing, and would
be intolerable in a commercial film
scanner. It was decided, therefore, to
attempt to eliminate this vertical jitter,
not by perfecting the mechanical pre-
cision of the component parts, since such
perfection would probably require a
continuous, time-consuming main-
tenance effort, but rather by auto-
matically monitoring the departure from
uniformity in film motion, and using the
indications of such departures to control
some element of the system in such a
way as to counteract or nullify the
vertical jitter.

In the entire processing of motion
picture film from camera to projector,
the primary standard of registration is
the location of the sprocket holes in the
film. It is natural, therefore, to use
these sprocket holes as a means for

measuring the departure from proper
motion of the film. Assuming that such
"error" information is available, the next
question is where to apply it to compen-
sate for the error. In Fig. 3 there is
shown a fixed mirror for deflecting the
light from the projection lens onto the
viewing screen. If this mirror is made
adjustable around an axis in the plane of
the mirror and perpendicular to the
plane of the diagram, then such an
adjustment would impart a vertical
motion to the image on the screen. In
other words, if the "error" signal ob-
tained from monitoring the sprocket-
hole position is used to tilt the mirror
in such a way as to counteract the error,
then the image on the screen will stay
still in spite of nonuniform film motion.
This is exactly what is done by the
jitter-correcting control circuit, or servo,
incorporated in the machine, and the

12

January 1952 Journal of the SMPTE VoL 58

method employed is shown by the
schematic diagram in Fig. 11.

This figure shows the essential features
of the mechanism used as a film scanner.
The light from the raster of the spot
scanning tube is transmitted via the
correcting mirror through the projection
lens, and via the drum mirror through
the film onto the cathode of a photo-
multiplier tube.

An auxiliary light path through the
optical system is provided as follows:
The sprocket-hole area of the film is
illuminated by light from a small
incandescent lamp passing through a
right-angle prism mounted adjacent
to the film gate (see Fig. 11). As far
as this sprocket-hole area is concerned
the machine now functions as an optical
projector. The reflected light from the
film surface is passed back through the
system as indicated in Fig. 11 and an
image of the sprocket-hole area is formed
in a vertical plane marked "slits" in
the figure. A picture of this image is
shown as an insert in Fig. 11. The
sprocket holes themselves will appear
black in this image, while the film area
around the sprocket holes will show
uniform illumination.

In this image plane there is placed an
opaque mask with two narrow slits as
shown in the insert. The lower slit
covers part of the film image between
two sprocket holes and is used as a
reference source, while the upper slit
partly overlaps the image of the sprocket-
hole edge and is used as a control
source. By means of prisms the light
from the two slits is passed to two
separate photomultiplier tubes and the
electrical output from these is passed
through a differential amplifier to two
electromagnets controlling the position
of the correcting mirror.

The system is so adjusted that for the
reference position of the image as
shown in the diagram the output of
the two phototubes is the same. The
differential amplifier, therefore, passes
no current to the electromagnets and

the correcting mirror stays fixed. If,
on the other hand, a sudden perturbation
in the film motion causes the sprocket
hole image to move upwards, then the
output of the phototube corresponding
to the upper slit will increase. The
differential amplifier will then pass a
corresponding current to the electro-
magnets and tilt the correcting mirror
in such a direction as to restore the
sprocket-hole image and thus the main
image to its original position.

It is seen that this electrooptical
control system is indeed a servo or feed-
back system, in that it automatically
will tend to keep the "error" signal
small at all times. The gain of the
electrical part of the system must be
high enough to keep residual errors
down to a negligible amount, and the
frequency bandwidth of the system
must be sufficient to make the reaction
time short compared to the frequencies
of normal perturbations of film motion.
In the present system the loop gain is
about 50 db at low frequencies and
gradually decreases to zero gain at
about 250 cycle/sec. A measure of the
performance of the system may be had
by introducing a sudden electrical
disturbance into the circuit. With such
a disturbance introduced, the correcting
mirror will readjust itself in approxi-
mately one millisecond, without any
appreciable overshoot.

The mechanical construction of the
correcting mirror is shown in Fig. 12.
The glass mirror 1 is about 3X4^ in.,
fashioned from a plano-convex lens
with the plane surface polished flat to
about one wavelength of visible light.
The mirror is cemented to an aluminum
frame 2, which in turn is spring sup-
ported to the fixed frame 3. The sup-
porting springs are clamped in the
fixtures 4. The springs are 0.005 in.
thick, 0.03 in. wide and 0.005 in. long
between clamping points. The driving
electromagnets are shown at 5. The
peak-to-peak deflection of the correcting
mirror during normal operation is of

Jensen, Graham and Mattke: Television Film Projector

13

Fig. 12. Mechanical construction
of jitter-correcting mirror.

the order of 5 to 10 minutes of arc, and
the peak power required to drive the
mirror is less than one watt.

The sprocket-hole edge used for control
is the trailing edge in the passage through
the film drive, since this edge is not
subject to gradual deterioration due to
drive-sprocket pressure.

It may be asked why the servo system
uses light reflected from the film surface,
rather than light transmitted through the
film. The answer is that the reflection
coefficient of 'the film surface is prac-
tically independent of the transparency
of the film. If transmitted light were
used the control light would be affected
by the degree of exposure of the emulsion
around the sprocket holes, by surface
scratches in the film and, above all, by
the fact that some film manufacturers
print their firm name at frequent
intervals along this part of the film.

In this discussion of the position servo
system only nonuniformity of film motion

has been mentioned as a source of vertical
jitter. Other sources of jitter may be
present, such as gear teeth irregularities,
cam motion irregularities, optical mis-
alignment, etc. Since the servo system
in effect controls the position of the final
image, it will tend to minimize vertical
jitter due to any of these causes. Even
film shrinkage is to some extent com-
pensated for automatically by the servo.

Control of Illumination

As one mirror on the drum approaches
the end of its active cycle the light from
this mirror will gradually decrease, while
the light from the succeeding mirror
increases. In an ideal system these
opposite changes in light transmission
should exactly cancel each other, result-
ing in constant overall light efficiency
throughout the cycle. In the actual
machine this is not quite so. An analyti-
cal study involving ray tracing through
the cycle indicates that for the period
when two mirrors are contributing light
to the screen there is a small amount of
masking of the light falling on one mirror
by the edge of the previous mirror. Also
during this part of the cycle the pro-
jection lens is not entirely filled by light
from the two mirrors together.*

The result of this analysis is shown
in Fig. 13. It is seen that for about
three-quarters (15 degrees) of the active
cycle, one mirror, and therefore one
film frame, contributes more than 80%
to the illumination on the screen. For
the remainder of the cycle two adjacent
mirrors contribute to the illumination.
It is seen that for a small part of this
overlapping period the contribution
from one mirror (No. 2) falls off faster
than the contribution from the next
mirror (No. 3) increases. The result

* This study was made on the assumption
that the machine was used for optical
projections with uniform illumination of
the film gate. It is, of course, equally
valid when the machine is used as film
scanner with a cathode-tube raster of uni-
form illumination.

14

January 1952 Journal of the SMPTE Vol.58

100

90

80

70

60

50

~ 40

30

20

10

^

X

^"^TOTAL /
<-EFFICIENCY /

N

'

^

1

s

V
\

/
/
/

a = 0°

1.625" 0 LENS

\
\

1

\
\

1

\
\

/

\

I MIRROR 2

MIRROR 2\
V

/

/

\
\

/
/

\
\

1

/

/
/
/

^_

/
/

\
\

/

/
/

\ MIRROR 1

MIRROR 3 /

/

\
\

/

f

\

V

i

/

\

/

/
f

\

\

/

\
\

^

,/

1

\

S

1

V
\

-16

-8-4 0 4

FILM DISPLACEMENT IN DEGREES

12

16

Fig. 13. Variation of light efficiency through mirror compensating cycle.

is a decrease in overall illumination of
about 10%, lasting for a small fraction
of the active cycle.

The result of this light variation is a
certain amount of flicker, scarcely
noticeable in the projected image in
case of optical projection, but more
objectionable in the television image in
case of film scanning. In the latter
case, low-frequency beats are formed
between the 24-cycle film frequency and
the 30-cycle television frequency. These
low-frequency variations are more dis-
turbing than the small amount of 24-
cycle variation present in optical pro-
jection. The variation is clearly notice-
able when the machine is turned slowly
by hand. It manifests itself as a slightly
darker horizontal band traveling down
the picture as one picture frame fades
out and the next fades in.

The analytical study mentioned above
indicated that the light variation might
be decreased in either of two ways.
The edges of the projection lens might
be masked off, resulting in lower overall
light efficiency, or the number of mirrors
on the drum might be increased, result-
ing in a larger mechanical structure.

Since neither method would entirely
eliminate the light variation it was
decided instead to incorporate in the
machine a light-controlling servo system
which would compensate for all such
cyclic variations in light efficiency.

The principle of this light-servo system
is shown in the diagram of Fig. 14,
which again shows the principal parts
of the machine used as a film scanner.
The main light path is from the cathode-
ray tube raster through the optical
system, through the film in the gate
and into the signal photomultiplier.
By means of a plane mirror mounted
next to the raster an auxiliary light path
is provided which also sends light
through the optical system, but this
light passes through a clear gate at the
side of the film gate and from there to
an auxiliary photomultiplier. The out-
put from this phototube is then properly
biased and impressed on the intensity-
control grid of the scanning tube. As
long as the light efficiency of the optical
system is constant the auxiliary photo-
tube output is constant and is so biased
that no control voltage is impressed on
the intensity-control grid of the cathode-

Jensen, Graham and Mattke: Television Film Projector

15

FLYING SPOT

(CONTROLLABLE -

BRIGHTNESS)

SCANNING PATH

CORRECTING
MIRROR

AUXILIARY GATE

AUXILIARY
PHOTOTUBE

MIRRORS ON DRUM

•-PHOTO MULTIPLIER
SIGNAL PICKUP

Fig. 14. Simplified schematic diagram of flicker-correcting
servo system using separate phototube.

ray tube scanner. If the light efficiency
changes, then the corresponding change
in phototube output voltage is impressed
on the scanning tube grid in such a
manner as to restore the illumination of
the gate to its original value. Like the
position-control system discussed earlier,
this light-control system is also a servo
or feedback system, which automatically
will tend to keep the illumination at the
gate constant. The gain and the band-
width of the electrical part of the system
are such that residual light variations
are kept to a negligible minimum and
that the reaction time of the servo
system is fast compared to the periodicity
of the light fluctuations in the optical
system.

As shown in Fig. 14 the system has the
disadvantage of using two photomulti-
pliers, one for the television signal and

one for the light-control signal. This
requires that any nonuniformities in the
photosensitivities over the cathode areas
used must be absolutely identical in the
two tubes since the illuminated areas on
the photocathodes are not the same
throughout the cycle. If the variations
in cathode sensitivity are different, this
will result in a false indication of light
efficiency and will actually cause flicker.
To avoid this difficulty a modified system
was adopted as shown by the diagram
in Fig. 15.

In this diagram the drum mirrors and
the position-correcting mirror have
been left out for the sake of simplicity.
Figure 1 5a shows the spot scanning tube,
the projection lens, the film gate and
adjacent clear gate or monitoring slit,
and finally the common condenser
lens and photomultiplier. Figure 15b

January 1952 Journal of the SMPTE Vol. 58

\ ^-EDGE OF RASTER

\ x/x FOR PICTURE PURPOSES

FLYING' SPOT \ /

CATHODE RAY TUBE V ^~- EDGE OF OVER -SCAN REGION

A /

MONITORING
ei IT

MIRROR IMAGE'
OF SPOT AT "A"

(b)

Fig. 15. Schematic diagram of pulse-operated flicker-correcting
servo system using signal phototube.

shows a view of the film gate and
monitoring slit. As the scanning spot
travels horizontally across the tube face
from O to A, the image of the spot
travels across the film from O to A, the
light through the film thus producing
the usual television line signal at the
output of the phototube. At A the
light is cut off by the edge of the film
gate. The spot on the tube, however,

is allowed to travel a little further until
it is blanked off electrically at B. This
part of the travel, from A to B, is re-
flected through the system by a plane
mirror in such a manner that the
corresponding image travels across the
monitoring slit from A' to B'. The
light through the slit passes to the photo-
tube and produces a short pulse imme-
diately following the line signal, the

Jensen, Graham and Mattke: Television Film Projector

17

(*\ NORMAL

BLANKED

LIGHT MONITORING TIME

PULSES

WHITE

STANDARD COMPOSITE
VIDEO SIGNAL

SYNC
,' PULSES

SAMPLES OF VIDEO SIGNAL
AT END OF EACH LINE
x (DURING MONITORING PULSE
/ TIME) EXCEPT DURING
' VERTICAL BLANKING

STRETCHED, AMPLIFIED

FLYING SPOT

CATHODE RAY

TUBE

BLANKING PULSE

INHIBITORY PULSE

HORIZONTAL

DRIVE PULSES

FROM SYNC

GENERATOR

Fig. 16. Block diagram of electrical circuits in flicker servo
system.

amplitude of this pulse being a measure
of the light intensity in the gate.

The further processing of the photo-
multiplier signal is shown by the block
diagram in Fig. 16. After preamplifica-
tion the signal appears as shown at a,
consisting of a normal line video signal,
followed by a short pulse of amplitude
proportional to light intensity in the
gate. The preamplifier is followed by
an equalizing amplifier and a line
amplifier, where the light monitoring
pulse is blanked out and replaced by
standard synchronizing signals supplied
by the studio sync generator. The
output of the line amplifier is shown at
b and consists of a standard composite
video signal, ready for transmission.

From the preamplifier the signal also
passes to a box marked "sample and
hold." In this box the amplitude of
the light monitoring pulse is sampled

by means of a sample gate circuit and
then the sample is "stretched" in time
by a holding circuit until it occupies
almost . the entire time interval until
the arrival of the next sample one line
later. After some filtering the output
of the "hold" box therefore consists of
a quasi d-c voltage which is constant
in amplitude as long as the gate illumi-
nation is unchanged. This d-c voltage
is then fed to the grid-control amplifier,
which in turn controls the light intensity
of the scanning spot. If the gate illumi-
nation changes, the monitoring pulse
amplitude changes accordingly, thereby
changing the d-c holding voltage and the
spot intensity in such a manner as to
bring the gate illumination back to its
original value. Before the control volt-
age is applied to the cathode-ray tube
grid, blanking pulses are inserted to
blank off the beam at B in Fig. 15.

18

January 1952 Journal of the SMPTE Vol.58

Optical Components

The machine is presently designed for
scanning black-and-white film. It is
possible, therefore, to use on the spot
scanning tube a phosphor with tjie
shortest possible decay time, namely,
the PI 6 phosphor, which gives peak
light response in the near ultraviolet
region. The spot scanner is of con-
ventional design and uses an RCA
5ZP16 tube with an anode voltage of
30 kv. The signal photomultiplier is
an RCA 5819 tube with an S4 photo
surface, which is sensitive in the ultra-
violet region. The overall spectral re-
sponse of spot scanner and photo-
multiplier stretches from about 3500 to
4000 A, with peak response at about
3750 A.

At these short wavelengths it is
necessary to pay attention to the trans-
mission losses in the image-forming
components, i.e., the condenser lens
and the projection lens. The con-
denser lens is made of quartz and may
be assumed, therefore, to have very small
transmission loss in the wavelength
region used. The projection lens is a
modified Kodak Ektar projection lens,
100mm focal length, //3.5. It has
been redesigned for the present purpose
to work at a magnification of 4:1 and
to have best chromatic performance in
the region around 3750 A. The glass
of the lens measure's about 75% trans-
mission at this wavelength. Assuming
10% reflection loss at each mirror
surface, we thus have an overall trans-
mission efficiency of 0.9 X 0.9 X 0.75 =
0.60. The lens is stopped down to
about //4 and the overall effective speed
of the system is thus about //5.

The lamp used for the position control
is a 100-w, 110-v tungsten lamp operated
at about 60 v and the photomultipliers
used for this control are RCA 931 A
tubes.

Overall Performance

The geometrical resolution obtainable
with the machine at present is such as to

resolve clearly the bottom of the vertical
wedge on the standard RTMA test
chart. On pictures of the same RTMA
chart all ten strips in the gradation
wedges can be clearly distinguished on
the face of the monitoring tube. On a
10-in. kinescope contrast ranges over
200 to 1 have been measured from
pictorial scenes of the standard SMPTE
test film, with adequate gradation in the
halftones.

The residual vertical jitter of the pic-
ture has an rms value of about 1/2000 of
the picture height, or about 1/4 of a
scanning-line pitch. The sideways
weave of the picture is larger than that,
due to the fact that the film is not
guided as well as might be desired. It
is felt that this weave can be reduced
by proper mechanical guiding, but if
still better performance is desired, it
should be comparatively easy to monitor
the edge of the film with a photocell
arrangement, and to impart this monitor-
ing signal to a second pair of electro-
magnets on the correcting mirror.

The signal-to-noise ratio of the video
signal from the machine is about 35-40
db, peak signal to rms noise. Un-
doubtedly this figure can be improved
by using a faster lens, maybe an //3
or even an f/2 lens. It should be
mentioned, however, that a faster lens
will have less depth of focus and will
be unable to focus the flat tube raster
sharply over the curved film frame.
With the present lens there is no serious
lack of sharpness at the upper and lower
edge of the picture. With a much faster
lens it would probably be necessary to
provide a field-flattening lens to com-
pensate for the film curvature.

Bibliography
1920

C. F. Jenkins, "Continuous motion picture
machines," Trans. SMPE, no. 10: 97-102,
May 1920.

1921

C. F. Jenkins, "Continuous motion pro-

Jensen, Graham and Mattke: Television Film Projector

19

jector for the taking of pictures at high
speed," Trans. SMPE, no. 12: 126-127,
May 1921.

1922

C. F. Jenkins, "Prismatic rings," Trans.

SMPE, no. 14: 65-71, May 1922;

discussion, ibid., 72-73.
F. N. Stewart, "Note on new continuous

projector," Trans. SMPE, no. 14: 162,

May 1922.

1923

H. D. Taylor, "The feasibility of cinema
projection from a continuously moving
film," Trans. Opt. Soc., no. 25: 149-176,
May 1923-1924.

1924

L. Bowen and H. Griffin, "Is the continuous
projector commercially practical?,"
Trans. SMPE, no. 18: 147-151, May
1924.

1928

A. J. Holman, "A non-intermittent optical
projector," Trans. SMPE, vol. 12, no. 36:
1184-1188, Sept. 1928; discussion, ibid.,
1188-1190.

J. F. Leventhal, "A new optical com-
pensator," Trans. SMPE, vol. 12, no. 36:
1068-1072, Sept. 1928; discussion, ibid.,
pp. 1072-1075.

J. F. Leventhal, "Projectors with optical
intermittents," Trans. SMPE, vol. 12,
no. 34: 406-409, Apr. 1928.

"The Mechau projector," Trans. SMPE,
vol. 12, no. 36: 1193-1195, Sept. 1928.

1930

A. J. Holman, "Apparatus developed to
simplify manufacture of lens wheels for
continuous projectors," Jour. SMPE,
14: 623-635, June 1930.

A. J. Holman, "Prospects for non-inter-
mittent projection," Mot. Pict. Pro-
jectionist, 3: 9-10, Oct. 1930.

A. J. Holman, "The revolving lens wheel
projector," Jour. SMPE, 15: 20-38,
July 1930.

A. J. Holman, "Why and how of con-
tinuous projection," Mot. Pict. Pro-
jectionist, 4: 25-26, 31, 33-37, Nov. 1930.

F. H. Richardson, "Continuous projection:
an inquiry," Exhibitors Herald- World, 100:
49-50, July 5, 1930.

1931

A. J. Holman, "Continuous non-intermit-
tent projectors," Jour. SMPE, 16: 612-
618, May 1931.

M. Hue, "Non-intermittent motion pic-
ture projector," Bull. Soc. Franc. Phot.,
18: 128-136, June 1931. Abstract in
Eastman Monthly Bull., 18: 17, Jan. 1932.

Carlos Mendizabel Brunet, "New continu-
ous projector shown by inventor," Mot.
Pict. Herald, 103: sec. 1, 19, May 16,
1931.

"New non-intermittent projector," Kine-
mat. Weekly, 174: 55, Aug. 20, 1931.
Abstract in Eastman Monthly Bull., 18:
134, Mar. 1932.

W. C. Plank, "Art of continuous cinematog-
raphy," Internal. Proj., 1: 22-23, Nov.
1931; ibid., pp. 9-11, Dec. 1931.

W. C. Plank, "Some interesting properties
of continuous projectors," Jour. SMPE,
16: 709-718, June 1931.

1932

A. Bourgain, "Non-intermittent projec-
tors," Technique Cinemat., 13: 69-76,
Feb. 1932. Abstract in Eastman Monthly
Bull., 18: 337, July 1932.

Plank, W. C., "Inertia in the service of
cinematography," Jour. SMPE, 19:
565-578, Dec. 1932.

"A portable non-intermittent cine pro-
jector," Jour. SMPE, 18: 101-106, Jan.
1932.

H. A. Robizek, "Continuous projection
by optical compensation," Internal :. Proj.,
2: 14-16, Mar. 1932.

J. L. Spence, "Mechanical advantages of
the optical intermittent projector," Jour.
SMPE, 18: 593-599, May 1932.

F. Tuttle and C. D. Reid, "The problem
of motion picture projection from con-
tinuously moving film," /. Opt. Soc.
Am., 22: 39-64, Feb. 1932; Bibliog-
raphy, Jour. SMPE, 20: 3-30, Jan. 1933.

1933

"A new continuous projection attachment
just announced by Victor," Am. Cine-
matographer, 13: 20, Mar. 1933.

M. Pirani and R. Rompe, "Uber Kino-
projektion mit kontinuierlick abrollen-
dem Film," (on the projection of con-
tinuous film), Kinotechnik, 15: 131-132,
Apr. 20, 1933.

F. Tuttle, "A non-intermittent high-speed

20

January 1952 Journal of the SMPTE Vol. 58

16-mm camera," Jour. SMPE, 21: 474-
477, Dec. 1933.

1934

R. H. Cricks, "The non-intermittent : some
attempts at continuous projection/'
Ideal Kinema, 2: 40-43, Sept. 13, 1934.
Abstract in Eastman Monthly Bull., 21:
161, May 1935.

1937

Fernseh-Filmabtastgerat, "Television film
scanning device," Mechau Filmtechnik,
13: 163-164, Oct. 1937.

1938

H. S. Bamford, "Non -intermittent pro-
jector for television film transmission,"
Jour. SMPE, 31: 453-461, Nov. 1938.

1940

F. Ehrenhaft and F. G. Back, "Non-
intermittent motion picture projector,"
Jour. SMPE, 34: 223-230, Feb. 1940.

1944

R. H. Cricks, "The requirements of modern
projector design," Bibliography, Jour.
SMPE, 43: 129-147, Aug. 1944; discus-
sion, ibid., 147-148.

Discussion

Anon: I am sure we all agree that you
have done a wonderful job .... I'm
wondering if you feel that you have sold
yourself down the river at all in utilizing
the sprocket hole instead of trying to use
the frameline on the film in some way?

A. G. Jensen: We felt that the sprocket
hole, as far as we understand it, is the
primary standard that you have in the
motion picture. The whole registration
all through the processing is done by means
of these sprocket holes, so we felt that
they were the primary standard and we
were afraid we couldn't hope to do much
better than that. If you go by the frame
then you are in trouble, because depending
upon the film material you don't always
have a good reference. You might have
to make an artificial reference in order to

make sure that you always have a good
reference edge — a nice black-to-white
edge which will not vary with the content
of the picture. When you go by the
sprocket hole you don't have that at all.
Another thing — it is very difficult, I
think, to do a good job of monitoring if
you use transmitted light. If you were
to use a film frame you would almost be
forced to use transmitted light and depend
upon black vs. clear film, as I see it. By
using the sprocket hole we use reflected
light and that seems to be practically
independent of what has happened to the
film as far as exposure is concerned.
Whether the area around the sprocket
hole is clear film or completely black
exposed film doesn't matter at all. The
reflected light is almost independent of
that, which is a great advantage. If you
were using transmitted light the gain of
your servo, the dependability of your servo
would depend on the density of the film
at the point where you measure. We
have avoided that by using reflected light.

Anon: Are you of the opinion that you
could use old film and new film and the
results would be about the same? The
trouble with a lot of television operations
is that the prints aren't always good.

Mr. Jensen: Of course I might say that
the edge of the sprocket hole that we
monitor on is a trailing edge, the one that
is least chewed up by sprockets.

John Kudar: Mr. Jensen, a few months
ago, I think in Electronics, there was an
article about this projector and there was
a remark that the jitter control would be
able to compensate for shrinkage.

Mr. Jensen: Well, it is true that if the
shrinkage is not too bad it does do a fairly
good job of compensating. If the shrink-
age is severe then you do have to adjust
the gate, but it isn't too complicated to
do. The shrinkage would generally be
uniform enough through an entire piece
of film so that you don't have to adjust
while you're running, but if the film is badly
shrunk you may have to adjust initially
before you start that piece of film by
moving the gate a little bit closer or a
little bit further away from the mirrors.

Jensen, Graham and Mattke: Television Film Projector

21

Color Television Reproducers

By HARRY R. LUBCKE

Altering the velocity of traverse of the electron stream in combination with
a suitable heterogeneous reproducing screen is the basis of a device described.
It differs from the CBS mechanical, the RCA tricolor tube and the Geer screen
systems of color television reproducers.

GRIEVING the right combination of
elements for the reproduction of color
television is not easy. By counting the
issued patents concerned with this
problem and observing the accelerated
rate at which these are issuing we must
conclude that many are now engaged
in such research — from the individual
inventor to the lush corporate laboratory
working around the clock.

One of Zworykin's original patents1
on the iconoscope, filed in 1925 and now
expired, disclosed the elemental type of
color screen that is being reinvented to
the present day. Another example is
due to Bronwell2 of Chicago. Three
grids are provided for a three-color
system, arranged in lieu of the fluorescent
screen. The wires of one screen are
staggered with respect to the wires of
the others as regards electron flow. By
the expedient of raising the voltage on
the grid that carries the color phosphor
to be energized at a given instant, the
electrons are attracted to it in preference
to the other grids and impinge upon it

Presented on October 16, 1951, at the
Society's Convention at Hollywood Calif.,
by Harry R. Lubcke, Consulting Engineer,
2443 Creston Way, Hollywood 28, Calif.

with sufficient velocity to fluoresce the
phosphor on that grid alone.

The first all-electronic color television
reproducer to be produced in any
quantity is the well-known tricolor tube
of the Radio Corporation of America.
This tube is an improvement of the
original obturating principle of the
German corporation "Fernseh Aktien-
gesellschaft" (translated: "Television Cor-
poration") proposed in their French
Patent No. 866,065, filed in July 1939.

Fernseh showed rods disposed in
only one direction, like pickets on a
fence. These were in front of a lined
triphosphor screen. RCA did one better
than this, by stamping a plate full of
holes and arranging three dots of different
phosphors behind each hole on the side
near the viewer.

Still another scheme is the Geer
screen.3 This was a fundamental in-
vention. It reproduces a color image
upon a substantially single plane while
retaining necessary uniqueness of rendi-
tion of the color components that form
the image.

The practical importance of the single-
color image has become apparent as the
art has progressed. The science of

22

January 1952 Journal of the SMPTE Vol. 58

Fig. 1. A three-gun faceted-screen color television reproducing device.

The divergently formed electron streams are converged at the screen

by the coil assembly.

optics has taught long enough how to
combine three-color component images
into the full-color composite, but it is
inadvisable to leave the matter of regis-
tration to the cabinetmaker and the
serviceman if there is a better way.
The Geer screen allows combination of
the colors in proper registration without
complicating the electron optics of the
device at the screen proper.

The tube of Geer and of his con-
temporaries, A. N. Goldsmith4 and the
late John Logic Baird6 of England,
suffers from the need for (usually)
double keystone correction. This is
occasioned because of the considerable
physical separation of the three electron
guns of the devices. If there was a way
in which this separation could be
eliminated a great improvement would
result.

The author evolved an answer to this6
in the form of a cluster of three guns
that originate mutually divergent elec-
tron streams at essentially one point.
This arrangement is shown in Fig. 1 as
guns 2, 3 and 4. The electron streams
from these are deflected as a whole by a
single set of deflection coils — 11, 12,
13 and 14. This arrangement has the
advantage of simplicity and accuracy,
making it unnecessary to match three

pairs of deflection coils for congruence
of scanning deflection as is required in
the devices previously mentioned.

The new originating arrangement
operates with the tridirectional faceted
(Geer) screen because of coil 15.
Located near the screen and carrying
a direct current, it acts to converge the
three divergent electron streams 16,
1 7 and 1 8 because of radial components
of the magnetic field created. By ad-
justing the current in the coil the three
streams are made to converge in the
plane of screen 10. In this way, three
separate electron streams, independently
controllable as to intensity, can be con-
verged at the screen as though each had
come from a gun widely separated from
the others and each phosphor upon the
screen can be individually excited to
any degree.

The author has found that this ar-
rangement operates over a considerable
area of the fluorescent screen, but not
at the extremes. This can be taken
care of in two ways.

Firstly, the size of the coil 15 can be
increased. One of the advantages of
this system is that the deflecting and
converging instrumentalities are exterior
to the vacuum structure and so can be
altered, adjusted or replaced without

H. R. Lubcke: Color Television Reproducers

23

Fig. 2. The rise and decay of light
output of three phosphors as a function
of time; the basis of operation of cathode-
ray-tube color reproducer devoid of
geometrical screen structure.

affecting the most costly component, the
cathode-ray tube itself.

Secondly, the magnetic axis of the
coil can be inclined in synchronism with
the scanning of the combined electron
streams. In the figure, the two lower
streams are at 120° from each other
and from the upper stream. The com-
mon axis is thus inclined upward at the
instant in scanning for which the figure
was drawn. Coils 20 and 21 are pro-
vided to accomplish the inclination.
These are wound in horizontal planes
and supplied with a fraction of the
vertical scanning energy. The field
from these coils alters the original field,
reinforcing it below, weakening it above,
so that the magnetic axis is above the
geometrical axis of coil 15.

With coil 22 and a companion 23
behind, the magnetic axis is also altered
from left to right upon the face of the
tube. Most fortunately, the converging
action is constant over quite an area
around the axis and precision in altering
the magnetic field is not required.

After considerable further study of
this subject the author evolved another
method.7 This method does away with
the faceted screen, the magnetic coils
and any other physical feature that

would identify it as a color cathode-ray
tube. The chromatic action takes place
in the screen itself.

Man now knows of some two hundred
chemical compounds, which, when com-
bined with minute amounts of suitable
impurities, give off cold light in return
for the energy of impact of electrons.
These substances are known as phos-
phors. By energizing suitable phosphors
differently with respect to time, selective
excitation of different color components
of a heterogeneously constituted screen
can be accomplished.

In Fig. 2, for example, curve 12
represents the excitation and decay time
of zinc sulfide with a small amount of
silver as the impurity-activator, the
phosphor being hexagonally crystal-
lized. It will be noted that the response
and decay are rapid. This phosphor
fluoresces blue.

Curve 13 is for zinc silicate, with
manganous oxide as activator, rhombo-
hedrally crystallized. The response and
decay of this phosphor are average and
it fluoresces green.

Curve 14 is for zinc sulfate, with man-
ganese sulfate activator, orthorhombo-
hedrally crystallized. The response and
decay of this phosphor are slow. It
fluoresces red.

For any small interval of time, such
as A t shown, the variation of response
of the three phosphors is very different,
even though each phosphor be impacted
with the same number of electrons
accelerated through the same potential
in the electron gun or guns.

The response of light, L, is near
maximum for the rapid phosphor 12.
It is much less for the medium phosphor
13, with only the small area under the
resulting (dashed) decay curve being
effective in light output. The response
is even less for the slow phosphor 14,
the amplitude rising only to point 16.

Thus, if we traverse our heterogeneous
phosphor screen rapidly we secure a
nearly pure blue response. But what
happens if we traverse such a screen

24

January 1952 Journal of the SMPTE Vol. 58

slowly; do all the phosphors light up
and produce white light?

This would be true if the only property
utilized was speed of phosphor response.
Actually, by combining a number of
processes in an additive manner it is
possible to shut off the rapid phosphors.

Phosphor materials behave differently
under different temperature conditions.
It is possible to select particular phos-
phors and to give attention during the
preparation of them so that, for instance,
the temperature characteristic of the
above-mentioned fast phosphor is such
that the light output for a given excita-
tion reduces rapidly with increase in
temperature. At the temperature of
boiling water the light emitted can be
made only one-fifth that at room
temperature.

In the present device this phosphor is,
furthermore, formed in small particles —
less than 10~3 millimeter. Small par-
ticles heat much more rapidly and
attain a higher temperature than large
ones. Consequently, the temperature
effect is accentuated and under a slowly
moving or stationary electron stream,
such as is required to activate the slow
phosphor; the light from the rapid
phosphor has reduced to a small value.

The coefficient of secondary emission
of the phosphor is similarly utilized.
The coefficient of the above-selected
fast phosphor decreases with increase in
temperature. By selecting a proper
operating potential for the gun of the
cathode-ray tube in relation to the
secondary emission characteristic of
the phosphor the ratio of secondary
emission can be made less than one.
This means that the phosphor particle
accumulates a negative charge under
the influence of the slowly moving or
stationary electron stream and ceases
to glow because of the resulting lower
effective velocity of impact.

Not only can these factors, inherent
in the rapid phosphor and its prepara-
tion, be utilized to cause the blue light
to cease shortly after time A t when the

rate of traverse of the electron stream
is slower, but the rapid phosphor can
be covered with a thin layer of silica,
chemically deposited on the particles
before the screen is fabricated. Silica
has a secondary emission ratio less than
one at desirable cathode-ray tube operat-
ing voltages. Metals, such as thin
films of tungsten, have similar effects.
These substances do not appreciably
alter the effectiveness of the primary
electrons of the electron stream when
this impacts the phosphor during the
proper brief interval of excitation.

Without going into detail, the phos-
phor of medium time of response is also
formed in small crystals and is chosen
and/or treated to cease functioning
during slow traverses. It does not, of
course, become appreciably excited dur-
ing the rapid traverses. Should the
green signal be black at any particular
instant the grid of the electron gun
cuts off the stream during the moderate
speed of traverse.

The slow phosphor, in addition to its
slow response, was selected to perform
under the conditions that shut off the
more rapid phosphors. It is formed of
relatively large crystals that take longer
to heat and of a phosphor composition
that has a temperature characteristic
giving visual response at high tempera-
ture and being capable of operating
under slow or stationary electron streams
for the brief instants utilized in the
device.

These characteristics are accentuated
in an alternate screen construction that
includes a largely transparent, yet
"black-screen" metal deposit on the
inside of the glass face. This is con-
nected to the second anode. The slow
phosphor is laid down in contact with
the metal. This removes any possibility
of accumulating a negative charge
and also provides some measure of ther-
mal sink, preventing heating of that
phosphor. The two other phosphors
are deposited on top of the slow one and
out of contact with the metal coating.

H. R. Lubcke: Color Television Reproducers

25

Fig. 3. Incremental waveform u is
combined with horizontal deflection
waveform W to secure variation of the
velocity of traverse of the electron
stream over the phosphors.

Experimentally, the relative amounts
of each phosphor may be changed to
effect chromatic adjustment. The inte-
grated intensity of the corresponding
color is thus varied. The spectral
response is empirically determined, being
composed of the inherent response of
the phosphor and a small and fixed
response of another. Not every batch
of available phosphors can be used with
certainty. Because the impurities of
subspectroscopic amounts and the actual
lattices formed vary with ingredients
and the precise routine of preparing the
phosphor^ variation of performance will
be experienced unless the phosphor
used is selected upon the basis of test
under operating conditions.

Fast, medium and slow traverses of
the electron stream over the phosphors
have been mentioned. In monochrome
television the scanning speed is constant
over all of the visible portion of the
reproduction. In the present device
a high-frequency deflecting waveshape
is combined with the usual horizontal

deflecting waveshape to accomplish
speed variation. For a three-color sys-
tem, a truncated triangular wave is one
shape that is used; that is, a triangular
wave-shape with the tops of the tri-
angles cut off.

This is shown in Fig. 3. Waveshape
W represents a small portion of one
horizontal scan. Displacement D across
the field of view is represented vertically
and time horizontally as the abscissa.
Waveform u is the truncated triangular
one. When these two are combined the
third waveform results. Where the
slopes of waveforms u and W are the
same, the resulting velocity is greatest,
as at /. When waveform u effects no
displacement with time, the top trun-
cated portion, the resulting velocity is
medium, as at m. Where the slopes of
waveforms u and W are equal but
opposite, the resulting velocity is zero,
as at s. Thus, the scanning spot suc-
cessively travels rapidly, at medium
speed and stops, all at substantially dot
repetition rate.

Refinements are possible; an asymmet-
ric truncated waveshape can be produced
by attenuating the low-frequency re-
sponse of the truncated wave device.
The slanting truncated top then produces
the stationary spot, the rapid traverse
is then more rapid and the normal or
medium traverse is actually executed
in reverse. Also, the stiffness of the
electron stream can be altered in syn-
chronism with these incremental de-
flections and the ratio of velocities
further increased.

Two types of devices to provide the
truncated waveshape have been de-
veloped and tested. One is a resonant
oscillatory circuit employing a single
small triode. Two of the coils of that
circuit are placed astride the neck of the
cathode-ray tube and directly deflect
the electron stream in the desired wave-
shape. The other type of device is a
relaxation oscillator that gives a tri-
angular waveshape directly. This is
truncated with a diode. The resulting

26

January 1952 Journal of the SMPTE Vol.58

wave is fed into the usual horizontal
deflecting coils, into similar coils of a
few turns or impressed upon deflection
plates within the cathode-ray tube.

The period of each of the three portions
of the truncated wave is made equal to
the period of exhibition of one of the
primary colors of the color system. The
above-described oscillator is kept in
synchronism by feeding a small amount
of color change information to it.

The color image thus formed is com-
posed of a short blue dash, a shorter
green dash and a red dot successively
repeated along each line of scanning in
approximately the same manner as the
individual primaries are reproduced side
by side in the shadow mask tricolor
tube.

Other phosphor combinations have
been worked out so that the dot is of
green rather than red hue to favor

rendition of detail. Several relations
between detail and color standard are
possible.

Acknowledgment

The encouragement given to this
work by Willet H. Brown, President of
the Don Lee Broadcasting System,
Hollywood, is gratefully acknowledged.

References

1. V. K. Zworykin, U.S. Pat. 1,691,324,
Nov. 13, 1928.

2. Bronwell, U.S. Pat. 2,461,515, Feb. 15,
1949.

3. G. W. Geer, U.S. Pat. 2,480,848, Sept.
6, 1949.

4. A. N. Goldsmith, U.S. Pat. 2,481,839.
Sept. 13, 1949.

5. John Logic Baird, British Pat. 562,168,
application dated July 25, 1942.

6. U.S. Pat. application, GGRT I.

7. U.S. Pat. application, GGRT III.

H. R. Lubcke: Color Television Reproducers

27

Film-Spool Drive
With Torque Motors

By A. L. HOLCOMB

The characteristics of torque-motor drives are described in connection with
their use for take-up and feed spools in film-pulling mechanisms. A useful
but limited field of application for this type of drive appears to be indicated.

I

,N RECENT YEARS torque motors have
been successfully used in some applica-
tions as a drive for take-up spools
replacing the older friction-drive devices,
and it is the purpose of this paper to
consider the relative merits of both
methods with respect to performance
and convenience. The term "film
spool" is here used to cover all of the
various types of reels and similar devices
on which film or tape is wound or from
which it is unwound in the process of
recording or reproducing sound.

A "torque" motor is any motor which
produces maximum torque at standstill
and which provides a sufficiently high
input impedance to allow it to be stalled
without excessive current demand. Such
motors may be either a-c or d-c and are
usually rated on the basis of stalled torque
and the percentage of operating time
they may be stalled without exceeding
some acceptable temperature rise. The
type most used is an a-c induction motor
with either three-phase or single-phase

Presented on October 17, 1951, at the
Society's Convention at Hollywood, Cali-
fornia, by A. L. Holcomb, Westrex Corp.,
6601 Romaine St., Hollywood 38, Calif.

stator and equipped with a high-
resistance rotor. The development of
compact a-c capacitors has permitted
the use of two-phase stator windings,
one of which can be effectively resonated
with a series capacitor, thus providing
the necessary phase shift for operation
from a single-phase source. Such
"capacitor run" motors are more easily
switched and controlled than three-
phase motors and provide essentially the
same operating features.

The use of torque motors as a take-up
drive for film-recording equipment was
experimentally considered as long ago
as 1935. The motors then available
for such duty were three-phase, wound
rotor units and series d-c or universal-
type motors. While it was found
possible to operate such motors so as
to provide an approach to constant film
tension, it was decided at that time
that the advantage realized did not
justify the cost and circuit complications
which were found necessary. Torque
motors have been used with excellent
results in many of the magnetic-tape
recorders developed in recent years,
and it has logically been suggested that

28

January 1952 Journal of the SMPTE Vol. 58

the modern version of these motors
would be of value in sprocket-type re-
corders and reproducers using 35mm,
or 16mm films.

Film and Tape versus Flutter

Standard -J-in. tape as now used
employs a thin flexible base which
conforms readily to small-diameter
drums or capstans, permitting such
units to operate at relatively high speeds.
The base is too fragile and flexible to
utilize sprockets and sprocket holes as
a synchronizing means and, conse-
quently, a friction drive to the capstan
is satisfactory. Thus, without gears or
sprockets, good motion can be obtained
with a relatively simple mechanical
filter provided the drag and take-up
are smooth. Torque motors provide
not only a smooth drag and take-up,
but also a convenient high-speed drive
in either direction which is an essential
facility in most tape machines.

Motion picture film base is relatively
thick and provides sufficient longitudinal
rigidity and durability to withstand not
only the constant small synchronizing
impulses imparted by normal sprocket
action, but also the high acceleration
of intermittent picture motion. Syn-
chronism between films is obtained by
effectively gearing the film to the
sprocket, the sprocket to the drive
motor, and electrically gearing the motor
to similar motors or to a common supply
line. All of these gear trains present in
some degree the characteristics of inertia,
resilience or backlash, and the resultant
unfiltered film motion is far from de-
sirable by sound-recording standards.
Thus, the mechanical filter require-
ments are obviously more exacting for
sprocket-type film machines than for
tape machines. Uneven or erratic take-
up or drag can add to the total "flutter"
or "wow" which must be corrected and
it was one of the objects of this investiga-
tion to determine whether the substitu-
tion of torque motors as a drive for the
feed and take-up spools could provide

measurably better film motion than the
friction-type drive when operated in
conjunction with a good mechanical
filter.

The flutter which may be contributed
by feed or take-up spools may take any
or all of four forms:

1 . Low-frequency or erratic variations
due to uneven friction in clutch or belt
drive.

2. Sprocket-hole flutter (96 cycle/sec)
due to high film tension at beginning or
end of a reel.

3. Erratic shifting of the film with
respect to the sprockets at "crossover"
where the net tension on the film re-
verses.

4. Gear-train chatter due to unload-
ing the sprocket gears at crossover.

It will later be shown that the friction
drive may contribute 1, 2 and 3, but not
4, and while the torque-motor drive is
not likely to contribute 1, it may con-
tribute 2, 3 and 4.

In order to visualize the conditions
existing during the transfer of a standard
1000-ft reel of film from feed to take-up
spools, it may be worth while to consider
some of the factors common to a constatnt-
torque drive.

Film Tension — Constant Torque

The minimum safe take-up tension is
determined by loop formation at start
and is about 300 g, although somewhat
less than this value may be operable.
The maximum tension is determined by
film breakage or sprocket-hole mutilation
and is dependent to a large extent on the
film path, matching of sprocket teeth
and sprocket holes, the size of drive
sprockets and the acceleration charac-
teristics of the driving system. It can
be stated, however, that in general the
maximum tension should never exceed
2500 g. Drag tension need only be
enough to prevent a full reel from coast-
ing and 1 50 g minimum is satisfactory.

A standard 1000-ft reel presents
approximately a 2-in. diameter spool
when empty and 9J-in. diameter when

A. L. Holcomb: Film-Spool Drive

29

40C

200

100 200 300 400 500 600 700 600 900 1000

FEET OF FILM ON TAKEUP.
Fig. 1. Take-up and torque-motor characteristics; tension vs. feet of film on take-up.

full. Since the film speed is fixed at
close to 90 ft/min, the reel speed must
vary inversely with the effective diameter
of the spool on which it is wound or
from which it is unwound; The speed
of a 1000-ft take-up reel at start is thus
about 172 rpm and at the end of a
1000-ft roll is roughly 35 rpm, the feed-
reel speed varying in the same manner,
but inversely.

A good friction clutch, as used for
this duty, delivers essentially the same
torque to the driven element regardless
of the differential or slip speed between
driver and driven members. If the
diameter of the film spool were constant,
this constant torque would produce a
constant pull, or tension, on the film,
but the spool diameter varies from
2 in. to 9^ in., or a ratio of 4.9:1, and
as a result the tension varies by the same
ratio. If the torque is expressed in
gram-inches (grams pull on a 1-in.
radius), and this factor is constant, then
the film tension under any spool condi-

tion will equal the torque divided by the
spool radius.

The film-tension conditions for a
1000-ft take-up reel (curve A) and feed
reel (curve B) with 2-in. hubs are shown
in Fig. 1, plotted against the number
of feet of film accumulated on the take-up
reel. Since the minimum take-up ten-
sion of 300 g is desired, it is assumed that
the friction of the take-up clutch has
been adjusted so that this tension is
obtained with a full reel (radius 4.94
in.), and the torque which will remain
constant is then 1480 g-in. (300 X
4.94 = 1480). The minimum spool
radius of 1 in. increases the tension to
1480 g. It will be noted that reel
speed and film tension are directly
related and both are inverse functions
of spool radius. Thus, the two curves
showing film tension from the feed reel,
B, and from the take-up reel, A, are
similar, but reversed. The minimum
drag tension, as shown, is set at 150 g.
It will be apparent that most of the

30

January 1952 Journal of the SMPTE Vol. 58

soo

400

2

Q-'300
K

200

100

\

200

600

800 IOOO 1200 I40O

TORQUE GRAM- INCHES.

1600

2OOO"

Fig. 2. Take-up and torque-motor characteristics; speed (rpm) vs. torque.

tension change occurs when the hub
radius approaches minimum, and
demonstrates that the use of reels with
4-in. diameter or larger hubs would
materially reduce the tension ratio.
Curve C of Fig. 1 will be referred to
later in connection with Fig. 2.

It has generally been accepted that
2000-ft and larger reels present a more
difficult take-up problem than the
standard 1000-ft reel with a 2-in. hub.
This is true only with respect to inertia
since 5-in. hubs are generally used on
the larger reels and the ratio of minimum
and maximum diameters and film
tensions is more favorable for such reels
up to 6000 ft (ratio 4.8:1) than is the
case in 1000-ft reels. The inertia,
however, increases approximately as
the 4th power of the film-spool diameter
and, consequently, the torque required
to prevent loop formation at start
becomes excessive except as a smooth
and slow-starting drive system is em-
ployed.

"Crossover" occurs in the region
where the drag tension equals and then
exceeds the take-up tension and due
to the high-tension ratio (nearly 5:1)
of a 2-in. hub, this condition cannot

be avoided with a constant-torque drive.
With well-matched sprocket teeth
and sprocket holes and a film path
which provides a considerable belt
effect between the film and the body of
the sprocket, the film motion, with
respect to the sprocket, suffers very
little due to the change in direction of
the net tension; but a flutter condition
can exist due to gear chatter if the
crossover removes the load from the
sprocket-driving gears and drive motor.
This condition can occur with torque
motors, but when a friction-clutch take-
up is driven from the sprocket shaft it
is usual to provide an overdrive of about
20% above the maximum reel speed.
This presents a friction load on the gear
train and motor at all times, regardless
of tension between the take-up reel and
sprocket, which is in the same direction
as the drag from the feed reel. Thus,
a friction-clutch take-up will normally
present a film tension crossover, but if
properly designed will not unload or
reverse the torque on the take-up sprocket
and motor gears. To realize fully this
advantage in a dual sprocket drive, the
holdback sprocket should also be damped
or loaded.

A. L. Holcomb: Film-Spool Drive

31

Torque-Speed Characteristics

Figure 2 shows torque-motor charac-
teristics at A and B, and at C shows
the ideal torque-speed curve required
to provide constant film tension from a
1000-ft reel with 2-in. hub. Curve A
shows the relation between torque and
speed of a 12-pole torque motor fre-
quently used as a direct drive for take-up
reels. For take-up duty the speed of
the reel and of the torque motor is
determined by the film speed, and it will
be noted that the dotted lines carried
out from the minimum and maximum
reel speeds of 35 and 170 rpm both
intersect the motor characteristic at
about 2000 g-in. Thus, the torque of
this unit, as normally used, is constant
and behaves in the same manner as
does a friction clutch. Ideal torque-
motor characteristics shown by the
straight line B, drawn from stall torque
to free speed, would be some improve-
ment but would still fall far short of
matching the curve G.

The stall-torque or zero-speed point
of either A or B can be moved toward
zero torque by various means such as
series resistance, but the no-load speed
is chiefly a function of the number of
poles in the motor and, thus, it becomes
apparent that even an approximation
of curve C will require either a 36-pole
motor or approximately a 3:1 me-
chanical-speed reduction. The charac-
teristic D can be obtained in this manner
and while it does not provide constant
tension due to the curvature of C, it is
an approach thereto, as shown in C of
Fig. 1. The latter is replotted from
curve D in terms of film tension on a
take-up reel driven by a 12-pole torque
motor with a 3 : 1 mechanical reduction,
and represents the best relationship
that can be obtained between film
tension and number of feet of film on the
take-up with a torque-motor drive.
However, this destroys the ability for
fast runback except as a gear change is
employed, and spoils the mechanical

simplicity which is one of the most
attractive features of torque motors for
this duty.

Flutter Characteristics

In Fig. 3A the total flutter of a very
good tape machine is shown without
automatic speed correction and the
same record is shown in Fig. 3B with
speed correction. The average rms
value for either condition is about 0.06
or 0.065%. This relatively low average
of total flutter is a good example of
what can be done with a simple filter
and torque motors when gears and
sprockets are eliminated. An attempt
was made to produce approximately the
same type of drive and filtering action
as obtained in the tape machine, but
substituting a 16-tooth sprocket as a
drive member together with the neces-
sary gears. Torque motors were used
for drag and take-up duty and the drum
near which the recording head was
located carried a substantial flywheel.
The resultant flutter from a very good
record is shown in Fig. 3C. It will be
noted that the average flutter is some-
what greater than that of the tape
machine, and shows considerable low-
frequency variations and erratic charac-
teristics not present in the tape record.
These flutter charts are made on a time
axis in which one division equals one
minute and in which the vertical
ordinates represent 0.01% peak or
0.007% rms flutter for each small
division.

The performance shown at G probably
represents as good motion as could be
obtained from synchronously driven 35-
mm film with a mechanical filter such as
is used in the best magnetic-tape ma-
chines. The take-up and drag tensions
on the film machine were adjusted to
be the same as shown in Fig. 1. Al-
though 5-in. hubs were used, some
evidence of high take-up tension will
be noted at the beginning of the reel,
although the increase of drag tension
at the end of the reel does not reach a

32

January 1952 Journal of the SMPTE Vol. 58

Fig. 3. Flutter characteristics. A and B, tape machine;
C, 35mm sprocket machine.

Fig. 4. Flutter characteristics — 35mm sound recorder. A, friction take-up
and drag; B, torque-motor take-up and drag.

A. L. Holcomb: Film-Spool Drive

33

sufficiently high value to be significant.
An increase in the number and amplitude
of erratic excursions toward the end of
the record shows the effect of crossover.
Figure 4A is a chart made using the
same record as in Fig. 3C reproduced on
a standard Westrex RA-1467-A Re-
corder, described by Crane, Frayne and
Templin,1 with sprockets and mechanism
as described by Davis.2 This machine
was not given special treatment other
than a check to insure normal operation
of all elements. In Fig. 4B is shown a
similar chart made from the same record
on the same machine, but with torque-
motor take-up and drag. Within the
accuracy of measurement, which is
about 0.005%, the two charts are
essentially similar, neither one showing
any adverse effects from crossover or
other take-up disturbances. An analysis
of the flutter under the two conditions
is practically identical and is shown in
Table I, neglecting those frequency
bands where the measured values are
less than 0.01% rms.

Table I.

Cycles

Per cent rms

1-200 .

0.035

130-200 . . . .
80-130 . . . .
50-80 ....
34-50 ....
4f-7 . ... .
1-2* ....
0-1 ....

0.021
..... 0.014
0.010
:•..... 0.010
0.010
•: -. . -. . 0.014
. . 0.010

Using the torque-motor take-up, a
considerable number of tests were made
with various values of constant tension
and constant torque. The results in-
dicated that with this particular re-
corder the filter system was adequate to
eliminate undesirable effects of crossover
or variations in film tension up to about
2000 g. Thus, as far as this production
recorder is concerned, it would appear
that a good friction-clutch take-up
driven by a reasonably smooth belt is

capable of delivering essentially the same
flutter performance as when equipped
with torque motors.

Operational Features

Torque motors, however, do have a
number of advantages, for certain specific
duties, which the friction drive cannot
supply. They are readily operable in
both directions by simple switching, at
slow speeds for normal recording, or at
high speeds for fast rewind in either
direction. They are also capable of
being controlled to provide constant
tension if such control is deemed neces-
sary, and where 1000- and 2000-ft reels
are used interchangeably they can be
switched to provide the proper torque
for either condition. They also have
a further advantage, that if they are
conservatively engineered for the job
so that they do not overheat, there is
little, if any, maintenance required.
However, torque motors are inherently
slow in response, due to a poor torque-
inertia ratio, and they should be
equipped with electromechanical brakes
to prevent coasting and to provide some
rigidity of the reels for threading opera-
tions when the drive power is off,
because when film is allowed to develop
appreciable slack, torque motors are
likely to break film or tear sprocket
holes when the motors are excited and
this slack is taken up. Such motors
are, in effect, a separate motor system
which requires additional controls of
some complexity if the desirable features
are to be realized. Also, the weight of
a pair of these motors, together with the
necessary control equipment, will add
at least 25 Ib to any 35mm machine on
which they are installed.

Conclusion

For production recording in which
experience indicates that it is seldom
necessary to run forward or back at
high speeds, it does not appear that
torque motors contribute features which
justify the added weight, bulk, and

34

January 1952 Journal of the SMPTE Vol. 58

control complication. For re-recording,
however, high-speed operation in either
direction is a desirable feature and the
additional weight and bulk in such
stationary equipment is unimportant.

References

1. G. R. Crane, J. G. Frayne and E. W.
Templin, "A professional magnetic-re-
cording system for use with 35-, 17^-
and 16-mm films," Jour. SMPTE, 56:
295-309, Mar. 1951.

2. G. C. Davis, "An improved film-drive
filter mechanism," Jour. SMPE, 46: 454-
464; June 1946.

Discussion

Col. R. H. Ranger: Some time since, we
worked on torque motors for tape machines,
but Fm very much interested in this
application to 35mm. Isn't it true that
the reverse characteristic of a torque motor
applies only in the direction in which it
is trying to drive? In other words, if
you use it as the release motor, why, it
will not at all have the same characteristics
as it does for the take-up. In our work
we use a torque motor with a high-re-
sistance rotor for the take-up, but for the
release we found that d-c, applied to an
ordinary induction motor, gave very much
more the characteristics that we wish;
in other words, you get a very uniform
inverse ratio of torque to diameter with
that kind of a setup. Is that not correct?

A. L. Holcomb: The characteristics do
vary. However, when we use a 12-pole
motor with a characteristic similar to
that shown in Fig. 2 (curve A), the torque
is essentially the same ±172 rpm as it
is at standstill. As noted in this paper,
such a motor is a constant torque device
over the range of reel speeds.

Col. Ranger: As it goes through the zero
point, the action is entirely different, the
curve is practically flat, in the reverse
direction. Whereas if you use d-c on the
winding of an ordinary induction motor,
you get a very nice inverse curve.

Mr. Holcomb: That's true.

Col. Ranger: And it has the decided
advantage that it gives you high-speed
rewind; and so it gives you, I might
almost say, all the things you want.

Mr. Holcomb: That is quite true. You
would get a slightly better characteristic
with d-c, for drag duty, than you would
when reversing the direction of rotation
against a-c torque. This difference may
well be worth while for tape machines,
but for sprocket-type machines, the differ-
ence in performance is not apparent.

Col. Ranger: Plus the opportunity to
have a d-c brake.

Mr. Holcomb: Yes, a brake appears to
be very necessary. For sprocket machines
a "normally on" electromechanical brake
seems preferable in order to provide some
reel stiffness during the threading operation
when there may be no electrical excitation.

A. L. Holcomb: Film-Spool Drive

35

Heat-Transmitting Mirror

By G. L. DIMMIGK and M. E. WIDDOP

Radiant energy incident upon a glass plate can be divided into transmitted
and reflected bands by the interference effect in thin films of dielectrics de-
posited on the glass. The mirror described here reflects over 95 % of incident
visible light and transmits a large part of the energy beyond 7000 A. Such
mirrors have been produced and typical transmission characteristics are
shown. Several arrangements for use of such a mirror with a carbon arc
are also shown.

J. HE PROBLEM of producing "cold
light" has occupied the attention of
scientists and engineers for many years.
A number of methods have been success-
fully employed for reducing the relative
amount of radiant energy which lies
outside the visible spectrum. One ap-
proach to the problem is to employ a
light source which radiates a large
portion of its energy in the visible spec-
trum. The fluorescent lamp and the
mercury-vapor lamp are examples of
this type of source. Unfortunately, the
unit brightness of the fluorescent lamp
is too low to have much application in
optical systems of the projection type.
Fluorescent lamps are, however, used
extensively for general lighting where
the area of the source can be relatively
large. High-pressure mercury-vapor
lamps are capable of producing large
values of brightness, but they are de-
Presented on October 19, 1951, at the
Society's Convention at Hollywood, Calif.,
by P. J. Herbst, for the authors, G. L.
Dimmick and M. E. Widdop, Radio
Corporation of America, RCA Victor
Div., Camden 2, NJ.

ficient in red energy, and a large part
of their radiation is concentrated in a
number of discrete lines. The addition
of cadmium vapor into a mercury-vapor
lamp greatly improves the distribution
of energy in the visible spectrum and
makes this type of lamp a potential
competitor to the carbon-arc and the
incandescent lamp for application in
projection-type optical systems.

Another approach to the problem is
to employ a carbon-arc or incandescent
light source and to remove as much of
the infrared energy as possible with the
aid of absorption filters or with heat-
reflecting mirrors. Absorption filters
may be made of special heat-absorbing
glass or they may be cells covered on
both sides with ordinary glass and
having a liquid, such as water, flowing
continuously through them. The heat-
absorbing glass niters usually require
a current of air to flow past the two
surfaces to carry away the heat. A
well-known type of heat-reflecting mirror
is produced by evaporating a thin film
of gold onto one surface of a plate of
The thickness of the gold may be

36

January 1952 Journal of the SMPTE Vol.58

WAVELENGTH IN MILLIMICRONS

Fig. 1. Transmission curve
of a typical dichroic.

such that its transmission is maximum
for green light and its reflectivity is
high in the infrared region of the
spectrum. Heat-reflecting mirrors of
this type have a very limited application
because the transmitted light is peaked
in the green and the transmitting
efficiency is low even at its peak.

Still another approach to the problem
is to use the principle of interference
in thin films to build up the reflectivity
for light within the visible spectrum and
to permit the infrared energy to be
transmitted. It is toward this solution
to the problem that the present paper is
directed. The use of multiple films for
the production of dichroic mirrors has
been covered in the literature and will
not be described in detail here. It is
sufficient to say that efficient dichroic
mirrors may be made by evaporating
on glass alternate layers of two trans-
parent dielectric materials, one of which
has a relatively high index of refraction
while the other has a lower index of
refraction. The thickness of each layer
is usually made to be J wavelength
for light of the color which is to be
reflected. It is possible to make dichroic
mirrors which reflect as much as 95%
of the light of one color and transmit
90% or more of the light of another
color. A typical curve for a dichroic
reflector is shown in Fig. 1 . The peak
reflection occurs at about 450 m/i while

WAVELENGTH IN MILLIMICRONS

Fig. 2. Transmission curve of dichroic
consisting of two sets of layers.

the peak transmission occurs at about
650 m/*.

One of the important characteristics
of a cjichroic reflector is that the absorp-
tion of visible and infrared radiation
can be made negligibly small. This
means that radiation which is not
reflected from the multilayer film is
freely transmitted through this film.
It was this property which gave the
authors the idea for a heat-transmitting
mirror which would reflect efficiently
only in the visible portion of the spec-
trum. The idea was to deposit several
sets of multilayer dichroic films on the
surface of a plate of glass. Each set
would be so controlled as to cause its
peak reflection to occur at a different
wavelength. The peaks would be
equally spaced through the visible
spectrum so that all portions of this
spectrum would be reflected efficiently.
Light which did not reflect from the
outermost dichroic film would pass
through this film to one of the inner
films, where it would be reflected and
would then pass back through the outer
films to the surface.

The first test of the idea was made
several years ago in an RCA Advanced
Development Laboratory in Indian-
apolis. Two sets of dichroic films were
deposited in succession on the surface
of a plate of glass. The thickness of
the layers of one set was so controlled

Dimmick and Widdop: Heat-Transmitting Mirror

37

Fig. 3. Transmission characteristics of a heat transmitting mirror.

Since there is no appreciable absorption,
this means that the average reflectivity
over the visible spectrum is more than
90%. Beyond 700 m/i, the transmission
rises rapidly. The average transmission
between 700 m^ and 2.5 M is about 80%.
Since most of the energy from a high-
intensity carbon arc is below 2.5 M> the
transmission characteristics of the heat-
transmitting mirror beyond that wave-
length are not shown in Fig. 3. How-
ever, the transmission has been measured
out to 8 /n, and shows a sharp drop
just beyond 2.5 ». The average trans-
mission between 2.75 and 4.25 /x is
about 50%. Beyond 4.25 there is
another sharp drop and the transmission
from 5.25 to 8 /* is about 1%. The first
drop in transmission is characteristic
of absorption due to water vapor, and
the second is characteristic of absorption
of glass. It is unlikely that there is
appreciable reflection by interference at
this part of the spectrum, since the de-
posited films are thin in comparison to
the wavelength.

The effectiveness of the heat-trans-
mitting mirror was determined by
measurements made with the arrange-
ment shown in Fig. 4. A beam of
radiant energy from a high-intensity
arc lamp, a, was directed toward the
heat-transmitting mirror, b, placed at
an angle of 45° with the axis of the

Fig. 4. Setup used for measuring heat
transmission and reflection of mirror.

as to make the peak in reflectivity occur
at 490 m/z. The other set of layers
was made to have its peak in reflectivity
at 650 mfjL. The transmission curve of
the completed mirror is shown in Fig. 2.
As expected, a curve with a double
hump was produced by the above pro-
cedure and the efficiency of the reflector
was greatly improved. The results of
the first tests were so encouraging as to
warrant a systematic study of the
different parameters upon which the
overall effectiveness of a heat-trans-
mitting mirror is based. Development
work on this project continued inter-
mittently for several years. The im-
provements which were made resulted
in mirrors having a degree of reflectivity
which is greater than that of a back-
silvered glass mirror.

The curve in Fig. 3 shows the trans-
mission of one of the improved designs
as a function of wavelength. It will be
observed that the average transmission
from 400 to 700 m» is less than 10%.

38

January 1952 Journal of the SMPTE Vol. 58

WAVELENGTH IN MICRONS

Fig. 5. Radiant power vs. wavelength for blackbody operating at 5500 K.

beam. A portion of the energy passed
through the mirror and was absorbed
by a black solution in the liquid cell, c.
The remainder of the energy was re-
flected from the mirror, b, and was
absorbed by a black solution in liquid
cell, d. The liquid cells were identical
in size and each contained the same
amount of a mixture of water and India
ink. An accurate thermometer was
placed in each cell and the liquid was
allowed to come to room temperature
before turning the arc lamp on. The
arc lamp was started and allowed to
stabilize after which the shutter was
opened. The liquid in both cells was
stirred constantly and temperature read-
ings were taken once each minute for
ten minutes. The temperature readings
from both cells were plotted against
time, and a smooth curve was drawn
through the points. Straight lines were
drawn tangent to each of these curves
at the starting point, where the liquid
was at room temperature. The slope
of each of the straight lines is propor-
tional to the rate of absorption of
energy. The ratio of the two slopes is,
therefore, a measure of the ratio of the
total energy reflected from the mirror
to the total energy transmitted through
the mirror. In the case of the high-
intensity arc, the above measurement

revealed that 46% of the total energy
was transmitted, while 54% was re-
flected. Another measurement made
with a 750-w incandescent lamp as a
source revealed that 75% of the total
energy was transmitted, while 25% was
reflected. These measurements were
made with the mirror at 45° for con-
venience. A test was made to determine
the change in transmitted energy when
the position of the mirror was shifted
from 45° to normal-to-the-beam. There
was no significant change.

The energy reflected from the mirror
may be divided into two parts. The
first part is due to the useful visible
light between the limits of 400 and 700
m/z. The second part is the unwanted
infrared energy which the mirror fails
to transmit. The first value can be
obtained from a curve of radiant power
versus wavelength for the light source
operating at a temperature of 5500 K.
This is the approximate color tempera-
ture of a high-intensity carbon arc of
the type used for motion picture pro-
jection. By measuring the area under
the whole curve in Fig. 5 and comparing
this with the area under the visible por-
tion only, it is found that about 35% of
the total energy from a high-intensity
arc is radiated within the visible spec-
trum. Using this value, together with

Dimmick and Widdop: Heat-Transmitting Mirror

39

HEAT REFLECTED
BY MIRROR

HEAT TRANSMITTED
BY MIRROR

the previously obtained values of total
reflected and transmitted energy, we
can easily determine the overall per-
formance of the heat-transmitting mirror.
This is shown by means of a chart
(Fig. 6). From this it may be seen
that the mirror transmits more than
two-thirds of the unwanted heat due to
infrared radiation. It transmits nearly
half of the total radiation with a loss
of less than 10% of the visible light.

When used with an incandescent
source, the performance of the heat-
transmitting mirror is even more im-
pressive. In this case, 75% of the total
energy of the lamp is transmitted
through the mirror with a loss of less
than 10% of the visible light. A gas-
filled incandescent lamp operating at a
color temperature of 3000 K radiates
about 15% of its energy in the visible
spectrum between 400 and 700 m^i.
Nearly 85% of its energy is radiated in
the infrared region between 700 m/x and
infinity. The second chart in Fig. 6
shows how the heat-transmitting mirror
performs when the light source is an
incandescent lamp. About 88% of
the unwanted heat energy due to infra-
red radiation is removed by the mirror.
Seventy-five percent of the total heat

Fig. 6. Distribution of energy by
reflection and transmission using heat-
transmitting mirror: A, high-intensity
carbon-arc source; B, incandescent-lamp
source.

energy is removed, with a loss of less
than 10% of the visible light.

The heat-transmitting mirror might
be used in a number of ways to reduce
the temperature of the film as it passes
through the gate of a motion picture
projector. Figure 7 shows an arrange-
ment in which multilayer films replace
the usual silver reflecting layer on the
convex surface of the reflector in a
motion picture projector. The glass
reflector shell, c, has its convex surface,
a, coated with the evaporated films
which transmit a large part of the heat
and reflect most of the light. A cor-
rugated metal shell, b, encloses the back
of the reflector and is spaced away from
the evaporated films. This metal shell
serves the double purpose of protecting
the reflecting surface from contamination
or mechanical damage, and absorbing
the radiation so that the energy may be
dissipated by convection currents.

One possible disadvantage of the
scheme shown in Fig. 7 is that elaborate
and expensive equipment might be re-
quired to evaporate thin films with the
required uniformity on the convex
surface of the reflector. This disad-
vantage would be overcome in the
arrangement shown in Fig. 8. Here

40

January 1952 Journal of the SMPTE Vol. 58

,„;;>-•« o

Fig. 7. Sketch of projection optics
using a heat-transmitting film
on back surface of reflector.

Fig. 8. Projection optics using heat-
transmitting film on front surface of
reflector, protected by another glass.

the multilayer film, a, is on the concave
surface of the reflector where it would be
relatively easy to obtain the required
uniformity. In order to protect the
surface from contamination and me-
chanical damage, a thin-glass shell is
placed in front of the reflector and in
contact with it all along the rim. This
shell might be removed for cleaning,
and it could be replaced when its con-
cave front surface gets badly pitted by
hot particles from the carbon arc.

Still another arrangement of the heat-
transmitting mirror is shown in Fig. 9.
The evaporated films, a, are placed on
the back surface of a flat plate of glass,
c. A thin, corrugated-metal housing
encloses the back of the reflector and
keeps it clean and free from mechanical
damage. The heat is dissipated by
convection currents of air flowing past
the thin metal housing. This arrange-
ment, with a single heat-transmitting
mirror and a normal silver-backed
concave mirror, requires a right-angle
bend in the illuminating system. If this
is a disadvantage, it could be overcome
by the use of two heat-transmitting mir-
rors like those shown in Fig. 9, arranged
to make an offset system with two right-
angle bends. This would result in a
two-stage heat filter which would be even
more effective than shown by the charts
in Fig. 6. If desired, a two-stage heat
filter could also be obtained by using a
combination of the systems shown in
Fig. 7 or Fig. 8 with the system shown
in Fig. 9.

Fig. 9. Flat heat-transmitting mirror
used in beams from silvered reflector.

References

1. J. Strong, "On a method of decreasing
the reflection from nonmetallic sub-
stances," J. Opt. Soc. Am., 26: 73-74,
Jan. 1936.

2. C. H. Gartwright and A. F. Turner,
"Multilayer films of high reflecting
power" (Abstract), Phjs. Rev., 55: 1128,
June 1939.

3. G. L. Dimmick, "A new dichroic re-
flector and its application to photocell
monitoring systems," Jour. SMPE, 38:
36-44, Jan. 1942.

4. G. J. Koch, "Interference mirrors for
arc projectors," Jour. SMPTE, 55:
439-442, Oct. 1950.

Discussion

Frank Carlson: What is the maximum
temperature the mirrors will stand?

P. J. Herbst: No tests have been made to
determine the maximum temperature the
mirrors will stand. No damage has re-
sulted from tests using the mirrors in the
beam from a high-intensity arc.

D. B. Joy: Has it been found that these
films will stand up satisfactorily in ordinary

Dimmick and Widdop: Heat-Transmitting Mirror

41

projection lamps used in motion picture
projection?

Mr. Herbst: The mirror has been sub-
jected to the beam from a high-intensity
carbon arc, focused to about a 3-in. diam-
eter spot, for several hours without
damage. However, actual life tests have
not been made on the mirrors.

Mr. Joy: The Motion Picture Industry
should be grateful to you people for having
done some work along these lines. We are
faced with a very practical and immediate
problem of trying to keep the heat down on
the film, while we are trying to force a
great quantity of light through the film in

out-door theaters. Therefore, anything
along this line, coming at this time, will
be of great help in giving us better movies,
and that's the thing we want.

Mr. Strickland: If you don't have the
shield in front of the mirror, and you get
pretty well pitted, does that have the
tendency to lower your light?

Mr. Herbst: You mean the dichroic on
the front of the mirror next to the carbon
arc?

Mr. Joy: That's right.

Mr. Herbst: This would not be recom-
mended. The dichroic surface should be
protected.

42

January 1952 Journal of the SMPTE Vol.58

Recent Improvements in Silencing
Engine-Driven Generators

Bv L. D. GRIGNON

A. gasoline engine-driven, 120-v, d-c generator of 150-kw output for set
lighting on location has been improved. The enclosing wall structures,
materials and carburetor air-intake were changed. When mounted on a
trailer the exhaust and radiator and noise are considerably reduced by methods
described. The improvements permit sound recording with the generator
as close as 250 ft under reasonably favorable circumstances and not exceeding
750 ft for critical conditions. Considerable saving in production costs results.

JL HE SILENCING of noisy equipment
used in the production of motion pictures
has been a continuous problem since
sound recording became a part of the
industry. One of the most offending
equipments has been the engine-driven
generator used for set lighting on loca-
tion, although the accumulated contri-
butions of many workers have produced
considerable improvement over the
initial situation.

The problem which led to the improve-
ments reported herewith was posed as
follows: Given a gasoline-engine-driven,
direct-current generator set of 150-kw
capacity of the basic design described
by Hankins and Mole1 and of similar
size, what changes or modifications can
be made to obtain a plant with less
noise?

The difficulty in work of this kind is
to find the best compromise between

Presented on October 19, 1951, at the So-
ciety's Convention at Hollywood, Califor-
nia, by L. D. Grignon, Twentieth Century-
Fox Film Corp., Beverly Hills, Calif.

size, weight, cost, operating features
and quietness. As is well known,
quietness is not compatible with the
first three items.

The design previously produced by
the Mole-Richardson Company was
carefully studied with these conclusions:
that some change in structure shape
would permit mounting the engine and
generator on a noise-insulated subbase;
that it was reasonable to expect improve-
ment in wall design; and that better
sound absorptive materials might be
used. The design of the subbase was
undertaken by the Mole-Richardson
Company with the application of con-
ventional vibration insulation methods.
Further, in consultation with the same
company and the engine manufacturer,
it was concluded that the engine could
be completely enclosed and adequately
cooled by water only.

The next step, in order, was to con-
sider the enclosing structure. Figure 1
illustrates the basic layout.

January 1952 Journal of the SMPTE Vol. 58

43

-ENGINE EXHAUST

COOLING-AIR
EXHAUST

ENGINE AIR
INTAKE

AIR CLEANER

RADIATOR

Fig. 1. Basic plan of enclosure.

Three principal factors determine the
efficacy of sound-insulating structures.
These are absorption, transmission
through the various media and element
resonance. When the maximum amount
of absorption at the sound source can be
provided, the two latter problems are
somewhat simplified. Of first impor-
tance, therefore, is the selection of ab-
sorptive materials for the inside surfaces
of the enclosing structure. Again, a
balance of thickness, weight and ab-
sorption must be determined and, to
complicate matters, the material must
be fireproof in this particular application.

A material having a good balance of
these factors is known commercially as
Spraycoat. This is a shredded asbestos
material with a suitable high-tempera-
ture flameproof binder. It is applied
by spraying and tamping, preferably on
wire mesh or plaster wire. If some air
space is provided behind a f-in. layer
of this material, very good low-frequency
absorption is attainable and since it is
of a semiporous soft nature, the high-
frequency absorptive qualities are excel-
lent. Whenever the material is applied
in this manner, the support must be
reasonably taut in order that the tamp-
ing operation will be satisfactory. The
material is not mechanically strong and
in areas where this is of importance it is
desirable to protect the surface with wire
mesh. An additional mechanical help

is to spray the surface with a thin
application of water-base casein paint.

Wherever possible, the interior sur-
faces of the enclosing structure have been
covered with Spraycoat. In a few
specific instances, for mechanical reasons,
an air-duct felted material with an
asbestos-cloth facing known as Dux-
Sulation has been used.

Panel or structure resonance is of
great importance for, if resonance exists,
the structure will apparently have small
transmission loss in contradiction to the
predicted loss value based on the ma-
terials used. One method used in the
past for minimizing resonance consisted
in designing the panels in random sizes.
This is of value in that such resonances
as do exist are distributed in the fre-
quency spectrum and the added bracing
provides some damping. A better solu-
tion would be some method which elimi-
nated resonance, regardless of the panel
spectrum distribution. This implies
that damping is the important factor
and accordingly design effort was di-
rected to this specific aspect.

The most generally used panel-
damping method is the application of
non-hardening asphaltic or rubberlike
materials. These do provide some
damping and lower the panel-resonance
frequency by virtue of the added weight,
but in many ways this method is not
very satisfactory. The most favorable

44

January 1952 Journal of the SMPTE Vol.58

Fig. 2. Types of wall sections showing juncture of engine
compartment and cooling-air exit duct.

1. Stainless-steel, 20-gauge Type 302
plus 2 layers Brownskin building
paper cemented with Minn. Mining
EC-1025.

2. Dux-Sulation ^ in. thick cemented
with Dux-Sul Glue.

3. Celotex \ in.

4. Minn. Mining Undercoater EC-831,
| in. thick.

5. Aluminum 0.040 2S plus 1 layer
Brownskin Grizzly Bear 30/40 build-

X f

ing paper cemented with Minn.
Mining EC-1025.

6. Metal Lath 3.4 Ib.

7. Spraycoat f in.

8. Hardware cloth 17-gauge, f in.
in. mesh.

9. Stainless-steel, 20-gauge Type 302
plus one layer Brownskin Grizzly
Bear 30/40 building paper cemented
with Minn. Mining EC-1025 to steel
and 0.040 aluminum 2S.

10. Air space.

aspect of the so-called "undercoaters" is
low cost. Certain heavy, creped build-
ing papers have been found to provide
excellent damping when cemented be-
tween two panels in laminate form.
In such form, panel vibration establishes
sheer forces within the damping paper
which dissipates the energy. A single
metal sheet with cemented paper is
reasonably well damped and may be
used, when indicated, with good results.
With good damping and high absorp-
tion attained, attention may be directed
to the reduction of sound transmission.

The classical considerations of trans-
mission loss now more nearly apply in
the practical case by virtue of the
significant reduction in resonance. It
is known that two separate structures
have greater transmission loss than the
same total amount of material in one
structure, but this scheme complicates
the construction, makes for increased
size, increased weight, difficulties of
access and maintenance. Wall struc-
tures with air spaces also provide more
transmission loss than the same amount
of material in a solid wall, but are

L. D. Grignon: Silencing Generators

45

usually more conservative of space
than the separate wall design, although
lower attenuation can be expected.
An additional factor to be considered
is that motion picture dialogue sound
recording is usually attenuated at the
low-frequency end relative to midband
and a high-pass filter is used. It is,
therefore, illogical to provide high
attentuation below 100 cycles.

Wall Sections

Considering all items discussed above,
wall sections as shown in Fig. 2 were
devised. The illustration shows the
joint between three different sections
as follows: the left-hand section be-
tween the engine compartment and the
outside; the right-hand section between
the cooling-air exit duct and the outside;
and the vertical section between the
engine compartment and the cooling-air
exit duct. The materials are shown in
the illustration, but the outer skin
requires more explanation.

For durability and appearance the
outside material is stainless steel damped
with two layers of creped building paper.
Since the paper is creped in only one
direction and light in weight, two layers
at right angles give good damping.
The paper is cemented together and to
the metal with a Minnesota Mining
nonhardening adhesive. The single
metal-paper laminate is used in this
location because it is desirable to face
the inside surface with an absorptive
material to minimize reflections in the
adjacent air space and to lower the unit
weight of the section.

Both the air duct and the engine
compartment must be faced with highly
absorptive material. Because of a di-
mension limitation, the wall between
these regions is modified slightly to
include a metal-paper-metal laminate
panel. The use of the metal-paper-
metal laminate maintains the transmis-
sion loss at a suitable value which would
otherwise have been decreased, due to
the removal of the damped aluminum

panel used in the other principal wall
section. Obviously, the air-duct wall
requirements are considerably less severe,
so here the principal attenuation is
provided by a metal-paper-metal lami-
nate with Spraycoat applied directly on
the inside.

The actual attenuation of the main
wall sections is not precisely known due
to a lack of facilities for this type of
measurement. By calculation and con-
sidered judgment, it seems reasonable
to assign an attenuation value of 43 to
50 db at midband and somewhat less
at 100 cycles. The attentuation is,
however, ample, since by methods to
be mentioned later whereby the residual
noise is considerably reduced, noise
through the wall sections is still far
below all other sources.

As a matter of refinement and pre-
caution, all structural members are
filled with Dux-Sulation. It will also
be noted in the illustration that the wall
section can be disassembled from the
outside if it is necessary for any reason.

Doors for access to the equipment are
a necessity and in the past the most
popular idea has been to use a stepped
jamb with multiple rubber or felt gaskets
and a latch which compresses the frame
and jamb upon the gaskets. This
construction is commonly known as the
"icebox door." This method is satis-
factory as a sound-stopping scheme,
but is cumbersome, requires heavy
hardware and loses effectiveness as the
gaskets are damaged or deteriorate with
age. If a door is made with sufficient
accuracy so that the residual crack is of
small dimensions, only high frequencies
will be transmitted by this path. Fur-
ther, high frequencies are readily ab-
sorbed by many different materials, so
that any material which can be intro-
duced within the residual crack will
considerably reduce the high-frequency
transmission. This is the basic design
idea used in the subject project, and the
actual construction is shown in Fig. 3.
The design requires no especially heavy

46

January 1952 Journal of the SMPTE Vol.58

EXTERIOR

ENGINE COMPARTMENT

Fig. 3. Section of engine compartment wall including access door construction.

1. Stainless-steel, 20-gauge type 302
plus 2 layers Brownskin building paper
cemented with Minn. Mining EC-
1025.

2. Dux-Sulation \ in. thick cemented
with Dux-Sul glue.

3. Celotex \ in.

4. Minn. Mining Undercoater EG-831,

in. thick.

5. Aluminum 0.040 2S plus 1 layer
Brownskin Grizzly Bear 30/40 building
paper cemented with Minn. Mining
EC-1025.

6. Metal lath 3.4 Ib.

7. Spray coat f in.

8. Hardware cloth 17-gauge, f in. X f
in. mesh.

9. Air space.

hardware, no pressure is required to
close the door and there is no audible
noise transmission through the door
joint. The gasket shown serves prin-
cipally as weather stripping and makes
the final closure of the door crack. If
the crack can be held to y1^ in. or less
by good construction, about a 40-db
noise attentuation can be expected.
An empirical rule to estimate the loss
through a joint as shown is to allow 1 db
for each unit of sound-path length, the
unit being equal to the crack width.

As is apparent from Fig. 1 , the genera-
tor compartment is open to the air
through the water-cooling radiator and,
therefore, it would be poor design to
provide the same excellent wall structure
for this volume as was used for the
engine compartment. Accordingly, the
wall structure used in these areas was
the identical simple section of the exit
air duct.

Air Exit Duct and Mufflers

The cooling-air exit duct is of interest
in that no baffles or turns are used.

The duct is straight and open from the
generator compartment to open air, a
distance of 63 in. Two vertical sepa-
rators are placed in the duct to provide
more surface for absorptive Spraycoat.

The duct has cross-section dimensions
of 15 in. X 50 in. and absorbs the
generator and fan noises so well that
this potential source of noise needs no
other attention.

Several good engine-exhaust mufflers
are available, hence the only precaution
to be observed on this item is that the
muffler itself does not become a noise
source. This difficulty can be mini-
mized by wrapping the muffler with sheet
asbestos held tightly to the muffler
surface with an external wrap of sheet
metal.

Carburetor Intake

A situation contrary to that of the
engine exhaust concerns the engine
carburetor air-intake. If outside cool
air is to be used for the engine, then the
noise from this source needs considerable
attention when, as in this case, the other

L. D. Grignon: Silencing Generators

47

IZ'DIA

AIR- INTAKE

VOL. ieoo cu IN

VOL 137 CU

a- VDIA HOLES

OUTLET TO
CARBURETOR

16-Vo.A HOLES

a" DIA

INTAKE INSECT SCREEN AND
EXTERNAL RESONATOR

HOLES

INTAKE ACOUSTIC SOUND FILTER & AIR CLEANER

Fig. 4. (A) Basic design of carburetor air-intake acoustic filter as combined with
standard Vortox air cleaner. (B) Design of external resonator and insect screen

for carburetor air-intake.

•TOTAL NOISE WITH AIR FILTER ONLY

3.

•TOTAL NOISE WITH SOUND FILTER ADDED

--'"I

S ? 8

FREQUENCY (CYCLES/SEC )

I I I I I !

§ I 1

Fig. 5. Noise frequency distribution, carburetor air-intake. One-half

octave band frequency analysis of source noise showing attenuation due

to acoustic filter and external resonator.

48

January 1952 Journal of the SMPTE Vol. 58

major noise sources have been minimized.
The usual sound attenuation is that due
only to the dirt filter; therefore, an
acoustical low-pass filter was designed
having the configuration of Fig. 4A.
It will be noted that two dissimilar
volumes are used: the through pipe is
smaller in cross section than the car-
buretor intake and the acoustic filter
has been combined with the Vortox
air cleaner to make one package.

The classical theory for design of
acoustic low-pass filters assumes that
such devices are inserted in the middle
of a long pipe or conduit. In the

practical case under discussion this is
not true, since at one end there is a
relatively short pipe, while at the other
end there is the volume of the air cleaner
and the acoustic resistance of the oil-
saturated meshes. These discrepancies
were neglected and the large single-
section filter computed by the simplified
formulas of Stewart,2 rather than those
more complete and complex equations
of Mason.3 One difficulty with acoustic
filters concerns the terminating im-
pedances, since an impedance match
occurs only at discrete frequencies. In
general, if the filter matches the con-

Fig. 6. Exploded view of radiator cool-
ing-air acoustic absorptive-type traps.

-CAR6URETOR INTAKE WITH AIR FILTER

-TOTAL NOISE LEVEL COOLING-AIR- INTAKE

* ! nsTs

FREQUENCY (CYCLES/SEC )

Fig. 7. Noise frequency distribution — radiator end related to carburetor

intake noise with air filter only, full load. One-half octave band analysis of

source noise from radiator end after installation of absorptive trap.

L. D. Grignon: Silencing Generators

49

necting impedances at some frequency
low in the passband, superior perform-
ance may be achieved. The smaller-
sized pipe in the filter assists in re-
ducing the filter impedance to improve
the match, but is not of such small size
as to restrict air flow.

Another difficulty with acoustic filters
results from passbands above the cutoff
frequency. When additional series-
connected-filter sections with different
cutoff frequencies, other than multiples
of the preceding sections, are used, the
spurious passbands are minimized. This
is the reason for the section containing
the second smaller volume. The cutoff
frequencies of the two sections are 80
and 350 cycle/sec, respectively. Using
the device as illustrated, considerable
improvement obtains, although there is
still a peak of transmission around 900
cycle/sec. Discrete frequency bands
above 300 cycles may easily be elimi-
nated by small Helmholtz resonators
coupled to the intake pipe at the open-
air end, and in this instance may be
combined with the insect screen. Such
a resonator is shown in Fig. 4B and the
total effect of the above-described filter
and resonator is shown in Fig. 5. The
measurement of total noise level shows
a 14-db improvement. The frequency
distribution was measured in one-half
octave bands from a recording made on a
standard production dialogue recording
channel.

In this particular application the two
volumes of the filter may actually be
operating as resonators coupled to the
pipe, rather than as a true low-pass
filter. This point needs investigation
before a clear-cut understanding of the
situation may be available, but in any
event the configurations described have
adequate performance for the require-
ments. The loss in horsepower due to
the acoustic filter is 0.33% at sea level,
full power, wide open throttle.

The principal remaining noise ema-
nates from the radiator end and results

from the large slow-speed fan and the
generator. There is very little that can
be done about this source without making
the plant considerably larger except to
absorb as much of the noise as possible.
This is done, as described, by using
Spraycoat and as much Dux-Sulation
around the generator end-bell as possible.

All of the access methods and operat-
ing features described by Hankins and
Mole have been retained in this overall
design.

As described above, the plant is a
complete unit of relatively low noise
level which can be shipped by any com-
mon means of transportation.

Further Quieting Methods

For the great majority of motion pic-
ture locations a generating plant as
described may be placed on a permanent
truck or trailer, provided that, when
required, it may be easily removed.
When considerable long-distance haul-
ing is to be done, a low-bed trailer is
of advantage. Consequently, the de-
scribed unit has been mounted on a
low-bed trailer with permanently in-
stalled fuel tanks. This procedure pro-
vides space which may be used for
additional silencing.

The plant was so placed on the trailer
that the added space came at the radia-
tor end. The volume from the radiator
to the end of the trailer was enclosed
with sheet metal, lined on the inside
with 1-in. preformed glass-wool sheet.
This enclosure included the fuel tanks.
Three sets of straight-through vertical
and horizontal partitions (commonly
called egg-crates) are placed in the air
stream, as shown in the exploded view of
Fig. 6. Each of these is approximately
16 in. long with all surfaces covered with
Dux-Sulation. The center set has the
vertical partitions constructed so that
a cross-sectional view of any two ad-
jacent partitions form an air path having
the approximate shape of a Venturi
tube. The three sets of partitions are
separated by free-air spaces of 14 in.

50

January 1952 Journal of the SMPTE Vol. 58

Fig. 8. Photograph of basic plant installed on trailer.

to 16 in. in length. Also, the number of
vertical and horizontal members are
different in each set so that if stacked
one upon another the combination would
appear as a set of irregular-sized open-
ings, each being smaller than any single
opening in any set of partitions. The
total area of the sound-absorptive ma-
terial within the traps is approximately
400 sq ft and though the air travel is
essentially straight through, a reduction
in noise power of 14 db is achieved.
The noise spectrum is shown in Fig. 7.
It is believed that considerable benefit
obtains from the free-air volume be-
tween the partition sets. There has
been no noticeable reduction in cooling
efficiency by the application of this
particular arrangement in the air path.
Indeed, there is some evidence that the
plant runs cooler under given conditions
with this sound trap.

With the noise reductions obtained by
the methods so far described, the exhaust
noise became noticeable. This source

is easily reduced by additional muffler
capacity and in this particular case was
most easily accomplished by the addition
of a second muffler essentially the same
as the permanently installed unit. The
second muffler was also lagged with
asbestos and sheet metal.

Final Plant

With all the methods and devices
described, the plant appears as in Fig. 8.
When necessity demands, the minimum
plant is removed from the trailer and is
used with some penalty in noise output
requiring longer cable runs or temporary
housing structures. When used com-
plete, as shown, the plant may be used
250 ft to 750 ft from the recording set,
the distance being determined by the
nature of the scene and the conditions
of the location. Distances of 300 ft to
400 ft are the usual placement for the
average scene. The savings in cost and
time on location resulting from close
generator placement are obvious.

L. D. Grignon: Silencing Generators

51

Acknowledgments

As is so often the case, this project
was successfully completed with the
assistance of many people. The plant
itself was contracted to the Mole-
Richardson Company where M. A.
Hankins was of considerable help in
designing the structure for the plant
enclosure; Standard Auto Body Com-
pany was of great value in suggesting
structure construction methods leading
to good mechanical design at reasonable
cost and for excellent assembly of the
enclosure; the Vortox Company was
very cooperative in making sample
acoustic filters combined with their
standard dust filter until a suitable
design was found. Lastly, the project
was guided by Walter Strohm and
Thomas T. Moulton of Twentieth
Century-Fox Film Corporation, Elec-
trical Engineering and Sound Engineer-
ing Departments.

References

1. M. A. Hankins and P. Mole, "Designing
engine-generator equipment for motion
picture locations," Jour. SMPTE, 55:
197-212, Aug. 1950.

2. George W. Stewart and Robert Lindsay,
Acoustics, D. Van Nostrand, New York,
1930.

3. Warren P. Mason, Electro- Mechanical
Transducers and Wave Filters, D. Van
Nostrand, New York, 1948.

Discussion

Anon: Thank you very much, Mr.
Grignon. On the egg crate, was there
any acoustic material?

L. D. Grignon: Yes, all the partitions are
covered with a felted flameproof material
5 in. thick.

David Joy: I noticed that you showed a
general lowering of the sound level and
also you had the curves showing the
lowering of the sound level for the indi-
vidual frequencies. Why are you so
interested in the individual frequencies,
if you have the general sound level low;
why do you have to worry about the
individual frequencies?

Mr. Grignon: Equipment noise seldom
has the sound energy uniformly dis-
tributed throughout the audible spectrum.
By making an analysis of the noise, cycle
by cycle, or in discrete bands as was done
in this instance, a determination can be
made of those portions of the frequency
spectrum contributing the greatest energy
relative to the total sound energy. In
some cases the noise source can then be
identified and corrections at the source
applied. When it is impossible to correct
the source, the greatest benefit can be
obtained by assuring that the external cor-
rective means is most effective in the fre-
quency bands containing the largest
percentage of the energy.

Harry R. Lubcke: Grig, could you estimate
what proportion of the weight was added
to the original weight of the engine-genera-
tor set by the sound -insulation job?

Mr. Grignon: I might make a guess at it,
but I think that we can probably get a
more accurate figure by asking Mr.
Hankins of the Mole-Richardson Company
whom I see in the audience.

M. A. Hankins: The total weight of this
particular engine-generator set is 10,660
Ib, including the sound-insulating housing
which weighs 2440 Ib. The weight of the
plant less housing is, therefore, 8220 Ib.
Assuming that the baffling added in front
of the radiator, etc., by Twentieth Century-
Fox weighs approximately 800 Ib, the gross
weight of all the sound-insulating com-
ponents is 3240 Ib, which is about 35%
to 40% of the basic weight of 8220 Ib.

52

January 1952 Journal of the SMPTE Vol. 58

Cinecolor Multilayer
Color Developing Machine

By JAMES W. KAYLOR and A. V. PESEK

The development of the various new and improved multilayer color films
emphasized the need for a standard-type motion picture film developing
machine that would be capable of handling any of the new types of multi-
layer color films. A machine of deep-tank, positive top-drive type embodying
bottom elevators, turbulation or spray facilities in all tanks and practical
flexibility, enabling it to be set up in any practical combination of solutions
and washes to develop the various types of multilayer color films available, has
been developed and put into operation as a production machine by the Cine-
color Corporation. A special arrangement of the geared drivehead allows
any of the racks to be removed without affecting the drive, and the drive has
been designed to provide for the attachment of desired auxiliary equipment.

WH

THEN THE Ginecolor Corporation
began the major conversion of its
facilities to the production of three-color
film, it was necessary to have a develop-
ing machine for the multilayer color
film which was to be used as the taking
medium. At that time it was decided
to try to design a standard-type machine
which would be flexible enough to de-
velop any of the color-coupling multi-
layer films available. Such a machine
could be set up as an experimental
machine to evaluate the possibilities of
the various films or as a production
machine to process any specific type of
film and still retain flexibility to permit

Presented on October 17, 1951, at the
Society's Convention at Hollywood, Calif.,
by James W. Kaylor and A. V. Pesek,
Cinecolor Corp., 2800 West Olive Ave.,
Burbank, Calif.

change-over from one system to another
with a minimum of rework.

The developing machine was de-
signed to operate at an average speed of
35 ft/min with a minimum and maxi-
mum speed of 10 to 60 ft/min. A
study of the developing techniques of the
several multilayer color films indicated
that 32 racks, each having a capacity of
approximately 100 ft or a time element
of about 3 min at 35 ft/min, would be
sufficient to provide for the different
combinations of solution and wash
times indicated for the various films.
Figure 1 shows diagrammatically the
general layout for several different
films.

In order to provide for different
combinations of solutions and washes
32 tanks are used, one for each film
rack. All tanks are identical, with side

January 1952 Journal of the SMPTE Vol. 58

53

LJS COLOR N6a- C.K.COLOR POSITIVE ANSCO POSITIVE £ NEGATIVE DuPQNT PO^ITIVP

START

&

CARBONATE -
REfcKIET^-
WASH J^

DEVELOPER"!

SPRAY WASH-
ACID STOP H
SPRAY WASH-{
BLEACH -£
SPRAY WASH-(
HYPO -[

WASH -t
FORMALIN —

xp

K

COLOR DEV-j

RINSE -
ACID STOP •{
WASH J"

BLEACH _f
WASH •{
HYPO 4

WASH -|

CAST

I

DEvELOPER-f

WASH —
FIX
WASH
BLEACH
WASH

CARJ

CARBONATE -
REM-JEfC^

•— i

1

1

II

-4=7

-J

{EF^-

-E^-

WASH J "
DEVELOPER-

•p^^

r=^

~F^

E — '-]

L— — ,

sE

\ =

1

5

1

i

1

i '

H — '

i '

1 '

L .

= — i

^E*

-^Eh

&^^-

£EF^-

SPRAY WASH-
ACIOSTOP -
SPRAY WASH-
BLEACH -£
SPRAY WASH-
HYPO

WASH — <

^^

"^^T

?=^^

^^

I '

:zi

1 1

1 1

^3

€EE^-

fErf3-

&=T

-&=•

7=^^-

WASH — |

T^3"

|—

I—'—1

1— 1

S — i

1

1

I '

1 '

1 '

1 — '

1

L— ,

1

"— 1

I —

1 — ]

1

1

1

1

1

I '

1 '

1 — -1

|

1

= 1

W

•Er^

^4

^4

\

f^—

1

1

L._n

,

1 ,

r J

I '

, '

, J

i ,

I ,

i ,

1 1

i --1

. '

, 1

r '

i ,

,L.._.n

i 1

I 1

—3

p. 1

1

.

t t t t

Fig. 1. Diagram of general layout for several color films.

inlets and outlets for supply, overflow
and drain and with six additional
fittings in each side to allow for the
insertion of turbulation or spray headers,
as shown in Fig. 2. On the drain side
of each tank are provided two overflow
fittings as well as a drain fitting at the
bottom, as can be seen in Fig. 3. The
additional overflow fitting is set lower
than the regular overflow to provide
for a submerged level control, when de-
sired, to eliminate excessive oxidation of
solutions if air is entrapped by overflow
into the usual open-type return line.

The tank drive frame can be seen
lowered into operating position in
Fig. 4 and raised for cleaning and
checking in Fig. 5. This is a tubular
frame, one side of which acts as an
integral gear and shaft housing for the
main gear shaft that drives all of the
film racks. Power is provided to the

head through a right-angle gearbox from
a telescoping power shaft which allows
the drive frame to be run in either the
raised or lowered position. The indi-
vidual film roller racks are attached to
the top of the tubular drive frame by
two identical castings. One carries a
helical gear which meshes with a similar
gear within the tubular shaft housing
and drives the top roller shaft through
a tongue-and-key joint; the other
carries a ball-bearing pin which sup-
ports the outboard or idler end of the
top roller shaft. An adjustable tierod
connects the two castings to maintain
alignment and stiffen the whole driving
head. Each casting supports two hanger
rods, which in turn support the back-up
roller (midway between top and bottom
rollers) and the bottom roller elevator,
as seen in Fig. 5. The extreme lower
ends of the hanger rods are clamped

54

January 1952 Journal of the SMPTE Vol. 58

Fig. 2. Inlet side of tanks, with chemical supply lines and turbulation headers.

Fig. 3. Drain side of tanks.
Kaylor and Pesek: Cinecolor Developing Machine

55

with phenolic hangers and tied together
with another tierod to insure proper
spacing. The phenolic hangers also
act as guides when the rack assembly is
being lowered or raised in the tank.

Positive drive is applied to the film
by means of a single drive sprocket
setscrewed to each top roller shaft.
Eleven film rollers are mounted in
connection with the sprocket on a full
standard rack, and are free-running on
the shaft with molded, phenolic bearing
inserts which also maintain correct
spacing of the rollers. Additional drive
is obtained, if necessary to relieve
tension, by slightly spring-loading the
rollers together.

The bottom roller elevator consists
simply of two phenolic hangers, each
clamped to two short lengths of stainless-
steel tubing that ride up and down
on the two hanger rods on each side.
The lower shaft is slipped into holes in
the hangers and centered by setscrew
collars at each end. The whole as-
sembly is tied together by a separator
assembly which also keeps the film
strands properly aligned with the bottom
rollers. These are free-running on their
shaft and spaced by molded, phenolic
bearing inserts, the entire weight of the
bottom elevator assembly being sup-
ported by the film strands.

The complete drivehead can be raised
or lowered by six stainless-steel cables
connected to an electrical hoist. The
cables are conveniently spaced three to
a side, as shown in Fig. 4, and are
adjustable with turnbuckles to equalize
tension and insure a straight and even
lift. The hoist is provided with an
electrically operated brake which mini-
mizes coasting and holds the head in
any position desired. Two limit switches
are mounted on the ceiling, one at
each end of the head to prevent it from
being raised too far.

Film is fed into the machine from a
clutch spindle mounted on a feed table.
The film passes first over a feed elevator
of approximately 100-ft capacity which

allows about 3 min for splicing on new
rolls. After passing through the tank
section, the film moves through an air
blowoff or squeegee and into the dry
box. The blowoff unit is hinged so that
it can be tipped down out of the way
when the drivehead is to be raised to the
ceiling. Air is supplied to the blowoff
from a Nash waterseal compressor at
about 2 to 3 psi, through a manifold
which is also extended the length of the
tank section with outlets at the center
and feed end for connection to various
auxiliary equipment.

The dry box, shown in Fig. 6, is of
sheet-aluminum construction with three
sliding glass doors on each side. The
drive is a positive top drive, similar in
construction to the tank drive, and
consists of a tubular frame with the
drive gear shaft and castings previously
described, but mounted inverted, with
the castings on the bottom, on four
corner supports. The drive has six
banks of rollers or film racks, the lower
rollers being ball-bearing and mounted
on elevators. Centered between the
top and bottom rollers on each rack
there is a back-wiping roller to remove
any drops or streaks from the base side
of the film as well as to keep the film
strands separated. The dry box holds
approximately 900 ft of film giving about
27 min drying time at 35 ft/min. Air is
supplied to the dry box under controlled
temperature and humidity conditions
from a heating and blower unit. Tem-
perature and humidity are provided by
a steam supply and are regulated by
Minneapolis-Honeywell controllers.

The dried film is taken up as it leaves
the dry box on either of two power-
driven, friction-clutch take-ups (two are
provided for rapid change-over) which
are shown in Fig. 6. Power for the
take-ups is supplied from a separate
gear-head motor of sufficient output
speed to drive the take-ups at the
maximum speed of 60 ft/min.

The machine speed controls and speed
indicator are mounted on a panel be-

56

January 1952 Journal of the SMPTE Vol. 58

Fig. 4. Tank section of machine with drivehead in operating position.

Fig. 5. Drivehead raised for maintenance or thread-up.
Kaylor and Pesek: Cinecolor Developing Machine

57

tween the tank section and the dry box
cabinet. Speed of the machine is
governed by a speedranger, located
below the floor, which is adjusted with
a chain drive from the speed-control
knob on the panel. Also on the panel
is one of three stop-start stations, the
other two being mounted on the feed
table and take-up table, respectively.

Water is supplied to the various wash
tanks through a common header. Deep
wash tanks are fed through the bottom
inlet and overflowed into the open drain
at the top of the tank. Spray wash
tanks are fed directly through the six
spray headers inserted through the sides
of the tank, and drained through the
bottom drain.

The chemical solutions are supplied
to the proper tanks from storage tanks
located in the basement through Saran
piping and headers. Solution return
is effected by gravity flow from over-
flows to the storage tanks. Individual
pumps are employed for each solution,
and flow is metered through Schutte-
Koerting Rotameters. Submerged
drains are provided on the developer
tanks and the level is controlled by
means of wet-type liquid level switches
actuating solenoid valves on the lower
end of the drain lines.

Turbulation of the chemical solutions
is accomplished where desired by with-
drawing the solution from the machine
tanks through the bottom drains and
pumping it back through the spray
headers. The turbulation flow is also
metered through Rotameters.

Temperature control of the solutions
is maintained by passing either hot or
cold water through stainless-steel, heat-
exchange coils installed in the solution
storage tanks. However, the problem is
primarily one of cooling the solutions,
so Taylor Temperature Recording Con-
trollers are used to regulate the chilled
water supply to each coil, hot water
being used only to warm the solutions
after shutdowns during cold weather,
when occasionally the solution tempera-

ture drops below the required point.
A portion of the basement section can
be seen in Fig. 7, showing tanks, tempera-
ture controllers, flowmeters, and solution
supply and turbulation pumps.

The machine described in this paper
has been operated very satisfactorily as
a production machine for over twelve
months. Considerable interest in it
has been shown in the industry. A
second unit, similar to the first, is now
being readied to increase the capacity of
color-coupling multilayer film processing
by the Cinecolor Corporation.

Acknowledgment: We wish to acknowl-
edge with thanks the helpful cooperation
of R. W. Lorenzen, O. W. Murray,
J. K. Stewart, E. W. Rutherford and the
many others who contributed to the
design and development of the Cine-
color Multilayer Color Developing Ma-
chine.

Discussion

John G. Stott: Did I understand that
you turbulate your bleach solution?
Could I ask why?

James W. Kaylor: It was recommended,
more or less, in the instructions for the
technique of developing the EK color
film that we use at the present time.

Mr. Stott: Well, the usual use to which
turbulation is put, is in a stage of chemical
processing, where the process doesn't
actually go to completion, but where you
want to make all the film go to the same
point at the same time. In other words,
you want all the film to be processed uni-
formly, but the fixing operations and the
bleaching operation are usually considered
as going to completion, so that turbulation
is, as I see, of absolutely no use.

Mr. Kaylor: Well, I can say that we have
tried it both ways, with turbulation and
without, and it was decided to use the
turbulation method. I believe that that
question probably could be answered a
little more fully in the paper that Mr.
E. W. Hart is going to read tomorrow
night, I believe, describing our three-color

58

January 1952 Journal of the SMPTE Vol. 58

Fig. 6. Dry box and take-ups.

Fig. 7. Section of basement showing storage tanks, temperature
controls, chemical pumps and flowmeters.

Kaylor and Pesek: Cinecolor Developing Machine

59

process, also going through the various
steps of the developing of our EK negative.

Mr. Stott: What is the total amount of
film in the developer section?

Mr. Kaylor: At the present time, it is
pretty close to 3000 ft. I say at the present
time because the machine can be set up
to take a total of about 3200 ft of film,
but in some of the tanks we do short
strand because we need much less than
three minutes' developing time, three
minutes', let's say, chemical time in those
particular tanks; in fact, in one or two of
the tanks we have only a couple of feet
of film.

Mr. Stott: Are your bottom rollers
floating rollers?

Mr. Kaylor: Yes, they're floating rollers.
The bottom rollers are all mounted on
elevators, with one exception where we

have them fixed to maintain a certain
amount of tension. I might say, perhaps,
that in this auxiliary equipment that was
mentioned — in the EK machine, for
example — it was found necessary to
install what we call a rem-jet remover
roller, to take off the jet-black backing
of the film. We had to install a velvet-
covered roller midway down into a tank
and use that to buff off the anti-halation
backing.

Edward H. Reichard: In the pictures of
the machine, I noticed that your top
rollers in the developing section are out
of the solution. Is that right?

Mr. Kaylor: No. Perhaps I should have
explained that in our developer section
we actually use submerged drives — the
nine shafts there are submergible beneath
the solution level.

60

January 1952 Journal of the SMPTE Vol. 58

New Magnetic-Recording Head

By MARVIN CAMRAS

A three-pole magnetic head produces a magnetic field at the recording gap
which is more uniform throughout the thickness of the magnetizable layer,
and decays more rapidly at the trailing edge. With this head, optimum bias
is practically the same for high as for low audiofrequencies. High audio-
frequencies are recorded at a 3-db to 7-db higher output level before dis-
tortion as compared with a similar head of conventional design.

wo IMPORTANT FACTORS have made
possible the present-day low magnetic-
recorder speeds: (1) thin, uniform
recording tapes with high magnetic
properties; and (2) efficient magnetic
heads with very short gaps.

Effect of Short Gaps

Efficiency and short gaps do not go
together, unfortunately, because, as
the gap is shortened, more of the useful
flux tends to be lost across the pole
faces. The situation for a playback
head is shown in Fig. 1. A recorded
signal such as A produces a certain
external flux which is utilized more or
less efficiently to thread a voice coil.
At the pickup gap some of the flux from
the record follows path B through the
core and through the voice coil. But
other lines of flux such as G and D prefer
to take the short path across the faces
of the pickup gap. Still others, such as

Presented on October 18, 1951, at the
Society's Convention at Hollywood, Calif.,
by Marvin Camras, Armour Research
Foundation, Illinois Institute of Tech-
nology, 35 West 33 St., Chicago 16, 111.

E, complete their circuit through the air
or backing material on the side opposite
the gap.

The flux divides inversely according
to the reluctance of each path. A gap
measured in tenths of a thousandth of
an inch may have less reluctance than
an inch or two of high-permeability core
material. Consequently, it may waste
a major portion of the playback flux.

When we use such a short gap for
recording as in Fig. 2, we find that
practically all of the flux concentrates
in the easy path across the short gap,
and a much lower field acts on the
record material, especially on the side
away from the gap.1 As we increase
input to the head to produce an adequate
recording field, we find that the head
core approaches saturation, and in,
many cases we cannot reach optimum
recording conditions regardless of how
much input we feed into the voice coil.

A third problem of magnetic heads,
that of maintaining good contact with
the record, becomes much more im-
portant as the gap is decreased. Studies
show that even with ordinary heads.

January 1952 Journal of the SMPTE Vol. 58

61

VOICE*
COIL

Fig. 1. Flux paths in a magnetic
head on playback.

2. Flux paths in a magnetic
head during recording.

/ I / Ss:^\\ \ \

A. Magnetic field produced by gap. B. Superposed cross field adds vectorially

to the gap field.

Fig. 3. Flux paths of an X-field head.

a separation of only 0.0001 in. causes
marked fall-off in high-frequency re-
sponse.8 With shorter gaps, a 0.0001-in.
spacing would be fatal. Thus, a de-
crease in record or playback gap, by
itself, does not solve the problem of
obtaining better resolution.

The Cross-Field Head

With a conventional head the mag-
netic field at the recording gap is as
shown in Fig. 3A. As an approxima-
tion, it has a semicircular direction with
respect to the gap center and decreases
inversely with the distance. Ordinarily
we have little control of the shape of
this field. Suppose we now provide
a vertical field as in Fig. 3B and combine
this field with the semicircular field of
the gap. (The additional field is termed
a cross field or X-field, and a head which

provides such a field is called an X-field
head.} Vector addition of the com-
ponents, as, for example, a, b, c, d, e
and f in Fig. 3B, gives some interesting
results:

1. On the left-hand side of the gap,
the components are additive, and the
resulting field is stronger and more
nearly vertical than the gap field alone.

2. On the right-hand side of the gap,
the vertical components oppose each
other, and can be adjusted to cancel
at some point near the right-hand edge
of the gap. For example, in Fig. 3B
near the surface of the right-hand pole,
at x, the gap field predominates; while
at z the cross field is stronger. Some-
where in between, near y, cancellation
occurs, leaving a minimum field. This
means that we have a steep gradient of
field in the region of x.

62

January 1952 Journal of the SMPTE Vol. 58

The resultant field is mapped in
Fig. 4. A head which produces such
a field has certain advantages over
conventional heads for magnetic re-
cording:

1. The field falls off more sharply as
the record leaves the gap. This im-
proves the resolution and minimizes
"recording demagnetization."3

2. In a direction away from the pole
pieces, the field changes less rapidly,
giving more uniform magnetization
through the thickness of the record, and
less variation due to poor contact be-
tween head and record.

3. The field at the right-hand pole
edge is more nearly longitudinal.

4. The main recording head operates
at lower flux density.

5. The shape of the resultant field
can be controlled by varying its com-
ponents.

Figure 5 is a photograph of an early
X-field head with taps to allow adjusting
the relative number of turns on each
of the legs. The auxiliary pole piece
overhangs the head proper and is spaced
about 15 mils above it, which gives
clearance for threading and for splices
in a 2-mil tape. Figure 6 shows typical
connections for this head.

Test Results

To measure the improvement result-
ing from the cross field, the head was
first tried with the center coil discon-
nected, so that it functioned as an
ordinary head. Output vs. input curves
were then run at 10,000 cycles, using the
bias at which maximum possible un-
distorted output level occurred. Then

V////V/////I

##^<xy / / /

Fig. 4. Resultant of gap field

and X-field.

the cross-field coil was connected, and
adjustment was again made for maxi-
mum undistorted 10,000-cycle output.
The results are given in Fig. 7, and
indicate a 4j-db advantage for the X-
field connection, which means that
about 65% more flux can be recorded
at the high frequency on the same tape.

It is well known that when supersonic
bias is increased beyond a certain rather
critical value, the high-frequency re-
sponse of a recording system goes down
rapidly. The best explanation for this
effect is that partial erasing of the short
wavelengths occurs in the extended field
of the recording gap (see "recording
demagnetization" 3) . On the other hand,
for distortionless low-frequency record-
ing, we need enough bias to excite the
recording layer through its entire thick-
ness, and this turns out to be con-
siderably more than for optimum high-
frequency response. The result is
usually a compromise in which the high
frequencies suffer.

Bias requirements of the new head
were determined both with the con-
ventional and with the X-field connec-
tions. Results are shown in Table I.

Table I.

Connection

Bias required for undistorted
100-cycle output

Bias for max.
10-kc output

Loss at 1 0 kc due
to increased bias

Standard
X-field
X-field

1000 ma
650 ma
(Purposely overbiased
to 1000 ma)

600 ma
560 ma

~db
Odb
1.5db

Marvin Camras: Magnetic-Recording Head

63

i i f

Fig. 5. Experimental X-field head.

\ OUTPUT

OUTPUT 1

HIGH FREQ.
BIAS
OSCILLATOR

AMPLIFIER
AND
EQUALIZER

Fig. 6. Typical connections
for X-field head.

30

20

10

BIAS FREQ =40KC

rX- FIELD
CONNECTION

J.

/^

'

* WITHOUT X-FIELD

£_

/&

^-DISTORTION POINT

/t

'/

,

fi

'

10 100

INPUT CURRENT- MA

1000

Fig. 7. Output-input curves for
head at 10 kc.

64

January 1952 Journal of the SMPTE Vol. 58

Fig. 8. Modified X-field head designs: (A) Effect of moving X-field pole piece to
the left; (B) Double gap X-field head; and (C) Circuit for producing a rotating

field at recording gap.

We note that bias requirements for
the X-field head are practically the
same for the high and low frequencies,
and overbiasing has little effect. If the
tests of Fig. 7 are conducted on a prac-
tical basis of optimum bias for low fre-
quencies in each case, rather than bias
for maximum 10-kc output, the ad-
vantage of the X-field head is even
greater.

Variations

Some variations in the X-field head
are shown in Fig. 8. In Fig. 8A the
overhanging pole piece has been moved
to the left. This tilts the cross field
from the vertical direction and causes
better cancellation of the gap field at
the trailing edge. Also, the cross field
decreases in intensity as we move to the
right, so that it has less effect on the
record beyond the cancellation point.
The intense vertical field at the left of
the record gap can be used for erasing,
and its intensity may be increased still
further by sharpening the pole piece.
Or if we run the tape backward and
make appropriate adjustments, this head
becomes an efficient vertical recording
head.

Figure 8B shows a head that has
advantages of the previous one, but dis-
penses with the overhanging pole piece.
Two gaps on the same side of the re-

cording head are spaced closely enough
together so that the right-hand part of
the field produced by the large gap
acts as a cross field for the small recording
gap. The large gap may be used for
erasing. In this connection, we found
that by using audio in the main field
only, but retaining the X-field principle
for bias, excellent performance was
obtained with this very simple design of
erase-record head.

Figure 8C is a circuit which shows the
degree of control possible with the X-
field head. Here we use a condenser
of relatively high reactance in circuit
with the X-field coil to give a current
90° ahead of the voltage. A resistor
in the recording coil circuit gives a
current in phase with the voltage. The
result is a head which provides a rotating
bias field. Similarly, we can produce a
rotary signal field or even a rotary eras-
ing field.

We have found that although the
primary advantages of the X-field head
are in recording, there are also ad-
vantages in playback, since a reciprocal
magnetic relation holds.

Conclusions

The field of a recording head gap can
be modified to advantage by combining
it with a cross field.

An X-field head has better resolution,

Marvin Camras: Magnetic-Recording Head

65

and a more ideal bias adjustment than
an ordinary head.

Modified designs of X-field heads are
as simple to manufacture and to use as
conventional heads.

References

1. S. J. Begun, "Magnetic field distribu-
tion of a ring recording head," Audio
Eng., 32: 11-13, 39, Dec. 1948.

2. R. Herr, B. F. Murphey and W. W.
Wetzel, "Some distinctive properties of
magnetic-recording media, "Jour. SMPE,
52: 77-87, Jan. 1949.

3. O. W. Muckenhirn, "Recording de-
magnetization in magnetic tape re-
cording," Proc. IRE, 39: 891-897, Aug.
1951.

Discussion

Anon: Can you tell us what the channel
width was about?

Marvin Camras: This was done with an
eighth-inch wide channel on quarter-inch
tape.

Anon: Would the same results be ob-
tained with 35mm magnetic film?

Mr. Camras: I don't see why the medium
would affect the result. I think it would
be substantially the same.

Anon: Do you still find the same varia-
tions in output of low frequency with
this new type of head as you do with the
normal-type head; the change of ampli-
tude with frequency at about 100 cycles?

Mr. Camras: You mean the bumps in
the response curves? As I remember, I
don't think it had very much effect on
that. That's caused by something else.

Anon: I know, but the changing of the
contour of the field might affect that. As
a matter of fact some of those bumps are
likely playback effect, and if you use con-
ventional playback heads it wouldn't
change.

M. G. Townsley: When you use a cross-
field head for record, you improve resolu-
tion on the film and the improved results

are on the film. You said something very
briefly about using a cross-field head for
playback. What type of a cross field do
you put on the head when you are using
it as a playback head?

Mr. Camras: Of course, in a playback
head you are just picking up fields, using
it as a sensitive element for picking up
something that's recorded. By using it
for playback I meant that we used the
cross-field coil and left it connected in
the circuit during playback. We haven't
run exhaustive tests and I haven't shown
any results here of what it can do, but it
seems to give us advantages in playback.

Mr. Townsley: In other words you make
a fairly large gain. Suppose you use a
head like this as a combined record play-
back head, you'd make a fairly large gain
in record and a small further gain in play-
back?

Mr. Camras: Yes, I would say that.
You're utilizing the material throughout
the thickness of the layer to better ad-
vantage than you are with the conventional
head that operates on one side of the record.

Mr. Townsley: The cross field so to speak
is induced by the tape material itself as
it passes over the head if you've got a
double gap, for example, as you showed in
the last picture.

Mr. Camras: I don't think I understand.

Mr. Townsley: I don't either.

Mr. Camras: Those gaps incidentally
are very close together, just a few mils
apart; and in that case you might get
additional reinforcement or pickup of low
frequencies with your second gap. If
you try to work such a scheme with con-
ventional double-gap heads where the
record gap may be spaced a hundred mils
or more from the erase gap you'd get
bad echo effects, of course.

Mr. Townsley: It would seem to me that
you might, with the gap space closer to-
gether, get some interference cancellation
effects at fairly high frequencies, because
of the phasing.

Mr. Camras: I haven't noticed those
things. One gap is considerably larger
than the other, at least ten times as large.

66

January 1952 Journal of the SMPTE Vol. 58

Push-Pull Direct-Positive Recording —

An Auxiliary to Magnetic Recording

I

By LESLIE I. CAREY and FRANK MORAN

This paper explains the transferring of magnetic film from the daily okayed
production takes to push-pull direct-positive film. By using a protective
coating on the sound track, the cutting-room hazards are reduced 90%.
This coating can be "peeled off" just before dubbing, assuring a new clean
track from which to dub.

JL HE ADVANTAGES of using direct-
positive records as a part of a magnetic
recording program have been previously
described by Loren L. Ryder in the
JOURNAL.1 The records mentioned by
Mr. Ryder are of the variable-density
direct-positive type, utilizing the super-
sonic bias previously described.2 The
purpose of this paper is to describe the
use of variable-area double-width push-
pull records as an adjunct to the mag-
netic-recording program now in use at
Universal Studios.

The difficulties involved in editing
magnetic film and the expense involved
in cutting it for dubbing purposes sug-
gested the need of a medium which is
inexpensive and, at the same time,
capable of giving quality comparable
to that obtained from magnetic films.
In the opinion of the authors, the 200-
mil variable-area double-width push-pull
track adequately fulfills these require-
ments. The double-width track was
selected because of its advantage over

Presented on May 4, 1951, at the Society's
Convention in New York, by Leslie I.
Carey and Frank Moran, Universal-
International Pictures, Universal City,
Calif.

a single track, from a signal-to-noise
standpoint, and the push-pull feature,
as described below, was selected because
of the lack of critical processing problems
and improved signal-to-noise ratio.
Direct-positive was selected because it
eliminates the need for a negative record
with its accompanying processing ex-
pense and printing losses.

The underlying principles involved in
making a direct-positive record with the
light valve have been described in the
JOURNAL.' The double-width push-pull
variable-area track is obtained by
applying a signal to the center ribbon
and noise reduction to the two outside
ribbons of a three-ribbon variable-area
light valve.4

In order to obtain the type of track
under discussion, a standard Western
Electric R A- 1231 Type Recorder,5
equipped to record either a single- or
double-width variable-area negative
track, was modified to incorporate the
direct-positive feature. No major
changes were made in the film-pulling
mechanism of this recorder, with the
exception that facilities were provided
so that the recorder could be run either
in the normal forward or in a reverse

January 1952 Journal of the SMPTE Vol. 58

67

DIRECT POSITIVE
POSITION

Fig. 1. Optical schematic.

OBJECTIVE
LENS L2

RE.C

1.0 1.2 1.4 18

VISUAL DIRECT -POSITIVE DENSITY

Fig. 2. Cross modulation characteristic.

direction. This permitted the recording
of the normal negative track in the
regular forward position, and the direct-
positive track by the simple process of
reversing the direction of the film during
recording to eliminate the necessity of
changing the position of the light valve.
The studio has found it convenient to
use a synchronous motor with the re-
corder on all occasions except for
dubbing when, of course, the interlock
motor is essential. In order to make
both these drive facilities readily avail-
able, the recorder was equipped with
two drive motors — one synchronous and
one interlock — coupled through a single
shaft to the drive mechanism. Either
motor could then be selected by a
simple switching arrangement.

The changes incorporated in the

optical system to obtain the direct-
positive type of track are noted below.

The Western Electric RA-1247 Three-
Ribbon Light Valve4 was restrung with
reflecting surface ribbons. The optical
system of the modulator was modified,
as shown schematically in Fig. 1 , so that
the direct-positive type of record is
obtained with the light source in the
upper position. In this position the
light passes through the condensing
lens, LI, to a mirror, Ml, where it is
deflected downward to the slit mirror,
M2. The latter directs the light through
the objective lens, L2, in the light valve
onto the reflecting ribbons. The light
reflected by the ribbons is returned
through the slit in the slit mirror, thence
through a clear-glass plate, PI, and
through the objective lens to the film.
The clear-glass plate reflects a small per-
centage of the recording beam back to
the photocell for PEG monitoring.

The double-width push-pull track
offers two distinct advantages over the
standard track, first, by virtue of the
extra width, an improvement in the
signal-to-noise ratio and, second, the
push-pull feature offers low and rela-

68

January 1952 Journal of the SMPTE Vol.58

Fig. 3. Direct-positive recorder. Left: rear view. Right: front view.

Fig. 4. Magnetic channel. Left: front view. Right: rear view.

lively constant cross-modulation products
over a large range of densities (Fig. 2).
The latter feature greatly simplifies the
processing problems which are normally
encountered on tracks which do not
have the push-pull feature. For
example, a track density of 1.8 on
Eastman Kodak Emulsion 5372 was
found to be completely feasible. When
this is compared to a "balance" density
around 1.3 for a single track on the same

emulsion, the advantages of increased
signal output and reduced grain noise
are immediately evident.

The 200-mil push-pull variable-area
method of recording was put into service
at Universal Studios on January 10, 1951,
on the production The Iron Man, and
subsequently on a musical short, Qggy
Elman and His Orchestra. It is currently
being used on the production Fiddlers
Green. All original takes are being made

Carey and Moran: Direct-Positive Recording

69

on 35mm sprocket-hole-type magnetic
film and at the end of a day's work all
"print" takes are transferred to direct-
positive. In this way the magnetic
film becomes the "negative" and is stored
intact until the picture is released, at
which time it will be erased and avail-
able for future productions. The direct-
positive track now serves as the "daily"
sound track and eventually, after editing
and cutting, as working and dubbing
prints.

The quality of sound obtained through
the magnetic-direct-positive procedure is
noticeably superior to and more de-
pendable than that obtained with the
regular photographic process. A careful
check has indicated that very little noise
is introduced as a result of running for the
"dailies" and during handling in the
cutting process.

Not only does the use of magnetic-
direct-positive system improve the over-
all sound quality of a production, but
economies in the cost of film and film
processing are realized at the same time.
While exact figures are not yet available,
approximate relative costs between the
regular photographic process and the
magnetic-direct-positive process have
been estimated on The Iron Man pro-
duction. A total of 104,000 ft of re-
cording was made on this production.
Actually 52,000 ft of magnetic film were
used with separate tracks at either edge
of the 35mm film. One half, or 29,000
ft, of this recording represented "print"
takes which were transferred to direct-
positive. If this production had been
originally recorded on a photographic
negative, from which working prints
were to be made, a total of 104,000 ft
of negative stock, plus 29,000 ft of
positive stock, plus the negative and
positive processing cost would have
involved an expense of approximately
$4100. As the production was actually
recorded first on magnetic film and then
with the "print" takes transferred to
direct-positive, 29,000 ft of direct-
positive stock plus its processing involved

a cost of approximately $884. The
initial investment in 52,000 ft of mag-
netic film is approximately $2080.
However, inasmuch as this film can again
be used after erasure, only a portion of
the cost should be applied against this
production.

Sufficient time has not yet elapsed to
make a sound determination of the life
of the magnetic film, inasmuch as some
of the stock obtained at the beginning
of our magnetic program three years
ago, and used frequently since that time,
is still in good condition. If it is as-
sumed conservatively that the stock can
be used 25 times, the cost per production
is then approximately $84. The total
cost of the film together with the proc-
essing of the direct-positive is, therefore,
$884 plus a prorated cost of $84 for the
magnetic film, or a total of $968, as
against $4100 for the negative-positive
method, or a saving of approximately
$3132 per production.

The use of the push-pull variable-area
direct-positive recording as an adjunct
to the magnetic recording program at
Universal Studios has proven to be
completely successful from the stand-
point of simplifying and cutting the cost
of the recording operations, as well
as retaining the high quality of the
original magnetic recordings.

References

1. Loren L. Ryder, "Motion picture studio
use of magnetic recording," Jour.
SMPTE, 55: 605-612, Dec. 1950.

2. G. R. Keith and V. Pagliarulo, "Direct-
positive variable-density recording with
the light valve," Jour. SMPE, 52: 690-
698, June 1949.

3. Lewis B. Browder, "Direct-positive
variable-area recording with the light
valve," Jour. SMPE, 53: 149-158, Aug.
1949.

4. John G. Frayne, "Variable-area re-
cording with the light valve," Jour.
SMPE, 51: 501-520, Nov. 1948.

5. G. R. Crane and H. A. Manley, "A
simplified all-purpose film recording
machine," Jour. SMPE, 46: 465-474,
June 1946.

70

January 1952 Journal of the SMPTE Vol. 58

Proposed Standard
Enlargement Ratio for 16Mm
to 35Mm Optical Printing

EFFORTS TO reduce costs in color
cinematography have led, in the
past few years, to an appreciably
increased commercial use of 16mm
film as original negative for 35mm
release prints. Optical enlargement
printing is, of course, an essential
factor in this process. A standard
magnification ratio thus becomes a
necessity since the difference in as-
pect ratios of the two film sizes pre-
cludes the simple use of the 35/16
ratio.

The Laboratory Practice Com-
mittee, chaired by John Stott, tackled

the problem in February 1951; a
first draft was submitted by Gordon
Chambers in May 1951 and ap-
proved by the Committee a few
months later. A revised draft was
subsequently approved for publica-
tion by the Standards Committee
and is published on the following
page for a 90-day period of trial and
criticism.

Please forward any comments, in
duplicate, to Henry Kogel, Staff
Engineer, at Society headquarters,
by April 15, 1952.

January 1952 Journal of the SMPTE Vol. 58

71

Proposed American Standard

Enlargement Ratio for 16Mm
to 35Mm Optical Printing

PH22.92

In the enlargement printing of 16mm film
to 35mm film, a magnification of 2.21 ± 0.01
shall be employed and the center of the 16mm
frame as enlarged shall coincide with the
center of the 35mm aperture in the enlarging
printer.

This will mean a scanned area on the 16mm
frame of 0.272 inch =b 0.002 X 0.373 inch d=
0.002 will be projected through the 35mm
projector aperture when the print is used in
the theater. This corresponds to a frame of

0.284 inch X 0.380 inch if the 16mm original
were projected directly.

The scanned area of the 16mm frame in the
printer as enlarged to the 35mm camera aper-
ture is 0.286 inch ± 0.002 X 0.393 inch =t
0.002.

Attention of camera users is invited to the
desirability of using a camera finder matte
0.272 inch ± 0.002 X 0.373 inch ± 0.002
when exposing 16mm film to be enlarged to
35mm film.

Note: In enlargement from 16mm positive or reversal
original to 35mm negative a black frame line will
result on the final 35mm print. In the case of enlarge-
ment from 16mm negative directly to 35mm print,
white frame lines will result. If the height of the 16mm
aperture for enlargement from 16mm negative to

35mm print is made 0.300 inch, the resulting aperture
image on the 35mm print will be from 0.660 to 0.666
inch in height. While the frame line will not be en-
tirely black, there would be a black margin on either
side of the image which would give an additional
safety factor in projection.

NOT APPROVED

72

January 1952 Journal of the SMPTE Vol. 58

71st Semiannual Convention

The Spring Convention has for many
weeks been in the minds and work of those
generally responsible for conventions and
of those especially responsible for the
Chicago Convention, April 21-25, at The
Drake.

Bill Kunzmann has already spent a good
deal of time in Chicago and has done all
the groundwork of planning with The
Drake and also already has a roster of
chairmen for the dozen major activities
and functions throughout the convention.
The complete roster will be published in
the February Journal.

John Frayne, at an editorial meeting
during the Hollywood Convention, ap-
pointed as Program Cochairmen R. T.
Van Niman and George Colburn. They
and others are already at work on the
papers program under direction of Papers
Committee Chairman Ed Seeley. Manu-
scripts and suggestions should go promptly
to any of the Papers Committee, listed
below; but all manuscripts and authors'

forms (these are available from members
of the Committee or from Society head-
quarters) should reach Cochairman George
Colburn, 164 N. Wacker Drive, Chicago 6,
111., as soon as possible.

John Frayne also confirmed at the
Hollywood editorial meeting the choice of
Richard O. Painter to be Vice-Chairman
for High-Speed Photography for Chicago.
In addition to the planning of at least
two high-speed photography sessions, the
editorial meeting and subsequent planning
have evolved the following tentative
schedule of session subjects: two sessions
on 16mm; one on sound recording;
three or four sessions on television; one
on laboratory problems; and one general
session.

C. E. Heppberger, Secretary-Treasurer
of the Central Section, has the very im-
portant duties of Local Arrangements
Chairman. He put out a solid two-page
memo in late November to begin tying
together the long roster of all the arrange-
ments he must be sure about.

PAPERS COMMITTEE

Chairman: Edward S. Seeley, Altec Service, 161 Sixth Ave., New York 13

71st Convention Program Cochairmen: R. T. Van Niman and George W. Colburn. Address

manuscripts and authors'' forms to George Colburn, 164 N. Wacker Drive, Chicago 6, III.
Vice-Chairmen

For New Tork: W. H. Rivers, Eastman Kodak Co., 342 Madison Ave., New York 17
For Washington: J. E. Aiken, 116 N. Galveston St., Arlington, Va.
For Los Angeles: F. G. Albin, Station KECA-TV, American Broadcasting Company

Television Center, Hollywood 27, Calif.

For Canada: G. G. Graham, National Film Board of Canada, John St., Ottawa, Canada
For High-Speed Photography for Chicago: Richard O. Painter, General Motors, Proving

Ground Section, Milford, Mich.

Committee Members

A. C. Blaney, RCA Victor Div., 1560 N.

Vine St., Hollywood 28, Calif.
Richard Blount, General Electric Co.,

Nela Park, Cleveland, Ohio
R. P. Burns, Balaban & Katz, Great

States Theaters, 177 N. State St.,

Chicago 1, 111.
Philip Caldwell, American Broadcasting

Co., 6285 Sunset Blvd., Hollywood,

Calif.
F. O. Calvin, The Calvin Co., 1105 E.

Fifteenth St., Kansas City 6, Mo.

Howard Chinn, Columbia Broadcasting

System, 485 Madison Ave., New York
J. P. Corcoran, Twentieth Century-Fox

Film Corp., 10201 W. Pico Blvd.,

Beverly Hills, Calif.
G. R. Crane, Westrex Corp., 6601 Ro-

maine St., Hollywood 38, Calif.
E. W. D'Arcy, De Vry Corp., 1111 W.

Armitage Ave., Chicago 14, 111.
Farciot Edouart, Paramount Pictures

Corp., 5451 Marathon St., Hollywood

38, Calif.

73

F. L. Eich, Paramount Film Laboratory,
1546 Argyle Ave., Hollywood 28, Calif.

Dudley Goodale, National Broadcasting
Co., 30 Rockefeller Plaza, New York 20.

Charles Handley, National Carbon Div.,
841 E. Fourth PI., Los Angeles 13, Calif.

R. N. Harmon, Westinghouse Radio Sta-
tions, Inc., 1625 K St., N.W., Washing-
ton, D.C.

Scott Helt, Allen B. Du Mont Labs., Inc.,
2 Main Ave., Passaic, N.J.

C. E. Heppberger, National Carbon Div.,
230 N. Michigan Ave., Chicago 1, 111.

J. K. Hilliard, Altec Lansing Corp., 1161
N. Vine St., Hollywood 38, Calif.

L. Hughes, Hughes Sound Films, 21 S.
Downing St., Denver, Colo.

P. A. Jacobson, University of Washington,
Seattle, Wash.

William Kelley, Motion Picture Research
Council, 1421 N. Western Ave., Holly-
wood 27, Calif.

E. P. Kennedy, Signal Corps Labs, Fort
Monmouth, N.J.

George Lewin, Signal Corps Photographic
Center 35-11 35 St., Long Island City
1, N.Y.

E. C. Manderfeld, Mitchell Camera Corp.,
666 W. Harvard St., Glendale 4, Calif.

Glenn Matthews, Research Laboratory,
Eastman Kodak Co., Rochester 10, N.Y.

Pierre Mertz, Bell Telephone Labs., Inc.,
463 West St., New York 14

James Middlebrooks, American Broad-
casting Co., 30 Rockefeller Plaza, New
York 20

Harry Milholland, Allen B. Du Mont
Labs, Inc., 515 Madison Ave., New
York 22

W. J. Morlock, General Electric Co.,
Electronics Park, Syracuse, N.Y.

Herbert Pangborn, Columbia Broadcast-
ing System, Inc., 6121 Sunset Blvd.,
Hollywood 28, Calif.

Edward Schmidt, Reeves Soundcraft, 10
E. 52 St., New York 22

N. L. Simmons, Eastman Kodak Co.,
6706 Santa Monica Blvd., Hollywood
38, Calif.

S. P. Solow, Consolidated Film Industries,
Inc., 959 Seward St., Hollywood 38,
Calif.

J. G. Stott, Du-Art Film Laboratories,
245 W. 55 St., New York 19

W. L. Tesch, Radio Corporation of
America, RCA Victor Div., Front and
Cooper Sts., Camden, N.J.

S. R. Todd, Consulting Electrical Engi-
neer, 4711 Woodlawn Ave., Chicago, 111.

M. G. Townsley, Bell & Howell, 7100
McCormick Rd., Chicago 45, 111.

Discussions in the Journal

Discussions are a valuable part of the
Society's functioning. Those which occur
on the floor at Conventions are now re-
corded as described in Ed Templin's
Committee Report in the December
Journal. The procedure and policy, once
discussion is on a disk, are:

Headquarters staff transcribes it almost
verbatim, pausing to correct only the most
obvious verbal slips. The typewritten
transcript is sent to the author, usually at
the time his paper is being processed for
Journal publication. Depending on the
length and clarity of the discussion, the
transcript is sent simultaneously or suc-
cessively to all discussers. Whatever the
timing, however, discussion is sent to all
persons named in the record and they
must clear it before it is published.
What worthy discussion cannot be identi-

fied as to source becomes that of Mr.
Anon.

Within a month after the close of the
Hollywood Convention, the Society's staff
had transcribed 105 pages of discussion
from that program.

In addition, 48 pp. have been transcribed
and mimeographed as the record of the
Panel Discussion on Emulsion Position of
16 Mm Positives. This has been sent
to all known interested persons. Let head-
quarters know if you are interested and
were overlooked. A copy will be sent to
you. When everyone interested has re-
turned his panel or subsequent discussions
to Society headquarters, a composite copy
will be made for review by Norwood
Simmons, who was moderator of the panel
discussion, and it will then be assessed for
Journal publication.

74

Engineering Activities

Three meetings of interest were held
recently, only one of a Society engineering
committee. The highlights of these meet-
ings are outlined below.

PH22 Led by its new Chairman, D. R.
White, ASA Sectional Committee
on Standards for Motion Pictures, PH22,
met November 29, 1951, and had a very
fruitful session with an agenda limited to
three key items.

Letter Ballots: Three letter ballots were
considered and acted upon:

1. Two proposed standards for 35mm
multifrequency test films, PH22.63 and
PH22.64, held in abeyance for some time
as a result of a major consumer's negative
vote, were returned to the Sponsor for
resolution of the existing differences.

2. A proposed revision of the standard
for 16mm reels, PH22.11, was approved
and forwarded to the Sponsor.

3. The ballot on the 16Mm Edge Num-
bering Proposal, PH22.83, was incomplete
and the Chairman was authorized to close
the ballot at his own discretion.

ISO: Questions relating to a con-
templated meeting of ISO TC/36 (Inter-
national Standards Organization Technical
Committee on Cinematography) in New
York in June 1952 were thoroughly re-
viewed. It was decided to canvass the
participating members concerning their
interest in attending such a meeting,
informing them of our willingness to call
one if there is promised attendance from
abroad. (The ASA, Secretariat of TC/36,
subsequently sent a modified version of a
letter drafted by PH22.)

PH22 Scope: The new scope, endorsed
at the last meeting, was criticized in the
interim as excessively broad and two
alternate proposals were offered for Com-
mittee consideration. A compromise be-
tween the two was approved as the Com-
mittee recommendation to the SMPTE,
which as Sponsor, has the final word on
the scope to be submitted to the ASA.

IRS The IRS was formed in April 1950

as a coordinating committee of

three Societies (IRE, RTMA, SMPTE) to

eliminate or reduce duplication of work
and areas of conflict in mutual spheres of
activity, primarily in the field of television.
Originally chaired by Axel Jensen and now
by Fred Bowditch, the Committee met
November 30 and December 20, 1951,
to consider two main points.

Committee Addition: In the light of
NARTB's renewed interest in standards
activity, discussions were held as to the
advisability of including it as a fourth
member. After an affirmative vote at the
first meeting, the NARTB was officially
welcomed as a Committee member at the
December meeting.

Recording Standards: CCIR's (Inter-
national Radio Consultative Committee)
program for standardizing radio program
recordings for use between nations was
outlined and the need for American
Standards on sound recording was re-
viewed. The Committee concluded that
the ASA Sectional Committee on Sound
Recording, Z57, should be reactivated and
proposed the procedure for achieving this.

New Name: The addition of a fourth
member required a change in the IRS
Committee name which was compounded
from the first initials of the three par-
ticipating Societies. "Joint Committee
for Inter-Society Coordination," to be
abbreviated "JCIC," won the day and is
the new name of the Committee.

Television Studio Since its inception in

Lighting January 1950 under

the chairmanship of

Richard Blount, this Committee has met
about every three months. The Chairman
noted, however, that very little has been
accomplished this past year. The main
discussion then centered on the cause of
this situation and how to remedy it.

This very practical approach resulted in
changes both in form and content of the
Committee's work with accompanying
changes in project responsibilities. Small
subcommittees were eliminated and the
entire Committee is to concentrate its
attention on two main projects: lighting
measurements and terminology. — Henry
Kogel, Staff Engineer.

75

Book Reviews

Three-Dimensional Photography:
The Principles of Stereoscopy

By Herbert C. McKay. Published (1951)
by American Photographic Publishing Co.,
421 Fifth Ave. So., Minneapolis 15, Minn.
334 pp. 98 illus. 6 X 9 in. Price $5.75.

Herbert C. McKay, FRPS, ASC, well
known to readers of American Photography
for his monthly column "Notes from the
Laboratory" and for his observations on
developments in photography and com-
ments on stereoscopy, has compiled a
text that is of interest to amateur photog-
raphers but it's hardly a book that has
much appeal to professional photographers
or serious stereographers. Some of the
theories on which the principles of stereos-
copy are based are blithely ignored, some
are attacked. It certainly is not to be
recommended as a reference work for any
motion picture engineer interested in the
stereoscopic process.

The author preaches such adroit doc-
trines as: "It has been repeatedly demon-
strated that a beginner knowing nothing
whatsoever about photography will have
a greater success in stereo than in con-
ventional photography"; and "... the
fact remains that the gravest trouble
encountered by projectionists in the stereo
field is the result of taking too much care."

The inference, to me at any rate, is that
knowledge of stereoscopic theory, skill in
photography, and careful craftsmanship
are handicaps rather than helps in the
stereoscopic art.

To sustain the mood, the author, in
referring to the projection of stereo slides
has this to say, "... You drop the stereo-
gram in the projector and enjoy it. The
headaches have all been removed. There
is nothing more than this that is absolutely
essential." Then, in taking stereograms
of close objects, "Some stereographers
erroneously use a narrow base when
making any stereogram nearer than ten
feet."

He evidently means that if you're photo-
graphing a flower at a distance of 2\ ft
with the normal base (lens interaxial) of
2$ in. and there is not provision on the

camera for converging the field of each
lens to a plane 2% ft away or nearer you'll
come out with a perfectly good stereogram.
This conflicts with some of the basic theories
of stereoscopy.

To quote the author again : "Those who
have seen modern stereo projection, now
predict that stereo movies will soon be
developed; they do not know that stereo
movies were presented in a Broadway
theatre a quarter century ago, and in
many other theatres throughout the land.
They do not know that polarized light
stereo movies were featured at both the
Chicago (1933) and New York (1939)
World's Fairs. There is little to be done
in that field, it has all been done time after
time and any amateur can with a minimum
of ingenuity make his own stereo attach-
ments which will enable him to project
perfect stereo movies." It will interest all
to know that "There is little to be done in
that field, it has all been done time after
time . . . . " And, that anyone with a
minimum of ingenuity can make and
project perfect stereo movies. I'm afraid
it takes just a bit more doing than Mr.
McKay seems to indicate.

But let's have some more light on the
subject from the author: "... we have
not emphasized the distinction between
motion pictures and still projection, for
one very good reason. Optical projection
remains the same no matter whether the
projected images are changed twenty
times a second or twenty times an hour.
A system which will work with one, will,
with few exceptions work with the other."
This reviewer and his associates have
been concentrating through the years on
these "few exceptions," to the exclusion
of the seemingly more direct and simpler
methods. All serious workers in cine-
stereoscopy must take into consideration
the problems of uneven illumination,
differential vibration between members of
the stereoscopic pair and other things
that can detract from complete visual
comfort for the audience viewing three-
dimensional motion pictures. — J. A. Nor-
ling, Loucks and Norling Studios, 245 W.
55th St., New York 19.

76

The Indian Film

By Panna Shah. Published by I. K.
Menon and the Motion Picture Society of
India, Sandhurst Bldg., Sandhurst Road,
Bombay 4, India. 289 pp. incl. 22 pp. of
appendix, bibliography and index. 20
illus. 5 1 X 8i in. Price Rs. 10/-.

Dr. Panna Shah has put film makers
both of the East and the West very much
in her debt by this searching study of the
conditions of the motion picture industry
in her native country. Thoroughly versed
in the film literature of the western world,
Dr. Shah has a useful yardstick for measur-
ing Indian accomplishments. The condi-
tions she reveals are indeed depressing.
In chapter after chapter she castigates
Indian producers, distributors and ex-
hibitors alike for the poor quality of Indian
films and the wretched conditions under
which they are shown. Yet her criticisms
are not merely destructive. It is evident
that they are inspired by a strong and
sincere wish to see indigenous Indian
films of high quality achieve success in
India itself and spread a greater knowledge
of India to the rest of the world.

Though vital statistics of the Indian
industry are seemingly scanty and in-
accurate, Dr. Shah collates them to the
best possible effect to show a state of
affairs resembling that of the U.S. industry
some thirty years ago, when bankruptcies,
ever-changing amalgamations and sudden
standstills of production were prevalent.
Nor are these conditions surprising in a
country where so high a proportion of the
population lives in the villages, which are
seldom or never reached by films. And
there are the further limitations of multi-
plicity of languages and tremendous
differences of taste and cultural back-
ground.

The history of the Indian film is thor-
oughly covered, and there are chapters on
Indian film stars, on audiences, on censor-
ship, on mythology, and on the social
influence of films, which is evidently the
author's particular field of study. This
is a book which all should read who wish
to learn more about the second largest
film industry in the world. — Raymond
Spottiswoode, Kingsgate, Sudbury Hill,
Harrow-on-the-Hill, Middlesex, England.

The Film Industry

in Six European Countries

By Film Centre, London. Published
(1950) by Unesco, Paris; U.S. sales agent,
Columbia University Press, 2960 Broad-
way, New York 27. 156 pp. Many
tables. 5 f X 8|. Paper covered. Price
$0.65.

This is one of the series "Press, Film
and Radio in the World Today" which
Unesco is publishing in following out its
constitutional obligation to "further by
all possible means the use of the instruments
of mass communications in the work of
advancing the mutual knowledge and
understandings of peoples."

Beginning on the strong basis of a Danish
report "Betaenkning . . . angaende Bio-
grafvaesenet" published in 1950, a de-
tailed study and comparison are developed
for the other two small countries, Norway
and Sweden, then chiefly a statistical
study is presented for Italy, France and the
United Kingdom. Making Denmark the
special part of this study is logical enough
when the facts are in on the Danish film
industry : for instance, Denmark a country
of only about 4,000,000 persons produces
more films a year than Belgium, The Nether-
lands and Switzerland together. This
small book has an amazingly large amount
of text and statistics about costs and results
in exhibition, distribution and produc-
tion.—V.A.

Charlie Chaplin

By Theodore Huff. Published (1951) by
Henry Schuman, 20 E. 70th St., New
York 21. i-xi + 354 pp. + 80 pp. illus.
6 X 9 in. Price $4.50.

The filmic Charlie Chaplin is here
given perhaps as well as he can now be
portrayed in a book, unless a book were
to contain even more than this volume's
generous collection of 80 pages of illustra-
tions. But of looking at stills there is soon
an end, and we go back whenever possible,
generation after generation the world
over, to seeing Chaplin films. And how
seldom we hear them referred to nowadays
as "old" films.

For the many who would like to find
out how old is each Chaplin film, this is
an excellent reference. One appendix

77

gives biographical sketches of the people
professionally associated with Chaplin;
another appendix indexes thoroughly all
the films: the Keystones in 1914, the
Essanay Films of 1915-16, Mutual Films
in 1916-17, the First National releases
of 1918-22, and the seven released by
United Artists in 1925-1947. Casts, re-
lease dates, length of films and other data
are given.

There is considerable text which will
varyingly inform or interest readers. Not
only is the production of each film de-
scribed but also there is given a frame of
timely reference of general and film
business conditions, international and
domestic political factors, and, without
being unnecessarily scandalous about it,
an adequate notice of what was happening
in the personal lives of those on or off the
sets. If this is not a thoroughly knit and
compact picture of the individual Chaplin,
perhaps we can forgive the biographer at
this time when it is doubtful if such could
be accomplished even autobiographically.
On one point, however, the author is
clear: the artist Chaplin has ever been
striving wholly and honestly to accomplish
more and more with the film, to make each
film somehow a greater accomplishment
than the preceding one.

That Chaplin's success has been con-
tinual and consistent may properly be
doubted by biographer and reader accord-
ing to his own artistic taste. This book
gives a solid basis for our understanding
the peculiar qualities of Chaplin and his
use of the film medium which led George
Bernard Shaw to call Chaplin "the only
genius in motion pictures." — V.A.

The Little Fellow

The Life and Work of Charlie Chaplin

By Peter Cotes and Thelma Niklaus.
Published (1951) by Philosophical Library,
15 E. 40th St., New York 16. 160 pp.
incl. 32 pp. illus. 5| X 8f. Price $3.75.
There is less about motion pictures in
this book than in the book briefly reviewed
above. There is much more of an effort
by the coauthors to accomplish a psycho-
logical analysis of Chaplin's background,
development and work. There is a deal
of detail beginning generally with Chaplin's

efforts to earn his way at the age of eight,
then on through his growing artistic and
financial successes. At the age of 11 he
successfully achieved the part of Billy in
Sherlock Holmes only by having his mother
drill him with the script, for he had not
yet learned to read or write.

The authors seem fairly occupied in
setting consistently right the considerable
record of matrimonial matters, of which
the public may have an undue aftertaste
from many doses of strong headlines and
lurid inks. The explanations of why
Chaplin's first three marriages were ill
fated and his present one apparently the
contrary are plausible and interesting
enough; but the authors do not quite
explain how anyone, genius or otherwise,
could often create such unbelievably bad
working conditions for himself and then
accomplish the almost superhuman in
completing the motion picture he
wanted — but at other times to be the
effective genius from the start in training
and directing as in The Kid. — V.A.

Acoustical Terminology is American
Standard Z24.1-1951 sponsored by the
Acoustical Society of America in coopera-
tion with The Institute of Radio Engineers.
This latest edition was approved July 31,
1951, and is now available at $1.50 from
the American Standards Assn., 70 E.
45th St., New York 17. A number of
special committees worked to revise this
standard since the first edition was pub-
lished in 1942. The section on speech
and hearing has been thoroughly revised
to bring it into agreement with the most
recent experimental results. Twelve sec-
tions, including six tables, and a thorough
index make up this 50-page standard.

John Wiley & Sons, Inc., is revising its
mailing lists and would appreciate re-
ceiving a postal with the proper address
and an indication of your interest in
scientific, technical or business books.
Address: Miss Clotilda Lowell, John Wiley
& Sons, Inc., 440 Fourth Ave., New York
16.

78

New Members

The following members have been added to the Society's rolls since those last published.
The designations of grades are the same as those used in the 1950 MEMBERSHIP DIRECTORY.

Honorary (H) Fellow (F) Active (M) Associate (A) Student (S)

Archer, Nicholas M., University of

Southern California. Mail: 5965 \

Ghula Vista Way, Hollywood 28. Calif.

(S)
Conner, Robert W., Director of Engineer-

ing, KLAC, KL AC-TV, 1000 Cahuenga

Blvd., Hollywood, Calif. (M)
Hittle, C. E., Design Engineer, RCA Victor

Div. Mail: 12544 Gilmore St., North

Hollywood, Calif. (A)
Jewell, F. Irving, Director, Visual Educa-

tion, National Council, Boy Scouts of

America, 2 Park Ave., New York, N.Y.

(A)
Kook, Edward F., President, Century

Lighting, Inc., 521 West 43 St., New

York, N.Y. (M)
Litecky, Paul A., Photographer, Cine-

matographer. Mail: 1306 Davis Ave.,

Whiting, Ind. (A)
Morris, Thomas C., Camera Operator,

Jerry Fairbanks. Mail: 10552 Tinker

Ave., Tujunga, Calif. (A)
Pierce, Cameron G., Television Engineer,

American Broadcasting Co. Mail: 555

Old Mill Rd., San Marino, Calif. (M)
Pon, S., Salesman. Mail: Corner Best

& Victoria Roads, Sophiatown, Johan-

nesburg, South Africa. (A)
Reisinger, Carl H., Photographer, Free-

lance. Mail: 1417 Kalmia Rd., N.W.,

Washington, D.C. (A)
Rothenberger, Warren Jack, First

Cameraman, Boy Scouts of America.

Mail: 1543 Sidney PI., East Meadow,

L.I., N.Y. (A)
Spriestersbach, Charles L., Television

Technician, KTTV. Mail: 15117

Germain St., San Fernando, Calif. (A)

CHANGES IN GRADE

Berg, Anthony, Chemist, Film Tech-
nician, Twentieth Century-Fox Film
Corp. Mail: 1835 South Burnside
Ave., Los Angeles 19, Calif. (A) to
(M)

Carreau, Gerald, Engineer, Columbia
Broadcasting System. Mail: 285
Avenue C, New York 9, N.Y. (A) to
(M).

Cu minings, James W., Assistant Director
of Photography, National Archives &
Record Services. Mail: 2151 Old
Georgetown Rd., Bethesda 14, Md.
(A) to (M)

Czarda, Theodore, Photographer, Johns-
Manville Co. Mail: Box 75, Sunset
Lake, Pluckemin, NJ. (A) to (M)

Del Valle, G. A., Design Engineer, RCA
Victor Div., Bldg. 10-5, Camden, N.J.
(A) to (M)

Denney, Bruce H., Sound Engineer,
Paramount Pictures Corp. Mail: 418
North Highland Ave., Los Angeles 38,
Calif. (A) to (M)

De Perez, Jose, Av Sonora 67, Mexico
City, D.F., Mexico. (A) to (M)

Dickinson, Edwin A., Recording Engineer,
WestrexCorp. Mail: RoomeRd.,Box
132, Towaco, NJ. (A) to (M)

Didriksen, Roald W., Television Engineer,
KPIX-TV. Mail: 1056 Cole St.,
San Francisco, Calif. (A) to (M)

Dworkin, Sol, Instructor, Motion Picture
Production, Audio-Visual Center. Mail:
121 College PL, Syracuse University,
Syracuse 10, N.Y. (A) to (M)

Fetherston, Joseph A., Sales Manager,
Kollmorgen Optical Corp., 2 Franklin
Ave., Brooklyn 11, N.Y. (A) to (M)

Frazer, Robert E., Chief Engineer, Pacific
Universal Products Corp. Mail: 978
Kent St., Altadena, Calif. (A) to (M)

Genock, Edouard P., Associate Editor,
Paramount News. Mail: 12 Hudson
Ave., Mount Vernon, N.Y. (A) to
(M)

Goldin, H., Engineer, Perkins Electric
Co., Ltd. Mail: 394 Avenue Rd.
Toronto, Ont., Canada. (A) to (M)

Hansen, Dane A., Projectionist, Con-
solidated Film Industries. Mail: 236
South Lincoln St., Burbank, Calif.
(A) to (M)

Howard, A. J., Plant Manager, Con-
solidated Film Industries. Mail: 488
Broad Ave., Leonia, N.J. (A) to (M)

Howell, Barton J., Research Physicist,
Universal Microfilming Corp. Mail:
287 Sixth Ave., Salt Lake City, Utah.
(A) to (M)

Hu, William C. K., Managing Director,
William C. K. Hu & Co. Mail: Rm.
#100, 14 Queen's Rd., Central, Hong-
kong, China. (A) to (M)

Isberg, R. A., Chief Television Engineer,
KRON-TV. Mail: 2001 Barbara Dr.,
Palo Alto, Calif. (A) to (M)

Kirnard, Charles F., TV Engineering,
National Broadcasting Co. Mail: 553
58 St., Brooklyn 20, N.Y. (A) to (M)

79

Klein dienst, Alfred F., Producer, 16-Mm
Sound Films. Mail: Beacon Lodge,
Webster, Mass. (A) to (M)

Kloepfel, Don V., Projectionist, Pro-
jection Engineer, Technicolor, Inc.
Mail: 1900 Keeler St., Burbank, Calif.
(A) to (M)

Knapp, Ned H., Laboratory Mechanic,
Shop Foreman, Columbia Pictures Corp.
Mail: 1525 North Courtney Ave.,
Hollywood 46, Calif. (A) to (M)

Krause, Peter, Control Engineer, Pavelle
Color, Inc. Mail: 69 Bourndale Rd.,
S., Manhasset, L.I., N.Y. (A) to (M)

Lumley, R. Rees, Service Engineer. Mail:
c/o A. S. Eisman, R.F.D. #2, Chitte-
nango, N.Y. (A) to (M)

Magargle, Hal, 162-20 North Hempstead
Tpke., Flushing, L.I., N.Y. (A) to
(M)

Malstrom, Vernon J., Projectionist,
Joseph L. Lawrence Theatres. Mail:
1844 South 17 St., E., Salt Lake City 5,
Utah. (A) to (M)

McCrea, M. W., Service Inspector, Altec
Service Corp. Mail: 476 East High
St., Manchester, N.H. (A) to (M)

Melnicoe, Samuel A., Transmitter Engi-
neer, National Broadcasting Co. Mail:
1751 38 Ave., San Francisco 22, Calif.
(A) to (M)

Mersay, Harry A., Head, Print Dept.,
Twentieth Century-Fox Film Corp.
Mail: 55 Winthrop St., Brooklyn 25,
N.Y. (A) to (M)

Monteleoni, Giulio Cesare, Ispettore
Tecnico, Society Ferrania. Mail: Via
Crispi 10, Rome, Italy. (A) to (M)

Moreno, R. M., Member, Technical Staff,
E. I. du Pont de Nemours & Co., Inc.
Mail: 5 Deerfield Rd., Parlin, NJ.
(A) to (M)

O'Brien, Bernard C., Radio Engineer,
Gannett Co., Inc. Mail: 83 Chelms-
ford Rd., Rochester 18, N.Y. (A) to
(M)

Papalia, Frank V., Precision Film Labora-
tories. Mail: 240 Whiteman St., Fort
Lee, NJ. (A) to (M)

Pavelle, Leo, President, Pavelle Color,
Inc., 533 W. 57 St., New York 19, N.Y.
(A) to (M)

Pesek, A. V., Assistant Plant Engineer,
Cinecolor Corp. Mail: 1334 North
California St., Burbank, Calif. (A) to
(M)

Rollins, Frank S., Jr., Manager, Motion
Picture Dept., E. R. Squibb & Sons.
Mail: 32-14 Northern Blvd., Long
Island City, L.I., N.Y. (A) to (M)

Roquemore, Everett E., Producer, Busi-
ness Films. Mail: 44 Mt. Vernon
Blvd., Hamburg, N.Y. (A) to (M)

Rousseau, Maurice, Writer and Amateur
Cinematographer. Mail: 31 7a St;
Joseph St., Quebec, Canada. (A) to
(M)

Schaeffer, Frederick H., Motion Picture
Engineer, De Luxe Laboratories, Inc.
Mail: 84-09 Talbot St., Kew Gardens
15, L.I., N.Y. (A) to (M)

Shearer, B. F., President, B. F. Shearer
Co. Mail: 2318 Second Ave., Seattle 1,
Wash. (A) to (M)

Sherman, Lawrence F., Jr., Motion Pic-
ture Editor, Free-lance. Mail: 325
E. 72 St., New York, N.Y. (A) to (M)

Shuey, Clyde W., Motion Picture Pro-
jectionist, Pacific Drive-In Theatres
Mail: 2627 McNally Ave., Altadena,
Calif. (A) to (M)

Sjolander, Eric T., 901 W. 34 St., Los
Angeles 7, Calif. (S) to (A)

Slater, Harvey B., Motion Picture Opera-
tor, RKO Albee Theatre. Mail: 64
Clematis St., Providence, R.I. (A) to
(M)

Smith, Holly, Motion Picture Producer,
Hollysmith Pictures, Inc. Mail: 106
South Church St., Charlotte, N.C. (A)
to (M)

Smith, K. R., President, Chief Engineer
K. R. Smith Co. Mail: 183 Leroy
Ave., Darien, Conn. (A) to (M)

Smith, Paul V., Senior Engineer, RCA
Service Co. Mail: Wayne Gardens,
Apt. 32, Collingswood 7, NJ. (A) to
(M)

Smith, Reginald W., Jr., Assistant Editor,
Lawrence F. Sherman, Jr. Mail: 111-
12—103 Ave., Richmond Hill, N.Y.
(S) to (A)B

Stern, Benjamin, Motion Picture Pro-
jectionist, Skouras Pelham Theatre.
Mail: 600 W. 218 St., New York 34,
N.Y. (A) to (M)

Sutherland, Edward P., Chief Engineer,
Motion Picture Section, Signal Corps
Engineering Laboratories. Mail: 158
Monmouth Rd., Elberon, NJ. (A) to
(M)

Taenzer, Erwin, Electronics Engineer,
Farrand Optical Co., Inc. Mail: 1711
Davidson Ave., New York 53, N.Y.
(A) to (M)

Turvey, Carl F., Superintendent, Motion
Picture Laboratory, U.S. Dept. of
Agriculture. Mail: 3710 South St.,
N.W., Washington 7, D.C. (A) to (M)

Underhill, Charles R., Jr., Product
Manager, Motion Picture Screens, RCA
Victor Div. Mail: 255 Rhoads Ave.,
Haddonfield, NJ. (A) to (M)

Varden, Lloyd E., Vice-President, Tech-
nical Director, Pavelle Color, Inc.
Mail: 24 W. 69 St., Apt. 8-A, New York
23, N.Y. (A) to (M)

80

Melvin L. Stewart is the designer of the
Society's new emblem which was adopted
by the Board of Governors last October
and described in the December Journal.
The symbol appears on the cover of this
issue of the Journal and it will gradually
be made a part of the many Society com-
munications. Mr. Stewart resides" at
10326 Orton Ave., Los Angeles. He is a
senior commercial design student at the
University of Southern California, ranks
high in his class and has received three
scholarship awards. He has exhibited
magazine cover designs, book jacket,
fabric and wallpaper designs. He is 23
years of age and is the son of George M.
Stewart who is employed in the Sound
Department of Twentieth Century-Fox
Film Corp., Beverly Hills, Calif.

New Membership Directory

A new directory will be mailed as Part II of one of the Journals late this spring. Plans
are to organize it generally as was done in 1950 — unless members send in valid sugges-
tions for a revised arrangement.

Each member should note that with his statement for dues for 1952 the Society sent
a copy of his last listing brought up to date according to the Society's records at the
beginning of December. In making the new directory, advice about changes in address
or employment will be taken into account at least as late as March 3.

Meetings of Other Societies

American Physical Society, Annual Meeting, Jan. 31 -Feb. 2, Columbia University,

New York

Inter-Society Color Council, Annual Meeting, Feb. 7-9, Hotel Statler, New York
I.R.E. National Convention, Radio Engineering Show, Mar. 3-6, Hotel Waldorf Astoria

and Grand Central Palace, New York

National Electrical Manufacturers Association, Mar. 10-13, Edgewater Beach Hotel,

Chicago, 111.

American Physical Society, Mar. 20-22, Columbus, Ohio
Optical Society of America, Mar. 20-22, Hotel Statler, New York
American Physical Society, May 1-3, Washington, D.C.
Acoustical Society of America, May 8-1 0, New York

American Institute of Electrical Engineers, Summer General Meeting, June 23-27,

Hotel Nicollet, Minneapolis, Minn.

American Physical Society, June 30-July 3, Denver, Cclo.

Photographic Society of America, Annual Convention, Aug. 12-16, Hotel New Yorker,

New York

American Institute of Electrical Engineers, Pacific General Meeting, Aug. 19-22, Hotel

Westward Ho, Phoenix, Ariz.

Illuminating Engineering Society, National Technical Conference, Aug. 27-30, Wash-
ington, D.C.

81

New Products

Further information about these items can be obtained direct from the addresses given.
As in the case of technical papers, the Society is not responsible for manufacturers' state-
ments, and publication of these items does not constitute endorsement of the products.

The Aminco Photomultiplier Micro-
photometer is a product of American
Instrument Company, Inc., Silver Spring
Md., designed for many applications
including film densitometry. This instru-
ment has ranges providing direct readings
for intensities from 20 micromicrolumens
to 20 lumens, densities from 0 to 9 and
phototube currents from 10~6 to 10~u
amp which can be extended with neutral
niters.

Full-scale deflection of the meter is given
with photomultiplier (or phototube) cur-
rents of 10, 1, 0.1 and 0.01 /ta. The latter-

value corresponds to a sensitivity of 20
micromicrolumens per meter division with
photomultiplier tube detector No. 4-6250
which is supplied with the instrument.
Commercial types of phototubes (blue,
green, red and infrared) may be used
by wiring them into an 11 -prong base.

The American Instrument Company
reports that it will supply filters from
Baird, Bausch & Lomb, Corning, Eastman,
Farrand or Fish-Schurman, which are 2
in. (50 mm) square and may be positioned
in the filter holder, either singly or in
combinations up to f in. thick.

SMPTE Officers and Committees: The roster of Society Officers and the
Committee Chairmen and Members were published in the April 1951 Journal.

82

A new professional camera dolly that will
go through most standard doorways
without being disassembled is being
marketed by The Camera Mart, Inc., 70
W. 45th St., New York 19, under the trade
name TV Camera Car. Equally useful in
the motion picture industry, the Camera
Car weighs 350 Ib, is 30 in. wide, and
provides lens angles from 26 in. to seven ft-
The dolly carries the cameraman and
one assistant, and one man can maneuver
it, either on or off dolly tracks. The two
front wheels are set and the two rear wheels
have an auto-linkage steering mechanism
for maneuverability or sharp turns. Two
floor locks steady the dolly for set shots,
and boom arm braces are designed to
prevent vibrations for extended dolly
runs. The tripod head has two leveling
finger-tip jacks for quick horizontal adjust-
ment. In addition there is a vertical
leveling rod attached to the boom arm, a
necessity when setting for a side shot.

Four powerful removable springs and a
cable are arranged to balance any weight
camera and blimp. Raising or lowering
the boom arm is accomplished by turning
the large counter-balanced wheel and
attached gears. The dolly is constructed
of aluminum alloy castings with bridge
supports for greater strength and flexi-
bility, and with 10-in. ball-bearing rubber-
tired wheels.

With the boom in a horizontal position,
the dolly may be lifted into a station wagon
for easy transportation. In addition, in
20 min it may be disassembled into its
three main parts and carried on a location
where the areas are too confined to admit
it otherwise. Reassembling then takes
approximately 30 min. This is considered
an especially valuable feature when shoot-
ing on locations in old buildings with
narrow stairways or no elevators. The
Camera Car is priced at $1,495 FOB
New York.

83

Westrex 1100 Series Magnetic Recording System

Correction and amplification: Running
back to the November 1951 Journal, p.
510, we should now record that the above
illustrates the 1100 Series Portable Mag-
netic System now being introduced to
the industry by Westrex Corporation.
The rest of the record is now played back
for your convenience:

The 1100 Series Portable Magnetic
System now being introduced to the
industry is a direct outgrowth of field
experience with the earlier 1000 Series
System previously described in the Journal
for March 1951. The number of cases
has been reduced to two as shown in the
photograph, the two-position mixer being
on the right and the recorder being on the
.left. The latter houses, in addition to the
film pulling mechanism, the a-c power
supply for the channel, the bias oscillator
and the film monitor amplifier.

New features of this system include two-
way talkback equipment between the
mixer and recordist, a talkback amplifier
being provided in the recorder housing.
Another new feature is a synchronizing
bloop unit which records an audible signal
when the recorder is up to speed on the

magnetic film in synchronism with an
optical bloop in the associated photo-
graphic camera.

The system operates from 115 v, single-
phase, 50- or 60-cycle a-c supply, pro-
vision being also made for motor operation
from 220 v, 3-phase, interlock or multi-
duty motor systems. Runback at normal
speed is provided. The power drain for
the electronic components is somewhat less
than 100 w and a 2-amp drain at 115 v
is required for the single phase motor
supply.

The weight of the complete system,
including cables, is approximately 170 Ib.
The system is available for 35mm, 17^mm
or 16mm operation. The track positions
are in accordance with the proposed ASA
magnetic track standards for 35mm and
16mm films. The recorder may also be
used as a magnetic film reproducer,
equalization being provided in the play-
back amplifier to give an essentially flat
response from 50 to 8000 cycles when
operating at 90 ft/min. By incorporating
some pre-emphasis in recording on 16mm
film, a flat response to 6000 cycles may be
obtained at the 16mm speed of 36 ft/min.

84

Factors Affecting the Quality of
Kinerecording

By P. J. HERBST, R. O. DREW and J. M. BRUMBAUGH

Limitations imposed by television and photographic processes, employment
of electrical compensation for degradations in detail and contrast rendition,
with experimental investigation of various aspects of the system, are re-
viewed. Conclusions regarding improved recording devices and techniques
are offered.

.INESGOPE RECORDING was initially
intended to provide program material
to television stations not connected to
the major origination centers by either
coaxial cable or radio relay facilities.
This served to expand the service, to
distribute the program expense and to
provide advertisers with a larger
audience. The rapid growth of inter-
connection facilities introduces a new
problem due to the time difference
between the point of program origina-
tion and the remote station carrying the
same show. The situation can be
appreciated by examining Fig. 1 which
indicates the problem arising from the
time zone difference across the conti-
nent. A program originating in New
York at 9:00 P.M. would reach the
West Coast at 6:00 P.M., much too
early for presentation. It appears that
the best solution is to record the program

Presented on October 15, 1951, at the
Society's Convention at Hollywood, Calif.,
by P. J. Herbst, R. O. Drew and J. M.
Brumbaugh, Engineering Products Dept.,
RCA Victor Div., Camden 2, NJ.

on the West Coast at the time of origina-
tion and to present a delayed broadcast
from this material at 9:00 P.M. Pacific
Time. Likewise, a program originating
in Los Angeles at 9:00 P.M. would reach
the East Coast at midnight, much too
late for presentation. One proposed
solution is to stage the show at 6:00
P.M. Pacific Time and to transmit it
to the East Coast for presentation at
9:00 P.M. Eastern Time, at the same
time recording the material for later
presentation over the local West Coast
Station.

The concentration of experienced
talent on the West Coast increases the
need for good recordings as does the
continuing necessity of providing pro-
grams for stations not yet connected by
common carrier facilities or unable to
transmit the program at the time of
origination due to other commitments.
While the appeal of television has been
sufficient for the public to tolerate a
considerable amount of degradation in
picture quality, it is obvious that the
system must eventually provide enter-

February 1952 Journal of the SMPTE Vol. 58

85

PACIFIC

MOUNTAIN

CENTRAL

EASTERN

— COMPLETED FALL OF 1948 —SPRING OF I9SI FALL OF I9SI

Fig. 1. Basic common carrier facilities for TV.

tainment of a technical quality consistent
with that of current studio originations.
In view of the importance of this opera-
tion to the entire industry, RCA em-
barked on a broad program of investiga-
tion aimed toward uncovering the sources
of picture degradation throughout the
system and developing the methods
whereby the losses and distortions might
be minimized. The number of indi-
viduals contributing to this effort is
too large for separate recognition here
as will be appreciated from the fact
that personnel at NBC, New York; RCA
Engineering Products Dept., Camden;
RCA Tube Dept. at both Harrison and
Lancaster; the RCA organization in
Hollywood; and the RCA Labora-
tories in Princeton were involved. This
paper is intended as a progress report
to the industry on the investigations
made to date.

Sources of Degradation

While excellent recordings are possible
under present conditions, and are being
obtained with increasing frequency, such
results are not obtained with consistency

and the quality of the poorer recordings
is so far inferior to studio origination as
to cause severe criticism. This picture
quality suffers in the loss of detail, the
distortion of the gray-scale rendition and
in the increase of noise or graininess.
In order to obtain the optimum in
recordings, the limitations of the system
must be understood and the details of
operation must be tailored to fit these
limitations until the various elements of
the system can be improved. The
sources of picture degradation are
illustrated in Fig. 2.

The first factor affecting quality is
the scene lighting. Very contrasty light-
ing or excessive brightness range is
almost certain to introduce spurious
shadowing or compression of the gray
scale in areas of interest. While the
results may not be too bad in the
original broadcast, the further distortion
introduced by the recording and re-
producing processes frequently serve to
exaggerate the original defects to a point
where the net result is hardly tolerable.

The second factor is the operation of
the studio camera. The range in which

86

February 1952 Journal of the SMPTE Vol. 58

Figure 2.

1. Scene lighting 8.

2. Camera operation 9.

3. Camera control and level setting 10.

4. Recording amplifying circuitry 11.

5. Recording kinescope 12.

6. Recording optical system 13.

7. Camera transport mechanism 14.

Film size and emulsion

Photographic processing

Reproducing optical system

Finished recording

Projector transport mechanism

TV camera pickup tube

TV film camera and operation

400 $00 600

Ttlt-VlSIOM LIWC-3

Fig. 3. Effective aperture response, no aperture correction.
Herbst, Drew and Brumbaugh: Quality of Kinerecording

87

an image orthicon will operate without
the introduction of excessive distortions
due to redistribution effects at the target
is not much greater than 30:1; there-
fore, careful control of the iris of the
camera, care in adjusting the operating
potentials to insure a reasonable range
of operation and precise setting of black
levels between cameras are necessary to
obtain a picture of optimum quality.
Unless the program director, the tech-
nical director and the operating per-
sonnel all cooperate in this respect,
there is nothing that the operator of the
recording equipment can do to rectify
their mistakes.

The next link in the chain is the ampli-
fying circuitry associated with the
recording monitor. In general, this
poses no problem since the bandwidth
and signal handling range can be made
adequate. In fact, it is possible to
include some corrective circuits at this
point. The adjustment and mainte-
nance of precise levels are more pressing
problems than any consideration of
losses in the electrical circuits.

The kinescope employed in the moni-
tor represents one of the limiting ele-
ments of the system. Considerable
effort has been devoted to the improve-
ment of this unit as will be described
later. Loss in detail and compression
of the contrast range can be introduced
at this point.

The optical system of the camera,
in fact any lens in the system, can
introduce losses in resolution due to
poor focus or lens flare. At the present
time, these effects are not limiting but
improvement in other elements of the
system may increase the importance of
the losses at such points.

The film transport mechanism in the
recording camera can introduce losses
by either improper motion of the film
or by vibration which causes loss of
interlace and smearing of detail.

The film size and the particular emul-
sion affect both detail and the gray-scale

rendition. The film processing also
introduces loss of detail and distortion
of the contrast depending upon the
exposures employed, the development
of both negative and print and the
precision of the printer.

The television film camera introduces
optical and mechanical losses but the
most important element in this unit is
the pickup tube and its operation.
Spot size and dynamic range affect
both detail and gray-scale rendition.
The latter varies with the particular
tube employed and requires either that
special compensation be employed or
that the characteristics of the recording
be adapted to the characteristics of the
pickup tube. This is one place where
a uniform characteristic is needed, in
order that both normal film and kine-
scope recordings be reproduced with a
minimum of distortion. A review of the
subject was presented by R. M. Fraser
in 1948."

The extent to which fine detail is
degraded even under the best current
practices may be appreciated by an
inspection of Fig. 3. This is a plot of
the effective aperture response of a
good studio pickup and the reproduction
of various types of film material. The
subject has been treated in detail by
O. H. Schade in previous publica-
tions.1'2-4'6 It is, therefore, sufficient to
explain that the plot is in terms of the
relative signal amplitude versus tele-
vision line number.

Several methods of electrically com-
pensating for such losses have been
described and circuits are currently in
use in many recording studios.5 Es-
sentially, these circuits are equalizers
which accentuate the higher video
frequencies representative of the fine
picture detail. The precise shape of the
response curve and the necessity of
including phase compensation to mini-
mize the edge effects due to transients
have been discussed in various publica-
tions. The effect of such an equalizer

88

February 1952 Journal of the SMPTE Vol. 58

or "aperture compensating" circuit is
shown in Figs. 4 and 5. Both are still
recordings of the same television signal.
Figure 4 shows the results obtained
without compensation while Fig. 5
shows the effect when electrical correc-
tion is employed. In this case ' the
compensation was excessive as indicated
by the pronounced edge effects. How-
ever, this was purposely introduced to
reduce the need for further correction
in the film reproducing equipment.
This picture, reproduced over a normal
and well-adjusted television film pickup
chain, gave very excellent results as
regards detail, as shown in Fig. 6. The
degree of improvement can be estimated
by comparing this result with the re-
production of the uncompensated re-
cording shown in Fig. 7. The extent
to which such compensation can be
employed is limited by the increase in
noise, the accentuation of defects in the
original pickup and the introduction of
unpleasant edge effects. It is likely to
vary with different originations and, at
the present time, requires the operator
to exercise good judgment in adjusting
the compensator for any one scene or
program.

The compression of the gray-scale
range as the signal progresses through the
system is shown in Fig. 8. This subject
has also been exhaustively treated by
Schade and others.2-3.4-6.7.8 It will be
noted that the linear range of the original
studio pickup, when properly adjusted,
is in the order of 30 to 1. The same
range for normally processed motion
picture film when televised without
compensation is in the order of 10:1.
When the original television studio
pickup is photographically recorded
and the kinescope recording reproduced
over the television system, the linear
range is only about 4 to 1. It is, there-
fore, essential that some means of ex-
tending the range of the system be
employed. Of course, it is obvious
that care must be taken to keep the area
of interest in the original pickup within

the range of the television system to
avoid washed out highlights and muddy
shadows.11-12

The circuits used for gray-scale
compensation are either expanders or
compressors. The type varies with the
type of tube used in the film reproducing
equipment. Circuits providing a re-
sponse in accordance with a power law
of less than unity are used with linear
pickup devices such as the flying spot
scanner in order to compensate for the
high contrast of the final reproducing
kinescope. Circuits expanding the high-
lights are used to overcome the com-
pression introduced by pickup tubes
such as the iconoscope and by the re-
cording kinescope as well. Circuits
expanding the shadows have been pro-
posed for overcoming the compression
of the lower tones by the toe of the
sensitometric characteristic of the film
stock. The BTL "Rooter" circuits9
and the NBC "Orthogam" amplifier10
are examples of such units.

The effect on the reproduced picture
may be seen in Figs. 9 and 10. The
former is a photograph of a televised
image with no electrical compensation
while the latter shows the results ob-
tained with the same signal by employing
electrical gray-scale compensation.
Aperture compensation was not used
in either case. Approximately 20% of
the original video signal, representing
the maximum "white" excursion, was
stretched to comprise about 50% of
the resulting corrected video signal,
with the gradient increasing toward
the peak of the "white" signals. Simi-
larly about 10% of the "black" signal
was stretched to 20% in the resultant,
in order to compensate for the compres-
sion introduced by the film characteris-
tic. Since the operation of the studio
camera and the film reproducing equip-
ment may both introduce gray-scale dis-
tortions which differ widely from the
representative curves previously shown,
it is difficult to establish an optimum
characteristic for the circuit to be em-

Herbst, Drew and Brumbaugh: Quality of Kinerecording

89

Fig. 4. Kinescope recording, no compensation.

Fig. 5. Kinescope recording, excessive electrical aperture compensation.
90 February 1952 Journal of the SMPTE Vol.58

Fig. 6. Television reproduction on overcompensated recording (shown in Fig. 5).

Fig. 7. Television reproduction of uncompensated recording (shown in Fig. 4).
Herbst. Drew and Brumbaugh: Quality of Kinerecording 91

Fig. 8. Contrast rendi-
tion transfer characteris-
tics of television system.

DtLOTIVt CXP05URt QRBITPODY UNITS

ployed at the recording position. It
should be appreciated that such com-
pensation may accentuate defects such
as flare, shading and halo in a studio
origination so that it is imperative for
the operator at the camera control
position to minimize such effects and
for the program director to avoid calling
for lighting which makes it necessary to
tolerate such defects in order to get a
picture at all.

One method of compensation recently
proposed does not employ electronic
circuits but depends upon photographic
techniques. This method depends upon
a well-known technique called "area
masking" which was described in some
detail at the Society's Spring Convention
in New York.8-13-14 It has the advantage
of accomplishing both gray-scale com-
pression and effective aperture correc-
tion in one process without an appre-
ciable increase in noise. It does, how-
ever, have the disadvantage of requiring
the preparation of a masking print from
the negative and, therefore, requires
evaluation as to its operational and
economic feasibility. The results ob-

tained can be judged from a comparison
of Fig. 9 with Figs. 11 and 12, which
show, respectively, the reproduction of a
television signal from a normal-contrast
subject: firstly, with no compensation
and normal film processing; secondly,
with no compensation but with a low-
contrast print to keep the range of signals
within the limits of the television
system; and thirdly, with a print of
low-contrast range prepared by the
area masking process. The correction
of the contrast range without destroying
the fine detail is easily observed. The
principal limitation of the method is in
the introduction of edge effects which
become objectionable when the com-
pensation is carried to excess.

Kinescopes

The kinescope used in the recorder
is far from perfect. In view of the im-
portance of its performance to the overall
result, a comprehensive program was
aimed at uncovering the manner in
which better performance could be
achieved. Figure 13 shows the details
of the tube which require consideration.

92

February 1952 Journal of the SMPTE Vol. 58

In order to examine the possibilities of
realizing better performance, a large
number of experimental tubes was
constructed and subjected to careful
measurements as well as tested in a
recording setup. The most important
variations tested to date as well as the
results obtained are tabulated in Fig. 14.
The performance of the experimental
tubes is referred to the characteristics of
the RCA Type 5WP-11, currently in
production. It was suspected that light
was dispersed in the phosphor itself
thus increasing the size of the scanning
spot and decreasing the effective aper-
ture response. Three methods of re-
ducing this effect were investigated.
The first consisted in aluminizing the
phosphor without the usual collodion
backing. This permitted the aluminum
to form light barriers between the
crystals. As indicated in row B of Fig.
14, tubes made in this manner exhibited
excessive grain, poor light output and
were hard to drive since the penetration
of the scanning beam was seriously
reduced by the greater thickness of the
aluminum layer. The second approach
consisted in reducing the thickness of
the phosphor deposited on the faceplate.
As indicated in row G of the figure, some
improvement in detail contrast was noted
in these tubes; however, the major effect
was an increase in light output since the
thickness of the deposit was more nearly
optimum for the potentials at which the
tube is designed to operate. The third
method consisted in mixing a small
amount of light-absorbing material with
the phosphor. In the experimental
tubes finely divided carbon was used.
It was found that the gain in fine-detail
contrast was small, that the tubes
exhibited serious graininess, and that
the light output was appreciably reduced
before any appreciable improvement in
detail could be observed. The qualita-
tive results are indicated in row D.
Attempts were also made to reduce the
flare light by decreasing the halation
in the faceplate. With gray glass

faceplates some improvement in detail
contrast was observed and the general
flare was decreased but the film exposure
was excessively reduced when an effec-
tive improvement was obtained. The
performance of these tubes is indicated
in row E of the figure.

Tubes of greater length and with im-
proved gun structures, row F, were tried
in an effort to obtain a finer spot. It
was concluded that the size of the electron
beam was not limiting but that spreading
of the spot in the phosphor and more
especially halation in the faceplate were
the major causes of loss in detail con-
trast and resultant loss in resolution.
The major improvement obtained by
increasing the length of the tubes, row
G, was to decrease the deflection angle
and thereby provide a more uniform
focus. The same result was obtained
by redesigning the deflection yoke
although this introduced geometric dis-
tortion in the form of pincushion. This
last effect can be corrected by suitable
optical means.11

Tubes of larger diameter, row H,
were built and tested in the hope that
the larger image would permit the
realization of improved detail. Since
the beam current had to be increased
to provide the same exposure of the film
stock, it was found that little improve-
ment in this respect was obtained. The
effect of the phosphor grain was reduced
and the image was more readily observed
by the operator and, therefore, easier to
monitor. However, the large size re-
quired to obtain a worth-while advantage
resulted in cumbersome construction
and did not appear to be warranted.

Since the screen brightness appears
to limit the use of methods of reducing
halation and since the Pll phosphor*
saturates at a current density which is
not sufficient to produce the desired

* The PI 1 phosphor referred to has a
spectral energy characteristic peaking
in the blue region, the maximum response
occurring at a wavelength of approximately
4600 Angstrom units.

Herbst, Drew and Brumbaugh: Quality of Kinerecording

93

Fig. 9. Kinescope recording, no compensation.

94

Fig. 10. Kinescope recording, electrical gray-scale compensation.
February 1952 Journal of the SMPTE Vol.58

Fig. 1 1. Television reproduction of a low-contrast print.

-

Fig. 12. Television reproduction of an "area masked" print.
Herbst, Drew and Brumbaugh: Quality of Kinerecording

95

GUN

SPOT SIZE (FOCUS)
EXPOSURE
CONTRAST RANGE

DEFLECTION

— OVERALL FOCUS

— GEOMETRIC DISTORTION

PHOSPHOR
— EXPOSURE
-CONTRAST RANGE
-DISPERSION (FOCUS)

FACE PLATE

— HALATION
- EXPOSURE

DIAMETER

— GRAIN

— MONITORING

'OPERATING POTENTIAL

-EXPOSURE

- CONTRAST RANGE

- DETAIL

LENGTH

— OVERALL FOCUS

— GEOMETRIC DISTORTION

Fig. 13. Features of recording kinescope affecting recorded image.

highlight brightness at the present
operating potential, it was decided to
construct a tube operating at con-
siderably higher voltage, row I. Ex-
perimental tubes similar to those used in
theater television equipment but having
screens of PI 1 phosphor were built and
tested. The results were highly en-
couraging. The available exposure of
the stock was increased several fold
permitting the reconsideration of gray
glass faceplates and possible reduction
of the aperture in the camera lens.
Tubes of this type are now being sub-
jected to tests in order to determine
the optimum operating potential, gun
design and phosphor thickness.

One method of increasing the effective
exposure from the present kinescope
consisted of applying a vertical deflection
in the order of 20 me to the scanning
beam. This deflection was just sufficient
to eliminate the appearance of scanning
lines. Measurements indicated a gain
in light output in the order of 2 to 1 by
the application of this "spot wobble"
technique.

Cameras

While high-quality lenses are em-
ployed in the camera, they are normally

adjusted for optimum correction at
infinite focus. In kinescope recording,
they are used at relatively short distances
and exhibit considerable curvature of
the field. The effect can be partially
corrected by the employment of a suit-
able portrait attachment.

One of the most serious defects which
can be introduced by the camera is the
displacement of the image on the film
by vibration. This can completely
destroy interlace and cause serious
losses in detail. The transmission of
energy to the light gun structure of the
kinescope is apparently the major source
of trouble. Good interlaced recordings
have been obtained by isolating the
camera and kinescope with proper
shock mounts. Care must also be taken
to avoid vibrations from other sources
affecting the relative position of the
kinescope and camera or from causing
mechanical displacement of the gun
structure in the kinescope.

The film transport mechanism can
cause misregistration of the frame and
thereby introduce jump into the re-
corded picture as well as aggravate
the effects of shutter bar. The short
pull-down cycle imposes severe me-
chanical problems in the design of this

96

February 1952 Journal of the SMPTE Vol. 58

EXPERIMENTAL
KINESCOPE

SPOT
SIZE

FOCUS
VS
BRIGHTNESS

CORNER
RESOLUTION

FLARE
LIGHT
HAZE

LIGHT
OUTPUT AT
USEFUL FOG

GRAY
SCALE
JS RANGE

DETAIL
CONTRAST

REMARK

=0

A

RCA
TYPE
5WPII

NORMAL

NORMAL

NORMAL

NORMAL

NORMAL

NORMAL

NORMAL

B

LESS

COLLODION

NORMAL

NORMAL

NORMAL

SLIGHT
IMPROVEMENT

LOW

ABOUT
NORMAL

SOME

IMPROVEMENT

EXCESSIVE
GRAIN

o£vEESS<VE

REQUIRED

C

THIN
SCREEN

NORMAL

NORMAL

NORMAL

SLIGHT
IMPROVEMENT

APPROX
2:1

SOME

IMPROVEMENT

SOME
IMPROVEMENT

<

D

CARBON IN
PHOSPHOR

NORMAL

NORMAL

NORMAL

SOME
IMPROVEMENT

LOW

ABOUT
NORMAL

SOME

IMPROVEMENT

EXCESSIVE

r,RAIN

^c

C

GRAY GLASS
FACE PLATE

NORMAL

NORMAL

NORMAL

SOME
IMPROVEMENT

LOW

ABOUT
NORMAL

RELATIVELY
WORSE

s<7

F

IMPROVED
GUN

NORMAL

NORMAL

SOME
MPROVEMINT

NORMAL

NORMAL

NORMAL

NORMAL

<

G

DECREASED
DEFLECTION
ANGLE

NORMAL

NORMAL

SOME
MPROVEMENT

NORMAL

NORMAL

NORMAL

NORMAL

^K

H

LARGE

TUBE

NORMAL

NORMAL

NORMAL

NORMAL
FOR FINE
DETAIL

ABOUT
NORMAL

ABOUT
NORMAL

BETTER
MONITORING

^

'

50 KV

UNDER INVESTIGATION

Fig. 14. Tube variations tested, with results of tests.

mechanism and sometimes the motion
of the film was not completely stopped
before the shutter opened. This re-
sulted in smearing of the television lines
over this region of the image. These
defects have been entirely eliminated in
experimental equipment by the use of
a modulated pressure pad and the re-
design of the registration pins.

The methods currently employed to
photograph the 60-field/sec image of
the television system on the 24-frame
film used in motion pictures all require
that the television image be "spliced"
at some point within the picture area.18
When mechanical shutters are employed,
an interval of several lines is required to
complete the operation of opening and
closing. During this interval, the rate
at which the illumination increases and
decreases must be carefully balanced so
that a uniform exposure is obtained.
When properly adjusted, the splice is

invisible. When this balance is not
obtained, the difference in exposure
causes a fluttering or local flicker to
appear in the vicinity of the "splice."

Instead of a mechanical shutter, it is
possible to blank out the picture on the
face of the kinescope by electronic means.
Such an electronic shutter was success-
fully demonstrated several years ago.
Recently, the technique was reinvesti-
gated to determine the practical ad-
vantages and limitations which it offered.
The realization of adequate precision
in the timing devices necessary to
produce the special blanking signal was
not difficult to achieve. However, the
stability and regulation required of all
associated circuits appeared to be exces-
sive since exact interlace must be ob-
tained at all times. Small displace-
ments of the raster not observable with
a mechanical shutter appear as white or
black horizontal streaks in the picture.

Herbst, Drew and Brumbaugh: Quality of Kinerecording

97

Moreover, the interruption of an image
of this brightness for a relatively long
period and at a 24-frame rate produces
a flicker which makes it difficult to
observe the image for any extended
period of time and prevents continuous
monitoring by the operator. In either
case, precise adjustment and the ap-
plication of proper mechanical or elec-
trical shading to balance the exposure
resulted in the elimination of such
defects. The matter is more one of
maintenance than of optimum per-
formance.

Photographic Processes

The choice of currently available
film stock suitable for kinescope recording
is limited. High-speed negative emul-
sions are generally too grainy, especially
for 16mm recordings, while fine-grain
emulsions are usually confined to rela-
tively slow-speed materials. With these,
it is difficult to obtain a sufficient range
of exposure due to the relatively low
brightness of the kinescope. Care must,
therefore, be exercised in the setting of
the black level in order to avoid com-
pression in the shadows since the stock
is usually worked at lower than average
densities while at the same time the
highlights must be adjusted to realize
the maximum exposure without phos-
phor saturation or spot defocusing. On
Eastman Kodak Co. Type No. 7273
film stock, comparatively good results
can be obtained by exposing to obtain
densities of 0.2 and 1.2, respectively, with
development to a control gamma of
approximately unity. In general, cur-
rent practices appear to be employing
the available emulsions to the limit of
their possibilities and little advantage
seems likely to accrue from variations
in processing parameters unless the
exposure can be appreciably increased.

Because the available highlight bright-
ness in the kinescope image at current
densities less than saturation does not
provide normal exposure of the re-
cording stock, the operators frequently

tend to overdrive the kinescope in
order to obtain an apparent good con-
trast range on the negative. This tends
to exaggerate any compression in the
original signal and is probably the
principal reason for the criticism leveled
at the recording technique.

There are several different approaches
to the recording, distribution and re-
production of kinescope recordings. The
first and most direct approach employs
a positive image on the face of the
kinescope and produces a negative
image on the recording stock. This
negative may be televised, thus elimi-
nating the losses involved in the printing
process, or positive prints may be made
from it for release to other stations.
The former process may be used for
local delay broadcasts or for cases where
the time required to transport the
original negative to the remote station
can be tolerated and where only one
such transmission is involved.

The second approach is to employ
a positive image on the kinescope but
to photograph this image on reversal
stock. Good results have been obtained
in this manner but the lack of positive
prints for distribution makes its applica-
tion limited as in the case of employing
a direct negative for reproduction.

The third approach employs a nega-
tive image on the face of the kinescope,
and provides a direct positive for re-
production over the television system.
The film thus obtained is likely to be of
low contrast and may not appear satis-
factory for direct projection. However,
good results have been obtained over
the television system. The method is
open to the objection that no release
prints are available.

The fourth method is to photograph
a negative image on the kinescope on
reversal stock. The reversal negative
thus obtained can then be used to pro-
duce release positive prints. Very poor
results have been obtained with this
technique.

98

February 1952 Journal of the SMPTE Vol. 58

Since prints for release should be
positives to permit the insertion of local
trailers and since such releases are
necessary for all conditions except for
network originations, it appears that
the conventional method of photo-
graphing a positive kinescope image 'on
normal stock offers the best practical
solution.

The losses introduced by the printing
process have also been investigated.
Carefully measured negatives have been
distributed to processing laboratories
with requests for processing under usual
conditions so that data relating to the
differences between various printing
methods might be obtained. Un-
fortunately, it has been difficult to
obtain the full cooperation of sufficient
laboratories to permit definite conclu-
sions to be drawn. The effort is con-
tinuing and it is hoped that eventually
it will be possible to recommend im-
provements in both printers and printing
techniques.

For some time, the use of 35mm film
has been suggested as a means of realiz-
ing greater detail in recordings. While
this may prove to be practical for net-
work usage where stations are equipped
with 35mm television film projectors,
it will be necessary to distribute release
prints on 16mm film since many local
stations operate with 16mm projectors
and do not have 35mm facilities. Fur-
thermore, even though most release
prints are on acetate stock, local fire
ordinances require special precautions
when 35mm film is used to insure
protection against the possible employ-
ment of nitrate film. It seems unlikely
that such restrictions will be removed
in the near future. It will, therefore,
be necessary to accept the limitations
on detail which the smaller film size
imposes. While the difference in detail
is significant, acceptable results can be
obtained on 16mm film by minimizing
the losses in the remainder of the system
and by employing optimum electrical
aperture correction. Fortunately, the

emulsions are identical so that no further
loss in gray-scale rendition is introduced.

Monitoring

It will be appreciated that the narrow
range of the television system and the
need to employ this range to the best
advantage in exposing the film will
require considerable precision in estab-
lishing brightness levels on the face of
the kinescope. Visual observation is
not sufficiently accurate for this purpose
and monitoring the video signal applied
to the recording kinescope does not
measure actual brightness and is limited
in the precision with which the observa-
tion of levels can be made on the
associated oscilloscope displaying the
video waveform. In order to provide
a more reliable means of establishing the
recording levels, a generator was de-
veloped which furnished a video signal
in the form of a series of steps. This
generator is, therefore, the electrical
equivalent of a photographic density
wedge. The overall amplitude of this
signal is adjusted to fit the video range
of the signal provided from a master
control. The picture produced on the
face of the kinescope is, therefore, a
group of bars of varying brightness
each of which represents a distinct video
voltage level. The pattern on the face
of the kinescope is picked up by a cali-
brated photocell and the output is dis-
played on an oscilloscope. The bars
are arranged horizontally in order to
minimize the effects of phosphor decay
which would tend to average the bright-
ness along a horizontal line if vertically
disposed bars were employed.

Film Reproduction

In order to complete the review of
the entire system, it is necessary to
include the performance of the pickup
tubes available for television film re-
production and the characteristics of
the projectors associated with them.

The pickup tubes of current interest
include the iconoscope, the image

Herbst, Drew and Brumbaugh: Quality of Kinerecording

99

4. 8 10 ZO 40

ORIGINAL LUMIWftMCC- (ORBITRORY UNITS)

Fig. 15. Effect of gamma compensation on signal-to-noise ratio.

80 100

orthicon, the flying spot scanner and the
vidicon. The gray-scale characteristics
of these devices differ in several ways
and affect the type of compensation
that is inserted in the system. This
aspect has been discussed previously
at some length in several publications.16-17
The noise characteristics of the several
tubes differ so that the degree to which
compensation may be employed to
advantage also varies. The relative
performance of several tubes as regards
noise in the reproduced image may be
estimated from the representative charac-
teristics plotted in Fig. 15. It will be
noted that a linear device such as a
vidicon is seriously limited if the noise
originates in the first amplifying stage.
This limitation will be alleviated if the
device can be made to have a power-law
response of less than unity. Image
orthicons of the presently available
types have been employed as film
pickup devices and are currently in use

to some extent. Under present condi-
tions the relatively narrow range over
which these tubes can be operated with-
out gray-scale distortion, the suscepti-
bility to burning-in of fixed images and
the rather critical adjustments required
for satisfactory operation present prob-
lems which must be considered in
evaluating their application to present-
day operations. Therefore, although
future developments may remove these
difficulties, interest is currently directed
toward the iconoscope and the flying
spot scanner for this application.

The iconoscope is a storage device
and as such, permits the use of projectors
of conventional type since the storage
permits the use of long pull-down cycles.
However, best results are not obtained
under such conditions and the use of
long application times with short pull-
down cycles has been shown capable
of reducing such undesirable charac-
teristics as shading and edge flare.

100

February 1952 Journal of the SMPTE Vol. 58

KINESCOPE

AUX. LENS

SEMI -SILVERED
MIRRQR —

MIRROR I

MAIN LENS

MIRROR 2

PHOTOCELL

Fig. 16. 30-Frame alternate path projector.

cx

FILM TRANSPORT

Moreover, the performance can be fur-
ther improved by the application of
proper edge lighting and back lighting
as well as circuitry to provide an essen-
tially constant background signal level
under wide changes in scene brightness.

The flying spot scanner does not make
use of the storage principle and, there-
fore, requires the scanning spot to be
maintained in proper register with the
film at all times. The two types of
projectors which have been suggested
to accomplish this are either the fast
pull-down projector, in which the film
is pulled down in the television blanking
interval, or the continuous projector,
in which the film moves at a constant
rate and proper registration is main-
tained by either optical or electrical
displacement of the image of the scanning
spot.

Several experimental models of fast
pull-down projectors have been built.
These devices have all been designed for
16mm film. Obviously the problems
associated with minimizing the wear on
both film and mechanical parts due to
the high accelerations involved will
require extensive life tests before the
practicability of any design of a film
transport mechanism can be evaluated.
The ability of the mechanism to provide
accurate registration in such a short
pull-down time also presents mechanical
problems. The results obtained on one
model have been highly encouraging.
After more than 400 passes through this
model, no damage to the film sprocket
holes was observed. Moreover, no per-

ceptible increase in jump was observable
with the SMPTE test film after more
than 50 hours of operation. While
these results must be confirmed by
further operation over an extended
period of time, it would seem that
flying spot scanner operation with this
type of projector is within the realm of
possibility.

Continuous projectors are of two
fundamental types: (1) those that allow
the film motion to accomplish a portion
of the vertical scanning and employ
optical or electrical means to deflect
the scanning raster to the proper position
at the beginning of each scan; and (2)
those that use continuously varying
optical means to maintain registration
of the raster and the film as the film
moves.

The first type has been used in Europe
where the 50-field television standard
permits a relatively simple alternate
positioning of the raster by running the
film at a rate of 25 frames. For U.S.
television standards, this would require
a special 30-frame film. One version
of such a projector is illustrated in
Fig. 16. Since there are only two
television fields for each film frame, only
two positions of the scanning raster are
required. The system shown employs
mirrors to deflect the beam and provide
the two alternate paths. Since the
mirrors can be relatively large, a pro-
jection lens of high speed can be used.
A projector of this type was built and
tested and found to operate extremely
well. Care must be taken to maintain

Herbst, Drew and Brumbaugh: Quality of Kinerecording

101

FILM POSITION

KINESCOPE

Fig. 17. Electronically deflected alternate path flying
spot scanner for 30-frame film.

equal transmission over both of the paths
to avoid flicker at a 30-cycle rate.

When 24-frame film is to be televised
at a 60-field rate, it is necessary to scan
one film frame with two television fields
and the next with three. In a projector
of either of the foregoing types, five
alternate paths must be provided. In
the multiple-lens case, this means a
still further reduction of lens speed. In
the second system, cams may be em-
ployed to position the unused mirror
while the other is in use and held sta-
tionary. An experimental model of
such a unit has been built but tests are
not sufficiently advanced to permit an
evaluation of its merits at this time.

Instead of shifting the effective position
of the raster by interposing optical
elements between the kinescope and the
objective lens, it is possible to accom-
plish the same result by displacing the
raster vertically on the face of the
kinescope. The principle, as applied
to 30-frame film, is illustrated in Fig. 17.
It will be appreciated that extreme
linearity of scanning and a minimum of
geometric distortion must be achieved
in order that the registration of successive
fields be obtained. The system has also
been tried for 24-frame film and U.S.
Television Standards. The problem of
registration is of the same order of
magnitude as before but an additional
problem of unequal duty cycle of various
areas on the face of the kinescope is

introduced due to the unavoidable
overlapping between the displaced
rasters. As the tube ages, objectionable
flicker may be introduced should this
produce unequal light output from the
several areas.

Projectors in which moving optical
systems have been employed to maintain
constant registration with a moving
film have been attempted for some time.
These have employed rotating lens disks
and drums, rotating prisms and rotating
mirror systems. The first two systems
require extreme precision in the optical
elements, are inherently low in light
efficiency and are difficult to compensate
over the desired range of travel. The
last system was originally developed by
Meccau in Germany and produced
satisfactory results although it was
difficult to maintain proper adjustment.
Recently this principle has been revived
and appears to offer considerable promise
of satisfactory and practical operation.

The promised improvements in film
pickup equipment will considerably
improve the results obtained from
kinescope recordings as well as from
normal film material by eliminating
many of the defects present under current
conditions, and by providing more stable
and uniform characteristics, thereby
reducing the variability injected by the
continual adjustment in accordance with
the operator's judgment.

102

February 1952 Journal of the SMPTE Vol. 58

Summary of Current Status

The present status of kinescope re-
cording may be summarized as follows:

1. Good quality is possible with
present equipment by careful control
of the lighting and staging and by
proper operation of both the studio
camera and the film reproducing equip-
ment. This imposes objectionable re-
strictions on programming but must be
tolerated until further improvements
can be realized.

2. Some improvement can be ob-
tained by the use of electrical correcting
circuits but care must be taken to avoid
overemphasizing the defects in the
original pickup.

3. Some improvement in kinescopes
has been obtained. Current limitations
are expected to be removed to some
extent by the use of higher operating
potentials.

4. Satisfactory camera mechanisms
can be realized and either the mechanical
or electronic shutters can be adjusted to
eliminate visible shutter bar. The latter,
however, introduces problems in main-
tenance and monitoring.

5. Better monitoring techniques can
be employed. The use of a step function
generator and a photocell monitor
promises to provide greater uniformity
and more precise exposure.

6. Present photographic processes ap-
pear to be capable of very little improve-
ment under present conditions. An
increase in kinescope brightness or the
introduction of a new emulsion might
make changes in processing desirable.

7. The photographic compensating
technique known as "area masking"
offers advantages but requires evaluation
from the operational and economic
standpoints.

8. Improved film pickup equipment
offers the possibility of minimizing the
losses introduced during reproduction.

Conclusion

It must be emphasized that the losses
and distortions in the system are the

summation of a number of individual
deficiencies and that an improvement
in one element may be masked by deg-
radation contributed by the remainder
of the process. No one element is re-
sponsible for the overall loss of detail
or distortion of contrast so that many
small improvements must be attained
before an outstanding improvement is
made in average reproductions. Until
such time as these improvements can be
included in the system, program direc-
tors and technical directors would be
well advised to maintain a careful control
over lighting, staging and camera opera-
tion if they expect acceptable quality
to be realized in the recorded program.

References

1. O. H. Schade, "Electro-optical
characteristics of television systems:
Part I, Characteristics of vision and
visual systems," RCA Rev., 9: 13-37,
Mar. 1948.

2. "Part II, Electro-optical specifications
for television systems," ibid., 245-286,
June 1948.

3. "Part III, Electro-optical charac-
teristics of camera systems," ibid.,
490-530, Sept. 1948.

4. "Part IV, Correlation and evaluation
of electro-optical characteristics of
imaging systems," ibid., 653-686, Dec.
1948.

5. E. D. Goodale and R. C. Kennedy,
"Phase and amplitude equalizers for
television use," RCA Rev., 10: 35-42,
Mar. 1949.

6. O. H. Schade, "Image gradation,
graininess and sharpness in television
and motion picture systems: Part I,
"Image structure and transfer charac-
teristics," Jour. SMPTE, 56: 137-177,
Feb. 1951.

7. F. G. Albin, "Sensitometric aspect of
television monitor-tube photography,"
Jour. SMPE, 51: 595-612, Dec. 1948.

8. P. J. Herbst, R. O. Drew, and S. W.
Johnson, "Electrical and photographic
compensation in television film re-
production," Jour. SMPTE, 57: 289-
307, Oct. 1951.

9. B. M. Oliver, "A rooter for video
signals," Proc. IRE, 38: 1301-1305,
Nov. 1950.

Herbst, Drew and Brumbaugh: Quality of Kinerecording

103

10. E. D. Goodale and G. L. Townsend,
"The orthogam amplifier," RCA Rev.,
11: 399-410, Sept. 1950 (abstract,
Jour. SMPTE, 56: 76-78, Jan. 1951).

11. W. T. Wintringham, "Tone rendition
in photography," Proc. IRE, 38:
1284-1287, Nov. 1950.

12. B. M. Oliver, "Tone rendition in
television," Proc. IRE, 38: 1288-1300,
Nov. 1950.

13. M. V. Johnson, "Print control with
blurred positive masks," Am. Phot.,
37: 14-16, Mar. 1943.

14. J. A. G. Yule, "Unsharp masks," /.
Phot. Soc. Am., 11: 123-132, Mar. 1945.

15. R. M. Fraser, "Motion picture photog-
raphy of television signals," RCA Rev.,
9: 202-217, June 1948.

16. R. B. Janes, R. E. Johnson and R. S.
Moore, "Development and perform-
ance of television camera tubes,"
RCA Rev., 10: 191-223, June 1949.

17. R. L. Garman and R. W. Lee, "Image
tubes and techniques in television film
camera chains," Jour. SMPTE, 56:
52-64, Jan. 1951.

18. F. N. Gillette, "The picture splice as
a problem of video recording," Jour.
SMPTE, 53: 242-255, Sept. 1949.

Discussion

Anon: Where on the response curve
does the corrected mask fit on the chart
that was shown for the response for the
live studio 35mm film and the kinescope
film?

P. J. Herbst: I don't get the question.
Where, on which response curve is what?

Anon: There was one of the charts
shown on the screen, one that showed the
response for a live pickup for 35mm and
for 16mm, and I just wondered where the
masked area. . .

Mr. Herbst: Oh, you mean area masking,
how much that improved it?

Anon: That's right.

Mr. Herbst: I'm sorry I hadn't under-
stood. I would say this. I think that
just from observation it brings it up to
something; it'll bring a 16mm recording
up to something that is not quite as good
as the 35mm reproduction over the TV
system.

Anon: In the interests of settling the
question, which is the more authoritative
source of technical information, the gossip
columns or the proceedings of this learned
Society? I would like to have you express
something about the future of photographic

kinescope recording vs. magnetic recording
of the image.

Mr. Herbst: A magnetic recording would
be fine if we knew how to do it.

Anon: You probably haven't been read-
ing the Hollywood columns in the last
couple of weeks, but it's supposed to be a
reality.

Mr. Herbst: I haven't seen any of it yet.
I'm sorry.

Anon: I wonder if we can still stay in
business.

Mr. Herbst: I think, to answer your
question, that unless someone has come
out with magnetic heads and magnetic
materials which are capable of much higher
frequencies than we've gotten so far (and
that may be possible), until that happens,
I don't think that photographic kinescope
recording will be abandoned.

F. JV. Gillette: Regarding the question
that was raised just a moment ago, is it
really proper to consider this area-masking
technique as a means of improving the
response curve of the system? It's more a
means of improving the tone values,
isn't it?

Mr. Herbst: Yes, but it also improves
the large-area tone values — in other words,
it reduces them to a value which the system
can handle. At the same time, it leaves
the fine-detail contrast where it was. So,
essentially it's exactly the same thing as
increasing the gain at the higher fre-
quencies. I think that Otto Schade's
old paper some years ago pointed out
that you could do that. It doesn't make
any difference whether you do it electrically
or photographically.

Dr. Gillette: But actually does it amount
to an increase in fine-detail contrast in
any region which was properly treated by
the techniques previously used?

Mr. Herbst: Oh, yes, it is increased in
every region. Look at it this way. The
mask is merely a way of reducing in any
given area the exposure which is given to
the print. If you did the same thing by
dodging in an enlargement you wouldn't
reduce the detail contrast any. You
would merely reduce the overall contrast
between large areas. The result is an
increase in detail contrast relative to large-
area contrast in all ranges of the picture,
not just in the highlights and in the
shadows.

Anon: You mentioned earlier some
improvements on iconoscope film chains.
Is information on these improvements
available?

Mr. Herbst: Well, we expect to have
that out shortly.

104

February 1952 Journal of the SMPTE Vol. 58

Multichannel Magnetic Film
Recording and Reproducing Unit

By C. C. DAVIS, J. G. FRAYNE and E. W. TEMPLIN

The multichannel magnetic film recorder and reproducer provides three
200-mil tracks in accordance with proposed ASA standard for 35mm film.
The effective crosstalk between adjacent tracks approximates —60 db and
flutter content does not exceed 0.05%. Complete recording and reproducing
transmission equipment is housed in the recorder and associated base cabinet.
The recording channels operate from a nominal input level of —30 dbm,
and a reproduced output of +16 dbm is obtained from each of the reproduc-
ing channels. Monitoring of each channel is provided from a separate
triple-track head.

_L HE COMBINATION magnetic recorder-
reproducer described in this paper was
developed to meet the needs of the
motion picture industry for a high-
quality triple-track magnetic recorder.
The use of a triple-track recorder was
anticipated by the industry in formulat-
ing the magnetic-track standards for
35mm sprocket-hole film, provision being
made that one of the three tracks re-
corded in such a machine should corre-
spond in position with that of a single
track recorded in an ordinary magnetic-
film recorder. In fact, the performance
specifications for the triple-track ma-
chine, especially those which specified
the crosstalk between adjacent tracks,

Presented on October 18, 1951, at the
Society's Convention at Hollywood, Calif.,
by a C. Davis, J. G. Frayne and E. W.
Templin, Westrex Corp., 6601 Romaine
St., Hollywood 38, Calif.

proved to be a determining factor in the
location of the single track in the regular
motion picture production magnetic
recorder. The final triple-track stand-
ards as adopted by the Society's Sound
Committee and which are now being
considered for standardization by ASA
are shown in Fig. 1. Reference to this
figure will show that three 200-mil
tracks are provided with a separation
of 150 mils between tracks, the edges of
the outside tracks being 50 mils from
the sprocket holes. The proposed stand-
ard calls for a crosstalk figure between
adjacent tracks of at least — 50 db.

At the time of the formulation of the
track standards, it was thought that the
— 50 db value of crosstalk would be
satisfactory for the then intended uses of
the triple-track recorder. However, as
the industry began to use this new
medium, the demand for a greater
reduction of crosstalk became imme-

February 1952 Journal of the SMPTE Vol. 58

105

RECORDING OR
REPRODUCING
HEADS IN LINE

TRAVEL

BASE DOWN

O.I89

0. ISO

-O.I50-

0.339

0.689

±0. 004

-35 MM

ALL DIMENSIONS IN INCHES
Fig. 1. Triple-track magnetic specifications.

diately evident and, as will be described
later in this paper, means have been
found for obtaining a crosstalk figure of
approximately —60 db at 1000 cycles
on a regular production basis. Ex-
haustive listening tests have shown that
with this amount of crosstalk at 1000
cycles, no audible sound is heard in
any track from a fully modulated signal
in an adjacent track, whereas with the
original value of —50 db, audible cross-
talk is very much in evidence. If
completely unrelated sound recordings
are to be made on the three individual
tracks, it seems highly necessary that a
successful triple-track magnetic recorder
meet the — 60-db cancellation figure.

The intended uses for this type of
magnetic recorder include the multi-
channel scoring operation in which this
single machine would replace three
existing single-channel recorders with
the attendant economy in film usage
and in manpower, as well as provide
a much greater convenience in operation.
A second major use of the triple-track
recorder is that of providing storage of
three individual tracks on a single film,
thus providing a marked saving in vault
space. These three tracks could be

entirely unrelated or they could be used,
for example, for storing dialogue, music
and effects tracks on a single film.
Other uses, of course, will be found
particularly in the re-recording opera-
tions as the studios get more familiar
with the possibilities of this type of
recorder.

In order to facilitate the early intro-
duction of this new recorder to the
industry, it was decided to utilize the
basic mechanism of the RA-1251 Re-
recorder1 which has had such wide
acceptance in the industry for both
photographic and magnetic re -recording.
In order to accommodate a triple-track
head providing three recording heads
in line and locating them in a low-
flutter film path, a double flywheel
drive was substituted for the customary
single unit. Two complete triple-track
heads, one used primarily for recording
and the second for monitoring or play-
back, are mounted in the film path
between the two impedance drums
mounted on the separate flywheel shafts.
The arrangement of the film path and
location of the two triple-track heads are
shown in Fig. 2. It will be noted* that
the elements of the Davis Drive, pre-

106

February 1952 Journal of the SMPTE Vol. 58

IMPEDANCE DRUM

Fig. 2. Triple-track recorder film path schematic.

viously employed in the RA-1251, are
retained. The single flywheel is re-
moved, a new subassembly is substituted
which carries the two individual fly-
wheels, the mounting for the triple-
track heads and a Permalloy shield box
to isolate the heads as far as possible
from magnetic pickup of extraneous
fields. The combined moment of inertia
of the two flywheels approximates that
of the single flywheel, so that the natural
period of the filtered film path remains
essentially the same. The performance
of this drive from a flutter standpoint
is quite comparable to that of the RA-
1251 Re-recorder when set up originally
for photographic purposes. The total
rms flutter for an average machine
amounts to approximately 0.05% based
on flutter frequency rates ranging from
2 to 200 cycles, the flutter at any given
rate not exceeding 0.03% rms.

Figure 3 shows an electrical analogue
of the film-drive filter mechanism, the
components being designated below the
illustration.

The basic elements of this circuit were
previously shown to illustrate the double-
arm or tight-loop filter mechanism for
a photographic film recorder.2 At that
time an explanation was offered for
the action of the common spring, Q
and the double-sprocket drive, Si and
S2, including the general characteristics
and attenuation curves of flutter dis-
turbance originating in either or both
of the sprockets.

The present circuit shows the addition
of six elements which represent the
additional flywheel and the significant
items associated with the film passage
over the magnetic heads. While six
elements have been added, the film-
filter performance remains substantially

Davis, Frayne and Templin: Multichannel Magnetic Recording

107

M, *i vs M2

^M^HP-J^/VvYAA^/^

M3

C5

C4

R2'

C7

C9

C8

C6

MI & M2,
MS & M4,

& Ga,

G4,

& Ce,

INTERNAL FILTER CIRCUIT
Fig. 3. Film-drive electrical analogue.

C? & Cg, Compliance

Inertia of flywheels

Inertia of upper & lower filter

arms
Compliance of spring common

to both arms
Compliance of arms when mov-

ing together
Compliance of arm-positioning

spring
Compliance of film between

flywheels & sprockets

unchanged from a single-flywheel type.
This is because the elements C7, C8 &
C9 and R2 & R3 represent relatively
small magnitudes, and MI & M2 tend
to become a single flywheel as these
elements decrease in value. C7, C8 &
C9 are short, stiff lengths of film and,
therefore, constitute small values of
compliance. Likewise, R2 & RS, repre-
senting the effective damping resistance
of the film friction over the heads, have
relatively small values. This results
from the well-known characteristic of
solid or sliding friction acting at con-
siderable velocity, as compared to
viscous friction, because of their differ-
ences in force-velocity characteristics.3
This may be illustrated by removing
the dashpot, Rb whereby the small
remaining amount of damping caused
by film friction permits highly undamped
oscillation of the filter arms following a
disturbance.

of film between
flywheels & heads
Cg, Compliance of film between

heads
Si & 82, Upper & lower film drive

sprockets

RI, Resistance of dashpot damper
R2 & RS, Resistance of film over heads
Vi & V2, Reference film velocity at heads

The displacing force created by film
friction over the magnetic heads is offset
by an adjustable spring, C4. By this
means the filter arms can be maintained
in their correct operating positions in
spite of large differences in the frictional
coefficient of various film samples.

In the present design, forward-running
speed of 90 ft/min is provided and the
customary high-speed rewind is retained.
For special applications where reverse
operation at standard speed is required,
a double torque-motor drive will be
furnished for each film-spool spindle.
A footage counter located in the central
angle bracket is an added feature of the
triple-track recorder.

The associated transmission equip-
ment providing for three complete re-
cording channels and three complete
reproducing channels, operating at a
nominal recording input level of —30
dbm and reproducing an output level

108

February 1952 Journal of the SMPTE Vol. 58

of +16 dbm, is housed in both the upper
section of the recorder cabinet and in the
associated base cabinet, as will be
observed from Fig. 4. This provides
for a maximum economy in recording-
room space as well as in all the facilities
and controls needed for operation of a
triple-track machine. Complete details
of the recording and reproducing trans-
mission circuits and controls are de-
scribed later in the paper.

Triple-Track Magnetic Head

The triple-track RA-1508 Magnetic
Head shown in Fig. 5 is based on con-
struction principles used in single-track
heads previously described.4 Basically,
it is a ring type with two stacks of
Permalloy laminations cemented in the
divided halves of a hollow ring. This
machined brass ring provides accurate
reference surfaces for the otherwise
irregular dimensions of the pile-up of
laminations. These serve as a founda-
tion for the manufacture of identical
units which are combined into multi-
track heads exhibiting close tolerances
relative to track placement and azimuth
alignment as a group. Thus, no indi-
vidual adjustment is required for azimuth
or track adjustment. The groups com-
prising the recording and reproducing
heads are placed on a mounting bracket
which provides facilities for adjusting
the record and reproduce heads as units.
The film is aligned with the entire as-
sembly by an adjustable guide-roller as
in the original photographic machine.

Interference or crosstalk in multi-
track magnetic recorders, wherein the
tracks are recorded simultaneously in
line across the magnetic medium, results
mainly from the fact that several heads
lie side by side, separated by incomplete
shielding since a considerable portion
of the heads must be exposed to contact
the recording medium. This type of
crosstalk may be referred to as electrical
because it evidences itself without the
presence of any recording medium. It
is the source of troublesome crosstalk

Fig. 4. Front view of RA-1506-A
triple-track recorder.

Davis, Frayne and Templin: Multichannel Magnetic Recording

109

in program material. A close phase
relationship may be shown to exist
between the original recorded track
and the track produced by the induced
flux in the adjacent head. Another
form of crosstalk exists only at low
frequencies or long wavelengths and
normally is not a source of trouble in
program material because of ear charac-
teristics and the energy distribution of
speech and music. It occurs when the
recorded wavelength becomes compar-
able to the distance between tracks and
the fringing flux becomes well defined
and of such intensity in the adjacent
track area as to constitute interfering
crosstalk.

Special means have been incorporated
in the recording- and reproducing-head
assemblies to reduce crosstalk among
the three heads. These consist of small
magnetic paths introduced diagonally
between one-half of each magnetic head
and the corresponding opposite half of
the adjacent head, and of such propor-
tions and phase relationship as to cancel
effectively the crosstalk leakage from
one head to the other. These sub-
stantially decouple the two heads elec-
trically or magnetically and are referred
to as decouplers. In the case of a triple-
track head only two decouplers are
required since their action is, for all
practical purposes, reversible. They
handle relatively small amounts of flux
and do not alter the normal charac-
teristics of the recording and reproducing
heads or the overall system in any way.

Without the application of decouplers,
heads similar to those described evidence
crosstalk interference of about —45 db.
This refers to the ratio of crosstalk
induced from a fully modulated signal
in an adjacent track relative to a fully
modulated signal in the track in ques-
tion. While experiment has shown
that crosstalk values better than —50
db may be obtained by increased shield-
ing, particularly of such form as to com-
partment the tracks on both sides of
the film, this method presents threading

difficulties and, as previously statedy
values considerably better than —50 db
are necessary for general professional
use. Therefore, the decoupler method
has been developed and values of at
least —60 db are obtained at 1000
cycles.

While a value of approximately —60
db of crosstalk at 1000 cycles has been
obtained in this design, it will be noted
from reference to Curve 1 of Fig. 6 that
although the crosstalk stays appreciably
constant from 150 to 3000 cycles, it
rises to a value of about —45 db at
50 cycles and —48 db at 10,000 cycles.
The increase at the low frequencies has
been discussed above. The increase
at the high frequencies simply reflects
the closer coupling between adjacent
heads at the high-frequency end of the
spectrum. In Curve 2 of Fig. 6, the
40-db ear-weighting characteristic has
been added to the experimental data
used in Curve 1 and it is obvious that
with this correction the effective cross-
talk at low frequencies is well below
audibility. Many listening tests with
a wide variety of recorded material
confirmed the selection of 1000 cycles
as a suitable frequency for adjustment of
the decouplers and for the attainment
of —60 db at this frequency as a
guarantee against any audible crosstalk
in the recording audio spectrum.

The decouplers consist of several
small strips of Permalloy extending
diagonally from a point near the re-
cording gap of one head to a similar
point on the adjacent head. A small
air gap at either end is adjusted for
optimum operating conditions with the
help of a wave-analyzer, after which the
decouplers are locked in place.

The individual heads are separated
by a double thickness of magnetic
shielding material to reduce hum pickup
from external sources. The complete
head assembly is enclosed in a box of
heavy magnetic material, the front half
being hinged to allow access for thread-
ing.

110

February 1952 Journal of the SMPTE Vol. 58

Fig. 5. RA-1508-A triple magnetic head.

V

m — 4O

\

g

z

\

*

rf -50

\

H
l

s

s

^

/

X

5 -60
g

s CURVE 1

^,

^^^-— •

==

^*

x

^

/•

>*«

—

—

• •

V-
<

_J

UJ _

\

s

x

x*

c: -'0

\

x

/

\

s,

CUV

s

•»

.— ^

—

J

20 100 000

FREQUENCY IN CYCLES PER SECOND

Fig. 6. Crosstalk as a function of frequency.

Davis, Frayne and Templin: Multichannel Magnetic Recording 111

-HO

2 0
8

-30

20

10000 20000

FREQUENCY IN CYCLES PER SECOND
Fig. 7. Frequency-response characteristics of triple-track magnetic head.

To avoid abrupt changes in the track
and shield relationship, the edges of
the individual shields are especially
contoured. This minimizes cyclic
amplitude variations in the useful low-
frequency-response characteristic. These
may otherwise occur at scanning fre-
quencies whose half-wavelengths effec-
tively encounter abrupt changes in
magnetic conditions. In this same con-
nection, the departure from a normal
6-db/octave reproducing characteristic
lies below the useful frequency range
because of the generous film wrap and
physical size of the heads.6 Further-
more, these conditions promote long
head life.

The frequency-response charac-
teristic of a typical RA-1508-type head
is shown in Fig. 7.

Transmission Equipment

Transmission components and their
circuit arrangement have been estab-
lished after considerable discussion of
the general requirements for such equip-
ment with major-studio sound personnel.

The cabinet contains all transmission
components, including 115-v, a-c power
supplies, for operating directly from
three mixer outputs when used as a
recorder, and for operating into three
high-level mixer inputs or power ampli-
fiers when used as a re-recorder, re-
producer or playback unit. Since these
three transmission channels must be
used simultaneously, special care must
be taken to maintain the high degree of
interchannel isolation provided by the
magnetic heads.

When used as a recorder it operates
at a nominal signal input of —30 dbm
which permits considerable separation
from the mixers without line-noise
difficulties. It also furnishes direct
monitor from each channel at a level
of +16 dbm which is sufficient for
operation directly into a small monitor
speaker or through power amplifiers
to a larger horn system. If the mixer
operator will monitor from only one
chaririel at a time, leaving to the re-
cordist the responsibility for monitoring
all channels simultaneously from the

112

February 1952 Journal of the SMPTE VoL 58

Fig. 8. Plug-in arrangement of amplifiers.

film, he may switch the input of his
speaker system to any one of the three
+ 1 6-dbm direct-monitor circuits. How-
ever, if he wishes to monitor two or more
channels simultaneously, these may be
combined as desired from low-level
( — 30-dbm) monitor circuits which are
also provided. These are bridged from
the high-level circuits with sufficient
loss such that when combined they
will not detract from the 60-dbm isola-
tion figure provided by the heads. It
is expected that the mixer operator will
not monitor from the film because of
the fractional-second delay required for
the film to move between the record- and
monitor-head positions.

When used as a re-recorder or re-
producer, an output from the film of
+ 16 dbm on each channel is provided
through the same output circuit as is
used for the direct monitor during re-
cording.

The requirement that the three com-
plete recording-reproducing systems, in-
cluding power supply, be contained
within the cabinet calls for special con-
sideration of components and mounting
arrangements. Three amplifiers are

provided in each channel — one for
recording, one for reproducing and an
output amplifier used alternatively for
direct monitoring or reproducing. For
both the recording and reproducing
amplifiers the compact RA-1474, as
used in the latest Western Electric
portable magnetic recording systems,
is used.6 This amplifier uses two minia-
ture 12AY7 vacuum tubes with push-
pull output of +22 dbm for 1% dis-
tortion. A feature of the amplifier is
a plug-in unit which connects to two of
the internal high-impedance circuits and
provides for gain adjustment in the range
from 40 db to 70 db and equalization as
required for the particular application.
As used in the recording amplifier, the
plug-in unit reduces the gain to 48 db
and in addition provides a low-frequency
pre-equalizer which is used if this equali-
zation has not already been inserted in
the mixer circuits preceding this equip-
ment. This pre-equalizer has approxi-
mately 4 db of boost at 60 cycle/sec and
is complementary to the low-frequency
shelf in the reproducing equalizer. In
combination, the low-frequency pre-
and post-equalizers provide a flat charac-

Davis, Frayne and Templin: Multichannel Magnetic Recording

113

! I

114

February 1952 Journal of the SMPTE Vol. 58

teristic at low frequencies and reduce
the effect of power-line interference on
the overall signal-to-noise ratio.

In the reproducing-amplifier applica-
tion, the plug-in unit provides the re-
quired 6-db/octave characteristic with
the low-frequency shelf as described
above, plus a high-frequency shelf
compensating for scanning and demagne-
tization losses.

A new amplifier has been designed to
meet the requirement for the direct-
monitor and reproducing output ampli-
fier where less gain and greater power
output are required than are provided
in the RA-1474 described above. As
used in this equipment it has a gain of
24 db and an output of +24 dbm with
1% total harmonic distortion. It is
also expected that it will have general
application as a 600-ohm-line amplifier
and a zero-gain bridging amplifier. It
is a single-stage push-pull unit using
one miniature 12AU7 vacuum tube.

Both types of amplifiers have been
equipped with new-type miniature plugs.
This permits ready removal or replace-
ment of any amplifier for maintenance
or test at a bench position. Separate
plugs at opposite ends of the amplifier
provide for optimum segregation of low-
and high-level circuits throughout the
equipment, thus minimizing noise and
crosstalk interference.

Figure 8 shows both types of amplifiers
in their mountings. Also shown are
the slotted guides in the mounting panel
which insure registration of the plugs
and receptacles. The complete mount-
ing panel for the group of amplifiers is
floated, thus making unnecessary the
separate isolation of each unit.

The high-frequency bias for the three
channels is supplied from one 60-kc-bias
oscillator. This eliminates the possi-
bility of audiofrequency beating which
could occur with the interaction of three
separate oscillators operating at slightly
different frequencies. A distribution
network from the oscillator output to
the three recording-head circuits pro-

vides 70 db or more isolation below
6000 cycle/sec for the audiofrequency
signals of the three channels which
would otherwise be coupled together by
the common oscillator supply.

The amplifiers are powered from the
RA-1479-type power units as used in
the previously described portable re-
cording system.6 One power unit
handles the six record-reproduce ampli-
fiers and the other handles the three
output amplifiers. The bias oscillator
contains its own separate power supply.

A simplified block-schematic and level
diagram of the complete system are
shown in Fig. 9. For recording applica-
tions, the nominal — 30-dbm signal level
is received from the mixer and the
recording attenuator is adjusted to
establish 3% total distortion from the
reproduced film for 100% modulation.
This normally is obtained with a level
of +10 dbm at the recording amplifier
output at which point the volume
indicator and direct-monitor circuit
are bridged. A series network in the
head circuit forms the load for the
recording amplifier. It also provides
high-frequency pre-equalization in five
1-db steps so that with the fixed equali-
zation in the reproducing amplifier a
flat overall response is obtained. A
60-kc suppression filter prevents the
bias signal from feeding back to the
volume-indicator and direct-monitor cir-
cuits.

The direct-monitor circuit also con-
tains a low-frequency postequalizer com-
pensating for that equalization intro-
duced earlier in the recording circuit
or in the mixer. Under recording condi-
tions it operates into the 24-db-gain
output amplifier to provide a flat overall
response to the mixer at a 1 00% modula-
tion level of +16 dbm. Switching
arrangements in the volume-indicator
circuit permit the meter to be used
alternatively for checking levels at the
recording-amplifier output, the monitor-
reproduce-amplifier output under either

Davis, Frayne and Templin: Multichannel Magnetic Recording

115

Fig. 10. Control panel.

recording or reproducing conditions and
also for measurement of bias current.

The recordist monitors by headset
from the output of the three reproducing
amplifiers. Normally, if all three chan-
nels are being used, he will listen to
them simultaneously. In the event any
trouble occurs, or for any other reason,
he can listen to each one individually
or to any combination by operation of
the separate cutoff switches. Since
these switches are in only the reproduc-
ing circuit, their operation will in no
way affect the recording. He can also
compare the reproduced output from
any one channel with the corresponding
direct-monitor signal by operation of
his monitor-selection switch.

Under reproducing conditions the
24-db-gain output amplifiers are switched
to the output of the reproducing ampli-
fiers to provide the +16-dbm output
level and the volume indicators are
connected across these output circuits.

To prevent the possibility of in-
advertent application of the bias- and
recording-circuit signal to the film under
reproducing conditions, a separate rec-
ord-reproduce switch operating on

all three channels is provided. In the
recording position the record and bias
circuits are connected through to the
head and a red warning light appears
on the front panel. In the reproduce
position all three record heads are
shorted, thus preventing the application
of either bias or audio signals.

The principal recording-reproducing
operating controls are contained on the
front panel covering the upper equip-
ment space. As shown in Fig. 10, the
controls for the three channels are
identical and are grouped separately.
These controls include, for each channel,
Record Attentuator, High-Frequency
Equalization, Bias Current, Volume-
Indicator Meter and Meter-Circuit Se-
lection. Other controls, including those
associated with headset monitor and the
drive motor, are mounted on either
side of the center section of the cabinet.

Performance

The overall performance of the equip-
ment amply meets the requirements for
recording and reproducing equipment
in major studios. Overall crosstalk
isolation between channels, including

116

February 1952 Journal of the SMPTE Vol.58

heads and all associated circuits, has
been maintained at approximately 60
db through the critical middle range of
the frequency spectrum.

Based on the allowable 3% total
distortion for the complete recorder-
reproducer system, the overall signal-
to-noise ratio is maintained at 55 db to
58 db, unweighted. The overall fre-
quency-response characteristic is essen-
tially flat over the frequency range from
50 to 8,000 cycles.

Demon stration

A demonstration film has been pre-
pared to show some of the operating
characteristics and intended usage of
this recorder. In the first part of the
film three separate recordings of organ
music, boys' choir and dialogue, re-
spectively, are laid down on the three
tracks. By switching outputs from the
three heads to a single reproducing horn
system, the high quality of each record-
ing as well as the lack of interference
among associated tracks may be ob-
served. To emphasize further the low-
level crosstalk conditions, the center
track later in the reel carries no modula-
tion, but the side tracks are heavily
modulated. Switching from either of
these tracks to the center one reveals no
audible evidence of crosstalk.

Conclusion

The machine described in this paper
permits the recording of three high-
quality magnetic tracks on 35mm film,
each one being comparable in quality
to that of a single track on 35mm. The
residual crosstalk value of —60 db
gives, in effect, complete isolation of
each track from the adjacent ones,
thereby permitting the recording of
completely unrelated material on any
track. The success of the first units of
this machine under studio operating
condition presages their wide adoption
in the near future for scoring, re-
recording and film-storage purposes.

Acknowledgments

The authors wish to express their
thanks to the other members of the
Westrex engineering staff who have
contributed to the success of this de-
velopment. We wish to express thanks
particularly to G. R. Crane for the
mechanical design of the double fly-
wheel drive, to A. L. Holcomb for the
"packaging" of the electronic com-
ponents, to H. R. Roglin for the testing
and alignment of the magnetic heads
and to P. F. Thomas for his painstaking
testing of the first model of this recorder
shown at the Society's Convention at
Hollywood, Calif.

References

1. W. C. Miller, and G. R. Crane,
"Modern film re-recording equipment,"
Jour. SMPE, 51: 399-417, Oct. 1948.

2. C. C. Davis, "An improved film-drive
filter mechanism," Jour. SMPE, 46:
454-464, June 1946.

3. W. J. Albersheim, and D. MacKenzie,
"Analysis of sound-film drives," Jour.
SMPE, 37: 452-479, Nov. 1941.

4. J. G. Frayne, and H. Wolfe, "Magnetic
recording in motion picture techniques,"
Jour. SMPE, 53: 217-235, Sept. 1949.

5. S. J. Begun, Magnetic Recording^ Murray
Hill Books, New York, 1949.

6. G. R. Crane, J. G. Frayne, and E. W.
Templin, "A professional magnetic
recording system for use with 35-,
\1\- and 16-mm films," Jour. SMPE,
56: 295-309, Mar. 1951.

Discussion

John G. Frayne: Before running the
demonstration films, which are recorded
in such a manner as to illustrate the
effective reduction in crosstalk between
tracks, I would like to express the thanks
of the authors to the engineering and
laboratory staff at Westrex Corporation,
Hollywood Division, for their invaluable
cooperation and assistance in this project.
The credit for the invention of the de-
couplers goes to my colleague, C. C.
Davis, and a U.S. patent application has
been made in his name for this invention.
The Westrex Corporation is pleased to

Davis, Frayne and Templin: Multichannel Magnetic Recording 117

make this information public since it
will undoubtedly stimulate constructive
thinking on the part of others and thus
will eventually aid in the improvement of
the magnetic-recording art.

John P. Livadary: In the Columbia
Pictures Sound Department we have been
using multitrack magnetic films since
November 1950 and we have accumulated
a lot of experience concerning the method
of compensating for losses in the magnetic
re-recording link used in our dubbing
operations.

To make it possible to reproduce the
same film satisfactorily on any three-track
channel, we found it necessary to standard-
ize on the equalization of the reproducing
circuits. This, in turn, has resulted in
dividing the overall compensation for
magnetic losses in two, and in equalizing
for part of these losses in the recording
and part in the reproducing circuits.

Consistent with this thought, we de-
veloped our own standard magnetic
frequency film which we use to calibrate
the reproducing circuits, the recording
circuits being adjusted to achieve a one-
to-one overall transfer which is desirable
for re-recording purposes.

We have also been using electronic feed-
back means for the decoupling of crosstalk
between adjacent magnetic heads with
reasonable satisfaction since November
1950.

L. L. Ryder: One further contribution is
that possibly, if the heads are moved a
slight distance further away from the drums
that may exist on some of the machines,
you can still retain the high quality of
movement which is desired and get away
from a large part of the head wear. We
have replaced one or two heads in our
work since the advent of magnetic re-
cording. I have in operation at Ryder
Services a head which has been operating
every day for about two years. It is
still not worn out. The angle of approach
of the film to the head and the angle of
recession of the film may contribute quite
a bit to the head wear.

C. E. Hittle: Since our good friend Dr.
Frayne of Westrex has volunteered to
provide us with information relative to
the design of their equipment, as long as
their design is covered by patent or patent
applications, I have a question to ask
relative to the design of their drum as-
sembly, particularly pertaining to the
design of the flywheels which they are
using on their twin-drum system. Are
they of equal mass, weight and size?

Dr. Frayne: Yes, they're identical as far
as we know. The combined moment of
inertia of the two flywheels is practically
the same as the moment of inertia of the
single flywheel on the RA1251 re-recorder.
That was done so we could retain the same
filter components.

M. Rettinger: I would like to ask Mr.
Livadary if his electronic decoupling
circuit provides crosstalk reduction equal
to what was demonstrated this afternoon?

Mr. Livadary: About nine months ago
we gave a demonstration, similar to the
one given today by Westrex, in which we
ran three tracks and cut off each track in
turn to demonstrate the amount of leakage
which existed. Our measured values of
crosstalk suppression at 400 cycles were
of the order of about 60 db to 62 db
between any two adjacent tracks. At
higher frequencies this figure was slightly
lower. However, according to our ex-
perience to date, having dubbed about
1,000,000 ft of released footage on multi-
track magnetic film, we haven't had a
single crosstalk problem to cope with, and
our decoupling methods have been satis-
factory for our work.

Dr. Frayne: When John called me up
and told me about this I asked him how
it worked. He said that it was mounted
in a little black box and that he could not
divulge the details.

Mr. Livadary: I regret to reply to Dr.
Frayne that this particular method was
indeed in a black box at that time and I
was more or less on a spot because we were
in the process of securing patents which
made it difficult to discuss this matter.

118

February 1952 Journal of the SMPTE Vol. 58

Magnetic Sound Track Placement

By LOREN L. RYDER and BRUCE H. DENNEY

This paper sets forth technical data indicating that in magnetic recording on
35mm film, high sprocket-hole modulation is encountered in the area between
50 and 100 mils from the sprocket holes. The paper suggests a change in
the proposed ASA standard for 35mm sound track placement.

_L HIS PAPER is presented after a careful
consideration of the present proposed
ASA standard for 17^mm and 35mm
magnetic sound track placement, Fig. 1 .
In the opinion of the writers there are
two basic reasons why this proposal
should not be accepted.

1. Recent tests show a very high per-
centage of sprocket-hole modulation in
the sound track area close to the sprocket
holes.

2. The present proposal was prepared
prior to active editing of magnetic film
and does not adequately meet editorial
requirements.

Sprocket-Hole Modulation

The sprocket-hole modulation under
consideration is a 96-cycle modulation
of the sound signal. In magnetic re-
cording and reproducing this effect
is largely an amplitude modulation. It
takes place in both recording and re-
production and usually is additive.
It is the result of a varying contact
and/or lack of contact of the magnetic

Presented on October 18, 1951, at the
Society's Convention at Hollywood, Calif.,
by Loren L. Ryder and Bruce H. Denney,
Paramount Pictures Corporation, 5451
Marathon St., Hollywood 38, Calif.

head with the magnetic coating of the
film. There are two primary causes
for this variation in contact. During
punching of the sprocket holes there is
a deformation of the film in the vicinity
of the sprocket holes that prevents
uniform contact. In winding the film
around a drum there is a polygonal
effect due to the weakening of the film
at the point of punching.

Tests made at Paramount indicate
that experts are conscious of about 2%
sprocket-hole modulation, that almost
anyone will observe 5% and that the
distortion becomes quite obvious at
8% to 10%.

A series of tests was made with record-
reproduce head widths of 250 mils,
200 mils, 150 mils, 50 mils and a 50-mil
head with a land so as to simulate con-
tact of a 2 50-mil head. Each of these
heads was tested at several distances
from the sprocket holes. All of these
tests were made with full-coated 3-M
35mm magnetic film on a Westrex RA-
1231 recorder modified for magnetic
recording-reproduction. This is a single
drum recorder with the head in the drum.
Comparable results were obtained both
for the condition of compliant head and
fixed head.

February 1952 Journal of the SMPTE Vol. 58

119

t <k

THACK NO. I T*ACK NO.2 .

TRACK MO.3

*CCOMD.~w „„.

REPRODUCING. I }— .2001 -W* .(
HEADS IN LINE

a
a
a
a

-.189

a
a
a

.1 50 --.

! 50

35M.M.

Fig. 1. Proposed American Standard for Magnetic Sound Track Placement.
(See Fig. 5A for new recommendation.)

120

,° -050 -100 -'^0 .200 .250 300 350 .400 IK

OrSTANCC FROM INSIDE OF' SPROCKET HOLES

Fig. 2. Record-reproduce 0.050-in. head with 0.200-in. land.
February 1952 Journal of the SMPTE Vol. 58

Figure 2 is a plot of the distance of the
record-reproduce head from the inside
of the sprocket holes vs. the per cent
sprocket-hole modulation, using the
50-mil head with a land so as to simulate
the contact of a 250-mil head. In this
case the slit is toward the sprocket holes
and the land is toward the center of the
film. On the chart the placement of
the slit is shown as a horizontal straight
line and the vertical position of the line
indicates the per cent modulation for
each slit position. A solid line is drawn
through the center of the respective slits
and a dashed line to the left of the solid
line crosses the slits at the point where
the per cent sprocket-hole modulation
for a small increment of the slit length
equals the per cent modulation for the
entire slit. The dashed line, therefore,
indicates the per cent sprocket-hole
modulation actually existent at any
distance from the inside of the sprocket
holes.

As indicated by the dashed line, the
sprocket-hole modulation is 5% at
130 mils from the inside edge of the
sprocket holes, 10% at 100 mils, 18%
at 75 mils and 32% at 50 mils. In
other words, that portion of the film
between 1 00 mils and 50 mils contributes
10% to 32% sprocket-hole modulation.
This area is within the scanned area of
the proposed ASA standard.

Figure 3 is a plot for a 250-mil record-
reproduce head. This shows how the
bad sprocket-hole modulation close to
the sprocket holes is masked and sub-
dued by the good reproduction far
removed from the sprockets. If a
250-mil sound track width is used,
starting 50 mils from the sprocket holes,
the sprocket-hole modulation will be
3.5% or 4% and the quality will be
impaired but marginal. If the 250-mil
sound track starts 100 mils from the
sprocket holes, the sprocket-hole modula-
tion will be approximately 0.5%.

Figure 4 shows the per cent sprocket-
hole modulation for a 50-mil (A), a
1 50-mil (B), a 200-mil (C) and 250-mil

(D) slit, each starting 100 mils from the
sprocket holes. Similar information is
shown at E, F, G and H for slits starting
50 mils from the sprocket holes. Line
G represents the present proposed ASA
standard (a 200-mil record-reproduce
head starting 50 mils from the sprocket
holes) which averages 5.6% sprocket-
hole modulation. Previous measure-
ments made by others and submitted
to the Motion Picture Research Council
indicate a sprocket-hole modulation of
4.5% for this condition. •

Line B shows that a 1 50-mil head
starting at 100 mils from the sprocket
holes and extending to the same inside
line as the proposed ASA standard
(right-hand end of both lines G and B),
will reduce the sprocket-hole modula-
tion from 5.6% to 2%. The loss in
level is approximately 2.5 db.

For both 35mm and 17^mm recording
the writers recommend the sound track
placement shown in Fig. 5A. For-
tunately, this placement also meets
the editorial requirements which are
set forth in the second part of this
memorandum. It is further recom-
mended that heads be aligned so that
the end of the slit shall be 131 mils from
the inside edge of the sprocket holes
and that this condition shall prevail
regardless of the length of the slit in the
head. Referring to Fig. 2, this 131
mils is the point at which the incre-
mental measurement is 5% and any
further encroachment toward the
sprocket holes is undesirable.

In this recommendation the sound
track is 200 mils wide; however, any
width head can be used and the re-
cordings played on any width head as
long as the alignment is as outlined
above. Paramount has many 250-mil
heads which will remain in service.

It is to be noted that recordings made
on the ASA proposed standard will re-
produce better under the conditions of
this recommended procedure than on
the ASA proposed standard. Further,
recordings made on the basis of this

Ryder and Denney: Magnetic Track Placement

121

\

.050 100 .150 200 .250 300 J50 .4001*

DISTANCE rH.OM INSIDE OF SPROCKET HOLtS

Fig. 3. Record-reproduce 0.250-in. head.

0 O&O 100 .ISO 200 250 300 350 400 IN.

DISTANCE FROM INSIDE OF SPROCKET HOLES

Fig. 4. Comparative sprocket-hole modulation — slits starting 50 mils
vs. slits starting 100 mils from sprocket holes.

700 mils

A, 50-mil head

B, ISO-mil head

C, 200-mil head

D, 250-mil head

50 milt

E, 50-mil head

F, 1 50-mil head

G, 200-mil head
H, 250-mil head

122

February 1952 Journal of the SMPTE Vol. 58

REPRODUCE HEAD

COATING UP

Fig. 5A. Recommended sound track placement.

RECORD OR
REPRODUCE HEAD

COATING UP

a
a

35MM

(..378-)-

Fig. 5B. Alternate sound track placement for 150-mil head.

recommended procedure will reproduce
better on the ASA proposed standard
than a recording made and reproduced
on the ASA proposed standard. In
other words, whenever the work and/or
equipment used is intermingled between
old and new, the result is always im-
provement and never degradation.

If existing equipment is to be modified
and if the 200-mil head cannot be moved
to the position specified in Fig. 5A, most
of the improvement can be gained by
using a 150-mil head as shown in
Fig. 5B.

It is expected that all three-track
recordings will be done on full-coated
magnetic film, although the future may
show a preference for striped film
having a clear area between each stripe.

For three-track recording the writers
recommend the sound track placement
shown in Fig. 6A. If greater track-to-
track isolation is desired, the placement
(Fig. 6B) can be used. Either of these
proposals will have less sprocket-hole
modulation and less distortion than the
ASA proposed standard.

As indicated earlier in this memoran-
dum, the tests were made with full-
coated magnetic film on equipment
having the head in the drum. This is
the basis on which the original standardi-
zation was contemplated and is the con-
dition of most magnetic recording.
Tests which have been made show a
slight preference in favor of magnetic
film that is not coated in the sprocket-
hole area and also there seems to be some

Ryder and Denney: Magnetic Track Placement

123

RECORD OR— «*• T

REPRODUCE HEADS.==

Fig. 6A. Three-track recommendation.

RECORD OR -*-.
REPRODUCE HEADS

COATING. UP

Fig. 6B. Alternate for three tracks.

improvement with recorders of the two-
flywheel type. However, as most of
the equipment and most of the magnetic
film used are the type used in this test,
the standard should meet these re-
quirements. Listening tests made sub-
sequent to the above measurements
indicate that under certain conditions
of single-drum handling of 17^mm
magnetic film, sprocket-hole modulation
is greater than with 35mm magnetic
film. As far as the writers know, no
measurement studies have been made
of this effect, even though over half of
the production recording is on 17^mm
magnetic film.

The tests reported in this paper were
made by recording a 3072-cycle fre-
quency on the film and observing the
ratio of peak amplitude of 96-cycle to

3072-cycle reproduction on an oscillo-
scope. Previous measurements made
with the harmonic wave analyzer also
included film irregularities and were
found to be misleading at these relatively
low percentages. Inter modulation
analyzers include the other film irregu-
larities in their measurement and are,
therefore, not indicative of 96-cycle,
sprocket-hole modulation. The per-
centage of 96-cycle modulation is almost
independent of signal amplitude.

In general, the sprocket-hole modula-
tion described above has gone un-
noticed and has caused little trouble in
recording. This is because most com-
panies are still working with only first
copy transfers or intermediate film
procedures. These modulation effects
will become more obvious when film

124

February 1952 Journal of the SMPTE Vol. 58

RECORD OR

COATIN& UP

169

. I IBS

4- — 225
|*-COATI*

Fig. 7A. 35mm magnastripe.

.020-H h-

CO AT IN 4

RECORD OR —

REPRODUCE HEAD

FI89
-

,11854- — -225 — H

I-.— COATING — -J

-689

COATING UP

Fig. 7B. 17fmm magnastripe.

losses are eliminated from the procedure
and when the number of generations is
increased. The reason why many
people prefer ^-in. tape as compared
to sprocket-driven magnetic film is
because of sprocket-hole modulation.
We should not accept a standard that
limits future development, especially
when a better standard is available
without increased cost and without
damage to the recordings already made.

Editorial Handling

Sprocket-driven magnetic film may be
assembled and handled in simple editing
much the same as -J-in. magnetic tape.
However, most motion picture editing
involves so much overlapping, modula-
tion matching and selecting of the best

place to cut that a more positive system
seems desirable.

Experience at the Paramount West
Coast Studio and Ryder Services in-
dicates that some form of visual modula-
tion is essential to convenient cutting
of magnetic film. The initial work
with modulation writing* on the mag-
netic coating was a step in the right
direction; however, it required front
viewing of the modulation writing,
whereas editors are equipped for and
are in the habit of viewing both picture
and sound by transparency.

After checking all combinations of
sound track placement and every known

* L. L. Ryder, "Motion picture studio
use of magnetic recording," Jour. SMPTE,
55: 605-612, Dec. 1950.

Ryder and Denney: Magnetic Track Placement

125

Fig. 8. Magnastripe film placed on
top of picture film for synchronous
handling.

form of modulation writing, Paramount
and Ryder Services have selected the
striped magnetic film shown in Fig. 7A
for 35mm editorial work. This has a
stripe of magnetic coating 225 mils wide
with center line 420 mils from the
edge of the film. A 20-mil stripe is
placed near the sprocket holes on the
side of the film opposite the sound track
so as to balance the roll for winding.
Figure 7B shows a coating of magnetic
film for 17^mm use. In manufacture
two stripes of magnetic coating are placed
on 35mm film and later slit to give two
17£mm films.

For both 35mm and 17^mm editing,
the modulation writing is placed on
the film in the clear area between the
striping and the sprocket holes. An
illustration of the 35mm film along with
picture is shown in Fig. 8.

By reviewing Fig. 8 it will be noted
that:

1. When the picture and sound films
are placed on top of each other, the
modulation can be seen by transparency
through the clear area of the picture.
This procedure is a common editorial
practice with optical sound film.

2. The picture can be seen by trans-
parency through the clear area of the
sound film. The area available for
viewing is the same as under the present
practice with 200-mil, push-pull optical
recording.

3. The code numbering used for
synchronization can be viewed in trans-
parency and matched.

4. The picture and sound film still
held together can be run through a
picture-only or sound-only magnetic
Moviola unit.

5. Markings can be made on the film
in crayon or ink and read in transparency
as at present. Crayon markings should
not be made in the sound track area.

6. These films can be handled in
regular existing editorial equipment
including synchronous rewinds. The
only conversion necessary is the magnetic
reproducer on the Moviola and a mag-
netic reproducer or conversion for the
review room.

7. These films can be handled by the
existing editorial techniques which have

126

February 1952 Journal of the SMPTE Vol. 58

been evolved after many years' practice
and experience.

8. The magnetic cutting print is also
used as the dubbing print.

9. The modulation writing is on the
base side of the film and can be removed
with carbon tetrachloride.

10. The film can be erased, cleaned
and re-used many times.

11. The magnetic sound track place-
ment leaves the photographic sound
area clear. This makes it possible to
intercut magnetic and photographic
sound films.

12. It is also possible to stripe photo-
sensitive film either before or after
processing. Under this proposal it will,
therefore, be possible to have both photo-
graphic and magnetic sound on the same
piece of film. This may be of great
value in newsreel work, narration re-
cording and editorial processes, includ-
ing scoring and dubbing.

13. For most reproducers these films
can be spliced on hot-lap splicers if the
blades are demagnetized.

Many combinations of sound track
and modulation writing position have
been tried and abandoned. There may
be a slight advantage in favor of having
the sound track in the so-called positive
position instead of the so-called negative
position. This would involve so many
changes in equipment that it is not
recommended. In these considerations
we have also reviewed the question of
under- vs. over-scanning of striped
magnetic film. Practice to date indi-
cates that there is no preference.

General

Fortunately the best-known specifica-
tion for editing is the correct speci-
fication in regard to sprocket modulation.

Paramount West Coast Studio and
Ryder Services, along with a goodly

number of the recording companies,
have from the inception of magnetic
recording used a sound track placement
such that the center line of the sound
track is halfway between the inside of
the sprocket holes and the center of the
35mm film. We are abandoning these
specifications in favor of the new sug-
gested procedure. We hope that others
in the industry will take advantage of
our work. We have no hesitation in
taking this step because, as stated earlier,
films can be interchanged and any film
that is either recorded or reproduced in
accordance with this suggestion will
play better than a film that is both
recorded and reproduced under condi-
tions of the ASA proposed standard.

We urge that the Society of Motion
Picture and Television Engineers and
the Motion Picture Research Council
reject the present proposed ASA standard
and along with other possibilities con-
sider the suggested procedure set forth
above.

Discussion

/. L. Pettus: It appears that the ex-
periences by us at RCA do not quite
agree with Mr. Ryder's. In fact, some
of our data might take exception by as
much as ten to one. With your permis-
sion, I would like to illustrate a point
or two by the use of a few slides. These
slides consist of measurements of 96-
cycle flutter as well as amplitude modula-
tion and I would like to present them in
view of the statement Mr. Ryder made
that his method of evaluation was (1)
by listening and (2) by measuring, and
in view of the method by which he
measured. Possibly we take exception
to the method of measuring.

At the top of Slide 1 is an oscillogram
of 96-cycle flutter measured from a

Ryder and Denney: Magnetic Track Placement

127

05% 96/v ,075 DISPLACEMENT

05% 96,w .000 DISPLACEMENT

;05% 96/v .IOOOI8PI.ACEMENT

:05% 96^ .025 DISPLACEMENT

05% 96-^ .050 DISPLACEMENT

SUde 1.

Slide 2.

:05% 96/v; .150 DISPLACEMENT

: LOCATION FROM SPROCKET MOLES

SUde 3.

SUde 4.

3000-cycle tone recording where the
magnetic track was laid down at a 0
displacement from the sprocket hole.
In other words, the edge of the track
was directly adjacent to the perforations.
The second oscillogram was made with
the head moved over 25 mils. You will
note in the first, that the average value
of 96-cycle flutter is somewhat greater
than 0.05% and when the head is moved
over 25 mils, the average value drops
a little less than 0.05%. In the third
oscillogram, the head was moved over
50 mils and the same amount of 96
cycles seems to prevail. Slides 2 and 3
show a continuation of flutter measure-
ments where the head was moved in
increments of 25 mils and measurements
were taken at 50-, 75-, 125- and 150-mil

displacement. Again, the amount of
96-cycle flutter shows little or no change
over that shown in the second oscillo-
gram where the head is just removed
from the perforations. Slide 4 deals
with the measurement of amplitude
modulation as indicated on an inter-
modulation set reading only the ampli-
tude variations. Here we notice that
the top of our curve shows a value of
6% modulation of the carrier and may
be compared to that in Mr. Ryder's
illustration which, I believe was ap-
proximately 60%. Our measurement
at this point was taken when the track
had a 0 displacement from the per-
forations. Now, if we move over 25
mils, we find 3.5% modulation followed
by 2% at the 50-mil displacement and

128

February 1952 Journal of the SMPTE Vol. 58

1% at the 100-mil displacement. It is
seen that the curve flattens out rather
quickly. If the signal is raised in fre-
quency to, say, 7000 cycles, this entire
curve moves slightly toward the per-
foration base line and also increases its
height. Judging from these data, 'we
see no real gain in changing the proposed
standards, but instead, we see several
complications arising in the use of triple
tracks and at the moment I have not
seen Mr. Ryder's proposal on how he
would arrange three tracks along the
35-mm film in satisfactory manner.

Dr. J. G. Frayne: I think Mr. Ryder's
figure of 60% was based on a 50-mil
track rather than a 200-mil track that
Mr. Pettus's figures were based on.
Is that right?

Mr. Pettus: That is correct. Those
tests were made on standard 200-mil
track.

L. L. Ryder: Please refer to the Slide
4 which was presented by Mr. Pettus.
The graph line indicates the percentage
of sprocket-hole modulation for different
positions of head placement with respect
to the sprocket holes. At a position of
the head such that the end of the slit is
50 mils from the sprocket holes, the
sprocket-hole modulation is shown as
2%. Now, please refer to Fig. 3 in
my paper. This happens to be for a
250-mil head and the reading of sprocket-
hole modulation is approximately 3.5%.
The data, therefore, are not out of
agreement by a ratio of 10 to 1, as
indicated by Mr. Pettus, but by a ratio
of 3.5 to 2, which is 1.7 to 1.

Now, referring to Fig. 2 in my paper,
it is to be noted that at 50 mils from the
sprocket holes the incremental sprocket-
hole modulation is 32%. If this is to be
reduced by the factor of 3.5 to 2 to con-
form with the RCA data, the amount of
sprocket-hole modulation introduced
would be 18%. It is my feeling that
we should not introduce 18% sprocket-
hole modulation in order to meet a
proposed standard.

The validity of our measurements
has been questioned. We believe our
measurements to be correct, but in any
case our measurements have been re-
lated to what can be heard, and what
we can hear is the thing about which I
am most concerned. Our first observa-
tions of this phenomena were the result
of listening tests made with the proposed
ASA standard. Both the RCA data and
the data prepared by the writer indicate
that a change should be made.

F. R. Wilson, Vice-chairman of the
Session, read a communication from
L. D. Grignon, Twentieth Century-Fox
Film Corp., reporting data from an
investigation of 96-cycle modulation
made recently on regular production
equipment:

"Recordings were made on a Westrex
RA-1231 recorder modified for magnetic
recording, using a 250-mil track with
the nearest edge 50 mils from the
sprocket hole. This recorder uses a
compliant mounted head adjusted to
90-g pressure and a special recording
drum which supports as much of the
film as possible. A signal frequency of
approximately 3000 cycles was used at
a level which produces 1% harmonic
distortion. During reproduction the
output was observed on an oscilloscope
by the same method as reported by
Ryder and Denney, the exact signal and
sweep frequencies being adjusted to give
the most readable traces. Since the
peak-to-peak 96-cycle modulation is of
the order of 3%, the reading error was
considerable due principally to random
amplitude fluctuations and noise; there-
fore, readings of total peak modulation
distortion products were made by the
use of an Altec TI 402 intermodulation
analyzer. Two film stocks were used
with the results shown in Table I.
When the 96-cycle modulation is less
than 2.5%, the oscilloscope reading
accuracy becomes seriously questionable
and, therefore, in some instances data
are recorded only for the intermodula-
tion analyzer measurement.

Ryder and Denney: Magnetic Track Placement

129

Table I.

Recorder

Oscillo-
scope %
96-cycle
peak
modu-
Reproducer lation

Intermod.
analyzer
% total
(peak)

Notes

RA

1231

RA

1231 3.3

4.

0

Roll 9601

Old film

RA

1231

RA

1251 2.5

3.

5

Roll

9601

Old film

RA

1231

RA

1231

1.

2

Roll

1336

New film

RA

1231

RA

1251

1.

2

Roll

1336

New film

RA

1251 (3-track)

RA

1251 (3-track)

1.

2

Roll

1336

Track 1

(outside)

RA

1251 (3-track)

RA

1251 (3-track)

1.

7

Roll

1336

Track 2

(center)

RA

1251 (3-track)

RA

1251 (3-track)

1.

0

Roll

1336

Track 3

(inside)

Table II. Recorded on RA 1231 and Reproduced on RA 1251.

Approx.

Oscillo-
scope %

no. of

96-cycle

Intermod.

Magnetic
film roll

Date first

times
used on

peak
modu-

analyzer
% total

no.

used

prod.

lation

(peak)

Remarks

9596

12-31-49

10

2.0

1.9

Many random variations

9915

2-2-50

12

1.80

1.8

0607

4-1-50

7

2.60

2.6

1415

5-19-50

9

2.10

2.0

Many random variations

1563

5-22-50

9

1.75

2.2

755

10-17-50

5

2.20

1.9

6650

11-29-50

6

1.75

1.7

925

3-8-51

5

1.25

1.4

1010

3-26-51

3

1.5

1.5

1298

9-7-51

1

2.0

1.7

"The considerable differences between
the two stocks prompted another series
of measurements of the same kind on a
variety of stocks. These results are
shown in Table II with pertinent his-
torical data concerning each roll. From
Table II it may be concluded that:
(1) with the recording and reproducing
equipment used at the subject studio
and the magnetic film stocks currently
in use, the maximum total amplitude
distortion products do not exceed 2.6%
(Note — Roll 9601 of Table I is used
only for preliminary maintenance tests);

(2) there has been some improvement
during the past 1^ years, due either to
reduced random amplitude irregularities
or improved perforations; and (3)
there is little correlation between usage
and amplitude distortion products.

"It can be expected that the amplitude
distortion products due to 96-cycle
modulation will increase in some fashion
when multiple generation recordings are
made from a given piece of material, but
this is also true of all other amplitude
irregularities, noise and flutter. The
number of good-quality generation rec-
ords which can be made is determined

130

February 1952 Journal of the SMPTE Vol. 58

by all these factors, not by 96-cycle
modulation alone.

"Considering the data and the par-
ticular equipment in use at Twentieth
Century-Fox, it appears that a change
in track position to some location other
than the standard now proposed by the
Society would be of very questionable
merit. It would seem that the greatest
benefit can be obtained by film improve-
ment, particularly with respect to uni-
formity of high-quality perforation and
low-valued random amplitude irregu-
larities."

Mr. Ryder: As an operator and as a
director of sound activity in the making
of motion pictures, it is not my good work
that causes me trouble but my bad
work. I am, of course, very concerned
about test data. I do want the data to
be correct and accurate, but with
respect to what I put in my plant, and
I should think this would apply to
others, I want first of all that it cause
me no trouble. The Fox data show
3.5% to 4% sprocket-hole modulation
on old film. These data correspond
almost exactly to the data shown in
Fig. 3 in my paper, which in turn
indicates to the writer that an incre-
mental measurement on the Fox equip-
ment would correspond to Fig. 2. I
should, therefore, expect Fox to be able
to hear the same sprocket-hole modula-
tion that we are able to hear at Para-
mount.

It is my belief that all users of magnetic
recording contemplate the re-use of film,
which means the use of old magnetic
film. In the taking of our data we found
variations and inconsistencies between
batches of magnetic film. The sprocket-
hole modulation of old films is generally
higher than that of new film. These
are the films that can cause trouble and
this is the trouble that can be avoided
by changing the proposed standard.

Dr. Frayne: It is interesting that
information presented by Mr. Pettus
was made on films using a double-
flywheel-type drive, whereas Mr. Grig-

CEIMTER OF BEAM
FROM EDGE OF TRACK

2.5 MILS

7.5 MILS

17.5 MILS

22.5 MILS

27.5 MILS

Slide 5.

non's information was obtained on a
recorder using the single-drum-type
drive with the head located at the drum.
So we have two different philosophies
of film drive and the results are more
or less comparable. Therefore, what is
wrong with Mr. Ryder's data? First
of all, 96-cycle sprocket-hole modulation
is nothing new and is not confined to
magnetic film. I wrote a paper on this
subject in 1935 and if we may have the
first slide (Slide 5 in this printed version)
I can show you some of the things we
found on a typical negative photographic
sound track. This was done not with
a 50-mil head, but rather with a 5-mil
head. At 2.5 mils from the sprocket
holes (in other words, the center of the
5-mil head was 2.5 mils from the edges
of the sprocket holes), we got severe
amplitude distortion on a 3000-cycle
track. At 7.5 mils the amplitude
modulation was a little better, at 12.5
mils a little better yet, and at 27.5 mils
it had just about disappeared. I'm not
claiming that the amount of sprocket-
hole modulation you get in photographic
is identical in amplitude to what you
get on magnetic due to the greater
depth of focus on photographic, but on
the other hand the trend is there showing

Ryder and Denney: Magnetic Track Placement

131

NEGATIVE

PRINT

5 10 15 20 25 30

DISTANCE IN MILS FROM EDGE OF SOUND TRACK

Slide 6.

how it varies as it moves in from the
sprocket-hole edge. Photographic re-
cording is all recorded on a drum and
reproduced at a drum; therefore, the
conditions as far as polygoning are con-
cerned are the same. This effect shown
here was not due to any laboratory effect.
The negative track described above
and the resulting print were developed
in a special tank which eliminated any
laboratory development sprocket-hole
modulation.

Slide 6 shows plots of results. There
are two methods of measurement. The
heavy solid curves were taken on a
microphotometer, which took in all the
peaks. The dotted points are the
measurements made with a device which

is quite similar to the modern inter-
modulation analyzer. And you will
notice, disregarding the amplitude be-
cause it shouldn't necessarily be the same
in photographic as magnetic, that the
effect flattens off pretty much around
25 or 30 mils. Based on this informa-
tion we set up, in cooperation with
MGM, the 200-mil push-pull track in
the so-called offset track position. Now
the intermodulation analyzer, when used
to measure amplitude modulation,
measures not only 96, it also measures
everything else. The statement, made
in Mr. Grignon's contribution above,
that they got 4% on old film was used
by Mr. Ryder as favoring his case; but
it doesn't favor his case at all because

132

February 1952 Journal of the SMPTE Vol.58

there is no reason at all why 96 cycles
should be affected by old film. The
old film is contributing the hash, the
96 is very much the same. I see no
reason why the 96-cycle should change
appreciably. If you're going to use old
film, you're going to get more amplitude
modulation of a general nature.

Dr. Wolfe and I set the track position
at 135 mils which incidentally is what
Mr. Ryder is proposing now. We found
that it was a very good position to
operate in because sprocket-hole modula-
tion was at a minimum. And we would
have been very happy to stay at 135
mils, if we hadn't had this kaleidoscopic
Hollywood, this ever-changing situation
to contend with. We no sooner got it
set there when up came John Livadary
with his bright idea of three tracks on
35mm film. As a result, RCA and our-
selves got together to study where we
could put these tracks. Westrex thought
that 75 mils in from the sprocket-hole
edge would be a good compromise. At
first nobody worried much about the
crosstalk between adjacent tracks. But
when it was decided to use these tracks
for storage of three independent films
the situation began to get fussier as to
crosstalk, and as a result we had to work
out a wider separation of the tracks,
resulting in the outer tracks being located
50 mils in from the sprocket holes.

At that time, in 1949, we measured
amplitude modulation products and the
information was presented orally to the
Research Council. I would like now to
show a slide (No. 7) which will show the
measurements we made in July 1949
on amplitude modulation with the
film as it was in those days. The upper
curve was made on the RA-1231 Mag-
netic Recorder drum and at 50 mils in
it shows 4%, but it dropped very little
until we went in as far as 200 mils. It
never dropped much below 3%. In
other words, it looked like it would never
get below 3% no matter where you
located the track. The only conclusion
I could draw from that was that the

SO 100 ISO 200

TRACK DISTANCE FROM SPROCKET HOLES

Slide 7.

residual amplitude modulations un-
doubtedly came from scratches and dirt
and hash. Now since Mr. Ryder made
his proposal to change the proposed
standard we repeated the measurements
in September 1951. This time we used
our new RA-1497 Recorder with drum
scanning. There are some significant
differences. The drum size is greater
on the 1497 than it is on the 1231 and
we know from previous studies that
sprocket-hole modulation is generally
more severe with small drums. It's also
a matter of film tension. In the case
of magnetic recording it is also a function
of the tilt of the head. We obtained the
lower curve shown in Slide 7. You will
notice that at 50 mils in we get slightly
over 1% which agrees with Mr. Grig-
non's contribution.

Mr. Ryder has said that amplitude
modulation experts could hear 2%,
halfway experts 5% and the public
10%. There are no published data,
although the Bell Labs did some work
on this problem, on how much ampli-
tude modulation a person can hear.
First of all, it depends on many factors,
the frequency that is being modulated,
whether it's 100 cycles, 1000 or 10,000.
It also depends on the modulation rate,
just as flutter does. I tried to work out

Ryder and Denney: Magnetic Track Placement

133

something that might help us see what
would be the minimum we could hear.
With an amplitude modulation of a
carrier only two sidebands are pro-
duced. If you have, say, 20% amplitude
modulation, 10% lies in each sideband.
In flutter, on the other hand, which is
an FM type of modulation, one obtains
an infinite series of sidebands, that is,
if the modulation index is high enough.
When the modulation index is low
enough in FM you also get only two side-
bands. So it's natural to suppose that
a flutter having this index of modulation
would sound to the ears just like an
amplitude modulation having the same
sidebands. Now, the maximum value
of the modulation index at which you
can neglect the higher orders of side-
bands in FM is of the order of 0.025,
the first-order sidebands being about
10% of the carrier. This index of
modulation is designated by the Greek
letter a.
Now,

A/o

Jm

where

A/o = frequency deviation of the carrier
and

fm = the modulation (i.e., flutter) rate.
Substituting:

a. = 0.025 and/m = 96,
A/o = 2.4 cycles

The flutter in a 3000-cycle tone is given
in % by:

flutter =

We noted previously that for this flutter
condition the first sidebands were 10%
of the carrier. In the case of amplitude
modulation, this corresponds to a value
of 20% since one-half, or 10%, lies in
each sideband. Similarly, a 10% ampli-
tude modulation corresponds to a peak
value flutter of 0.04% at the 3000-cycle

rate. Now, this is about as good a
commercial film reproducer as can be
built and it would seem to signify that
10% amplitude modulation would be
largely inaudible, at least at 3000 cycles.

What minimum 96-cycle flutter can
be detected is somewhat controversial.
It depends on the person and it depends
on the room in which the tone is being
heard. Manufacturers of sound-record-
ing equipment have tried to keep 96-
cycle flutter somewhere between 0.05%
and 0.1% and we think 0.05% is pretty
good. It would appear, therefore, that
we could similarly tolerate 10% ampli-
tude modulation. Since our graphs
show a little over 1% in new film, we
do not feel that there is any great
problem in the proposed location of the
track.

Mr. Ryder: With respect to the optical
versus magnetic comparison of sprocket-
hole modulation, it should be noted that
with optical film as long as there is a
signal across the film, it is seen by the
photoelectric cell. Sprocket-hole modu-
lation on optical film is the result of a
change in photosensitivity due to punch-
ing, a change in the developing effect
near the sprocket holes due to agitation,
and polygoning. In magnetic record-
ing and reproduction the effect that
we are noting is a result, partly at least,
of deformation from punching which
causes a fluctuation in magnetic head
contact with the film. All one has to
do is hold the film in reflected light and
observe the deformation from punching.

There is another phenomenon that
has been observed — that the effect of
sprocket-hole modulation varies with
frequency. We have not searched foi
the point where the highest modulation
takes place. We should expect the
modulation effects to be greater at
higher frequencies and lower in the
2000-cycle range where the more recent
Westrex tests were made.

With respect to the old films, our
definition of old film as presented here
a few minutes ago is not so much a

134

February 1952 Journal of the SMPTE VoL 58

question of age in time as a question
of age in usage. By examining film
that has been used many times, the
sprocket wear and deformation are
quite obvious, which can only increase
rather than diminish the problem. At
Paramount and at Ryder Services,
where we are using the sound track
placement suggested here, we use old
films interchangeably with new films
and have noticed no bad effect. We see
no reason for buying new film because of
any deterioration of the film or from the
standpoint of the hash mentioned by
Dr. Frayne.

I do not have data to show the effect
of increase or decrease in this hash.
I should point out, however, that the
system of measurement used by Para-
mount separates sprocket-hole modula-
tion from the so-called hash; whereas
all of the other data presented at this
meeting combine sprocket-hole modula-
tion and so-called hash to the point of
confusion. As pointed out in my
paper, Paramount changed from the
distortion and intermodulation type of
measurement to the use of the oscillo-
scope in order to avoid this measurement
trouble.

I should expect that in the future we
might develop recording machines and
magnetic film which will have less
sprocket-hole modulation than we are
now encountering. We have piesented
this paper on the basis of a recorder
under normal present-day conditions of
operation. If we were to repeat these
measurements as we have in the past on
several occasions, our results would be
the same.

I concur with the mathematics which
Dr. Frayne has placed on the blackboard.
I was careful and punctuated my word-
ing with respect to our observations of
percentage modulation and made it
clear that these observations were under
our conditions of use and measurement.
Our first concern is what we can hear,
what can be heard by the average
person and what annoys the average

person. Our measurement data are
related to these observations. It is
my feeling that any measurements and
any data which are tied into flutter
modulation may be quite different from
that which is now taking place with re-
spect to amplitude modulation. Many
people are confusing these two types of
modulation. We were confused at first,
but it is quite clear to us now that our
concern is amplitude modulation which,
incidentally, can be additive along with
generations of transfer. When each
generation of transfer adds up in the
same direction, you can gain a very
high percentage of sprocket-hole modu-
lation. Unfortunately, they never com-
pletely cancel. Again I say, it is the
occasional bad quality and not the good
quality that causes us trouble. We are
endeavoring to eliminate the occasional
bad quality. We hope that this elimi-
nation will also improve the good
quality.

We present this information to the
Society of Motion Picture and Tele-
vision Engineers and the Research
Council as a study which we have made
in all sincerity with the thought that
the knowledge that we have gained
should be made available to all. The
utilization of this knowledge, its ac-
ceptance or its rejection, is up to the
Society and the Research Council.

I should clarify one point and that
is I doubt if Paramount will move over
to 50 mils from the sprocket holes. In
any case, Paramount will make its
recordings and reproducers so that
they can be played interchangeably
with whatever standard is finally ac-
cepted. I am very much opposed to
getting into another turmoil of the type
that now exists on 16mm work.

Mr. Mueller: I think it is time that
you should hear from the Sound Com-
mittee of the Research Council who drew
up these present standards and who
presented them to the SMPTE. These
proposed standards were published in
July of 1951.

Ryder and Denney: Magnetic Track Placement

135

You have heard the pros and cons
as presented here which is really an
extension of the discussion in our com-
mittee, as most of the information shown
today was gathered at the request of
our committee and discussed thoroughly
at meetings extending over more than
two years. We finally decided that it
was very important to all of us that no
more delays be tolerated and that the
magnetic standards favored by the vast
majority of the committee be adopted.

Speaking as a member of the Sound
Committee of the Research Council and
as its present chairman, I want to state
that we propose to stand by the standards
as established and as published.

There have been 150 channels built
by the two major manufacturers here in
Hollywood based on the performance
given in the slides presented here today;
and as far as I know, Loren, you have
the only machine that does not work.
So I think that rather than move the
standards, perhaps you should fix your
machine.

Mr. Ryder: I wish to make a point
clear for the record; and that is, if the
Committee of the Research Council
have made up their minds before they
hear an honest debate of this discussion,
I don't think it's worth while to follow
the recommendation of the Research
Council. I don't believe that it is on
that basis.

Mr. Mueller: I have recently discussed
this with the other members of our
committee, of which you are a member,
and we see no reason for changing our

opinions which were based on a study
of more than two years.

L. T. Goldsmith: As Chairman of the
Sound Committee of the Society, I
wish to reconfirm that our Subcommittee
on Magnetic Recording, under the
chairmanship of Glenn Dimmick, had
given wide study to all the proposed
standards on magnetic sound track for
over three years before they were pub-
lished in the July issue of the Journal.
Any comments received during the
90-day trial publication period are
welcome and will be carefully studied
by the Subcommittee. I would like to
point out, however, that in the interest
of avoiding industry chaos it is to the
best interests of both the users and
manufacturers of magnetic-recording
equipment that standards which are used
and approved by the great majority
be adopted as rapidly as possible.

Added note by Mr. Ryder: Although it
was not discussed at the meeting, the
data collected at Paramount have been
questioned on the basis that some of the
measurements involve a larger-than-
normal space between the drum and
the record-reproduce head. We have
curves to show that very bad effects
can be produced by improper adjustment
and tension of the head, especially under
such conditions. The data presented
and the curves shown in the paper are
for the conditions where these effects
would not exist and further, the curves
as presented, and as shown by the
graphs, tie together with measurements
made with heads located as recom-
mended by the manufacturers.

136

February 1952 Journal of the SMPTE Vol. 58

New Principle for

Electronic Volume Compression

By HAROLD E. HAYNES

The principle described is a radical departure from those heretofore used
in compressors. The features of this compressor are extremely low thump,
very fast action (if desired), low distortion and freedom from the need for
special circuit components or selected tubes. Fundamental circuits are
discussed, and performance obtained with a complete compressor embodying
the system is presented.

A

VOLUME COMPRESSOR is an automati-
cally actuated variable-gain amplifier,
used for reducing the dynamic range of
program material. The timing charac-
teristics of the voltage derived from the
signal for actuating the variable-gain
amplifier are customarily such as to
provide a very rapid gain reduction
whenever the signal level rises abruptly,
but to increase gain relatively slowly
when the signal level drops. Very short
acting times, less than one millisecond,
are often used in order to minimize
unwanted initial peak amplitudes on
sounds having sudden large increases
in envelope amplitude, such as certain
spoken syllables.1 If a change of gain
is accompanied by a shift in d-c axis of
the wave, a spurious aperiodic signal,
commonly called "thump," will be

Presented on October 18, 1951, at the
Society's Convention at Hollywood, Calif.,
by Kurt Singer for the author, Harold E.
Haynes, Radio Corporation of America,
RCA Victor Division, Bldg. 10-4, Camden
2, NJ.

produced. The d-c component of this
shift will, of course, be filtered out by the
low-frequency cutoff characteristic of the
system; nevertheless, to the extent that
the gain-reducing action can be con-
sidered instantaneous, this shift is a
step-function and contains energy at all
frequencies. The more rapid the attack
and the better the low-frequency response
of the system, the more objectionable
will be the thump.

Background

Brief mention of a few commonly
used methods of varying gain will serve
to point out their shortcomings as far
as balance, or tendency to produce
thump, is concerned. The most com-
mon type of compressor employs as a
variable-gain device some nonlinear
electrical element, an element in which
the two electrical quantities employed
as input and output are related by a
curved characteristic. This type of
element is utilized in such a way that
the slope of the characteristic at the

February 1952 Journal of the SMPTE Vol. 58

137

operating point determines the gain of
the circuit in which it is connected (which
in general may be either greater or less
than unity). Variations in gain are
produced by superimposing upon the
input signal an adjustable control signal,
the amplitude of which determines the
operating point. Examples of this type
of variable-gain device are nonlinear
semiconductors, such as Thyrite, and
vacuum tubes as usually used in com-
pressors and limiters.

In the latter class is the familiar
"variable mu" or "exponential" pen-
tode, in which various points on a curve
of transconductance vs. grid voltage are
selected by adding a control voltage to
the signal in the grid circuit. It is clear
that in this case, as with all others in
which the control effect is merely a bias
superimposed upon the signal, there will
inevitably be an output component
produced by a change in gain, and hence
a thump.

There are other vacuum-tube variable-
gain circuits in which the controlling
voltage is not superimposed upon the
signal. One example is the "loading-
tube" circuit, in which the plate im-
pedance of a tube is shunted across a
relatively high-impedance signal source,
and the value of this impedance is
changed by varying the grid voltage.
Here a family of curves of plate current
vs. plate voltage exists, their slopes
varying as a function of grid voltage.
Unfortunately, however, changing from
one curve to another causes a change in
plate current, so that the same funda-
mental problem presents itself, as before.
A generalization may be made to the
effect that a change in any tube charac-
teristic causes a change in plate current;
hence, circuits of this class also suffer to
a greater or lesser extent from an in-
herent tendency to thump.

The obvious and almost universal
remedy is the use of push-pull circuits,
in which signal is applied to the two
variable-gain elements out of phase,
while gain-control voltages are applied

in phase. Recombining the outputs in
push-pull fashion then makes the signal
components in phase and the gain-
control components out of phase, so that
they tend to cancel. A great reduction
in thump is thereby obtained, but it
is apparent that in order for perfect
cancellation to occur under all condi-
tions the characteristic curves of the two
elements must be identical at every
point in their operating ranges. An
estimate of the degree of similarity re-
quired if the thump level is to be neg-
ligibly low may be made on the basis of
the following arbitrary but not un-
unreasonable assumptions: (1) that a
change of gain between any two values
within a range of at least 10 db should
produce thump of the order of 40 db
below signal level; and (2) that signal
level is limited to 5% modulation of
plate current by considerations of non-
linear distortion. These values lead to
the conclusion that the plate currents
must be equal (or differ by a constant
amount) within something of the order of
0.05% throughout the operating range.
Obtaining and maintaining the degree of
similarity of characteristics necessary
for such high-quality performance,
though sometimes adequately accom-
plished, is expensive and time-con-
suming, and it frequently entails special
selection and aging of tubes, plus
frequent checking of those in service-
Principles of the New System

A means of varying gain by employ-
ing vacuum tubes, but one which is not
based upon nonlinear characteristics,
was thus sought and an approach which
proved fruitful is described. It is
based upon the principle of keying a
transmission device between gain values
of zero and some fixed value, at a high
frequency, and obtaining different effec-
tive-gain values by controlling the rela-
tive durations of "off" and "on" periods.
Otherwise expressed, this means ampli-
tude modulating the signal with a
high-frequency rectangular wave or

138

February 1952 Journal of the SMPTE Vol. 58

Fig. 1. Sinusoidal signal of
frequency Js modulated by a
rectangular wave of fre-
quency fk.

SIGNAL (£a)

MODULATING WAVE
A '.25

MODULATED SIGNAL

series of rectangular pulses of varying
duty factor. Of course, such a modu-
lated signal contains high-frequency
components not present in the original,
but by proper choice of modulating
frequency, these can readily be made
inaudible and easily separable from the
signal by filters.

The action is illustrated in Fig. 1.
A sinusoidal signal of frequency Js is
shown modulated by a rectangular wave
of frequency /#, in which the duty factor
is k. It is shown in Appendix A that
the modulated wave contains the original
signal multiplied by the factor k, plus an
infinite number of modulation products
of frequencies nfk ± fs. It follows that
if the maximum signal frequency com-
ponent to be accommodated is, for
example, 15 kc, the lowest sideband will
be fjf — 15 kc. This sideband should be
substantially higher than the maximum
signal frequency, to facilitate removal of
the sidebands. (It is pointed out later
that the keying pulses should be as
nearly rectangular as possible; hence,
it is desirable to use the lowest per-
missible keying frequency, in order to
minimize circuit difficulties.)

With the unwanted components of the
modulated wave filtered out, there
remains only the desired signal multi-
plied by k; hence, if the value of k
can be varied in accordance with an
appropriate control voltage, compression

involving only linear electrical elements
will have been accomplished.

Since the keying frequency must be
at least 30 kc, a vacuum-tube circuit
appears to be the only promising type
of keying device. Hence, the same
objection that was raised previously to
tube circuits may at first seem valid,
namely that a d-c component of plate
current, which will change with changes
in gain, will still be required. The
important distinction here is that the
tube will need to operate only at one
mean value of plate current (corresponding
to "on"), and at cutoff (corresponding to
"off"). Thus, any two tubes can be
used in push-pull, and substantially
perfect balance can be obtained at their
single operating points. They can, and
should be, linear devices, and as such
will permit relatively large signal ampli-
tudes without objectionable distortion.
Furthermore, their linearity may be
enhanced by means of negative feed-
back, an expedient which would tend
to nullify the gain-changing properties
of conventional circuits.

Circuit Methods

Figure 2 shows the basic circuit of
such a keyed amplifier. Two cathode
followers are connected in push-pull,
with positive keying pulses introduced
in the cathode circuit. Pulse amplitude
is sufficient to cut off plate current com-

H. E. Haynes: Electronic Volume Compression

139

pletely even when peak signal amplitude
occurs. Additional positive bias volt-
ages, Eci and EcZ, permit desirable
operating points to be selected, one of
them being adjustable to permit balanc-
ing.

It is apparent that the keying pulses
must have negligible rise and fall times,
in order that the tubes will not be operat-
ing at points on their characteristic
other than the desired one during an
appreciable fraction of the time. This
means that a minimum of capacitance
loading should be permitted at any
point in the pulse circuit. Therefore,
resistors R3 and R4 are inserted to iso-
late the output transformer from the
pulse circuit. Unwanted modulation
products are removed by a simple low-
pass filter following the out-put trans-
former.

An essential adjunct to the keyed
amplifier, when used in a compressor,
is a source of pulses of controllable dura-
tion and of approximately constant fre-
quency, having the requisite relation
between duration and control voltage.
Appendix B shows that in a compressor
deriving control voltage from output,
as is customary, and having a slope of
£ on a decibel basis (2:1 compression),
numerical gain should be inversely
proportional to control voltage; hence,
a pulse generator was developed in
which the "on" (negative) pulse width
closely approximates this relation. Fig-
ure 3 shows the basic circuit of the pulse
generator. A 45-kc square wave,
generated by a multivibrator, is differ-
entiated by Ci and R5, to produce a
series of alternate positive-voltage and
negative-voltage pulses of very short
duration. The negative pulses cause
capacitor C2, which is also connected
to the grid of sharp cutoff pentode V3,
to be charged negatively once for each
cycle, through diode V2. C2 discharges
toward zero through R6, which is con-
nected to the source of control voltage,
the latter being variable from zero to a

relatively large positive value. Thus,
the plate current of V3 is cut off for a
portion of each interval between pulses
which becomes smaller as the value of
the control voltage is increased. It is
these periods of cutoff which eventually
become "on" pulses for the keyed ampli-
fier, their duration relative to the pulse
period being the factor k. The time
constant of G2 and R6 is made about
equal to the pulse spacing (22 ^sec), and
the potential to which C2 is charged by
the negative pulses is about ten times
the cutoff grid voltages of V3; hence,
V3 draws no plate current unless the
control voltage has a substantial posi-
tive value. This means that the signi-
ficant part of the discharge curve of C2
is reasonably linear, and it can be shown
that this causes the duration of the cutoff
period, and hence the value of k, to be
nearly inversely proportional to the
control voltage, as desired. The rapidity
with which the value of A: can be changed,
and hence the speed of action of the
compressor, in practice is limited only
by the properties of the circuit by which
gain-controlling voltage is derived.

The plate-current pulses of V3, which
are roughly rectangular because of its
sharp cutoff characteristic, produce
voltage pulses which are further shaped
by subsequent amplifier and limiter
stages so as to have very short rise and
fall times, and applied to the amplifier
circuit of Fig. 2.

These two basic circuits, with the
addition of conventional means of deriv-
ing control voltage proportional to
compressed output, and having the
desired timing characteristics, constitute
a complete compressor. Since this type
of control circuit is well known, and for
the present application need be little
different from those for other compres-
sors, this subject will not be discussed
further.

Performance

A complete compressor based upon
these circuits has been built and is

140

February 1952 Journal of the SMPTE Vol. 58

o o

MODULATING-
WAVE INPUT

OUTPUT

Fig. 2. Basic circuit of vacuum-tube keyed amplifier.

"LTLT

v v

i r II , A-, • \

f

f

^

c,

\\ 11

--- i

^J \

TIVIBRATOR
45 KC.

1 !

:*s cz -

s

<
<

1

':*•

V

<
«

»
>
>

<

t

•^ r

:b CONTROL

^ <
. VOLTAGE -£y £.

)

Fig. 3. Basic circuit of pulse generator.
H. E. Haynes: Electronic Volume Compression

141

Fig. 4. Complete compressor based on the circuits shown in Figs. 2 and 3.

illustrated in Fig. 4. Its operating
characteristics were made to conform
to those of existing compressors so that
its performance could be easily evaluated.
Although this model is somewhat more
complex than the simplest compressors,
no special tubes or other special com-
ponents are used. It affords a useful
gain-reduction range of more than 15
db. Performance, especially with regard
to thump, is excellent, both with respect
to the degree of balance obtainable and
to the long-time stability of this balance.
Two chief methods have been used
for observing and measuring the effects
of unbalance in compressors. One
which is typical of actual operating
conditions, described by Maxwell,2 con-
sists of abruptly raising the level of a
relatively high frequency sine-wave input
signal to the compressor and observing
on an oscilloscope the transient appearing
in the output. Although it depicts the
thump phenomenon very graphically,
this test is open to the objection that it
takes into account balance conditions
at only two specific points in the gain-

reduction range. Also, in a well-
balanced compressor the transient ampli-
tude is too small to be conveniently
observable. A second method, often
built into compressors as a balance
check, measures cross modulation be-
tween the gain-control circuit and the
signal circuit by applying a sinusoidal
test voltage in the gain-controlling cir-
cuit. A single test of this kind includes
the effects of unbalance at all points in
the gain range swept, and if the test
conditions are suitable it affords a good
overall evaluation of balance.

Measurements of both types have been
made on the pulse-modulation compres-
sor. In the first type, a 250-cycle low-
pass filter was used, following the com-
pressor, to reduce the carrier amplitude
and thereby make the transient more
easily seen. For a 10-db increase in
input (5-db gain reduction), signal-to-
thump ratios of 50 to 60 db were ob-
tained.

The cross-modulation method is felt
to be preferable for specifying un-

142

February 1952 Journal of the SMPTE Vol. 58

balance, because it does detect unbalance
at all points in the range used. A figure
of merit called "signal-to-unbalance
ratio" is proposed to describe the per-
formance of a compressor when tested
in this manner. It is expressed in deci-
bels and is defined as follows: Signal
level is the maximum output level, with
10 db of gain reduction, at which some
satisfactorily low value of total harmonic
distortion of a 1000-cycle signal is pro-
duced. In the present case, this value
is taken as 0.5%. Unbalance level is
the output produced, in the absence of
signal, by a 60-cycle control voltage
which varies the gain reduction through-
out the range of 0 to 10 db.

Using this cross - modulation test
method, excellent signal-to-unbalance
ratios have been obtained, along with
freedom from the need for special
tube selection and from the necessity
for frequent rebalancing. Tests have
shown that, except for the possible re-
jection of perhaps 10% of samples,
tubes selected at random for the variable-
gain stage (6SN7GT) will all produce
optimum signal-to-unbalance ratios of
55 db or more. Operation over periods
of a few hundred hours has indicated
that the balance does not deteriorate
more than 10 db during this length of
time, and that the original figure can
be readily regained by rebalancing.
Although unregulated heater and plate
supplies were used, line-voltage varia-
tions of 1 0% also increase the unbalance
only about 10 db.

By adoption of pulse-modulation tech-
niques, it has thus been possible to
construct a compressor whose perform-
ance regarding thump is equal or
superior to that of any now used in the
most exacting applications, without the
need for specially selected tubes or other
components. Its moderate added com-
plexity is felt to be of secondary im-
portance in the light of its very signi-
ficant advantages.

APPENDIX A

The modulating wave of Fig. 1 can be
represented by the expression3

+ - ( sin kir cos co*/ -f -= sin 2kv

7T \ /

+ - sin nkir cos

01

(1)

where :

a = instantaneous value of gain,
AO = gain value during "on" periods,
ojfc = 2irfk = fundamental angular

frequency of modulating wave,
k = ratio of pulse width to period of

modulating wave.

The signal wave is:

es = Es sin ust. (2)

An expression for the modulated wave
is obtained by multiplying (2) by (1):

aes = AoEs\ k sin <ast -\ — (sin dist sin for

cos <i>kt + « sm

-\ — sin Ugt sin nkir cos mo*/ J I . (3)

The first term, kAoE», is the desired
output. Each of the other terms is the
product of a sine term, a cosine term and
a constant, depending upon the value of
k. The general term:

- sin (ast sin nkir cos n&kt (4)

n

can be rewritten as:

- sin nkir\ « si

sin

2n

Since co* > co,, this is better rewritten as :

sin nkir [ — sin (nco*/ — co.O

+ sin (ncojt* + co./)] . (6)

H. E. Haynes: Electronic Volume Compression

143

APPENDIX B

If the gain-vs.-input relation in the
compression range is to be linear when
expressed in decibels,

Db0 =

i + c,

(7)

where:

Dbo = output level in db,

Dbi = input level in db,

k = slope of compression curve,

c = a constant.

On a numerical basis,

20 log Eo

20 k log Ei + e, or log Eo

= k log Ei + c', (8)

where :

Eo = output voltage,
Ei = input voltage,
c' = a constant.

If the gain-controlling voltage is derived
from and is proportional to compressor
output, it is of interest to express voltage
gain as a function of output :

(9)

and determine the nature of the function/.

From (8),

Eo = c'Ei*,orEi = (^)k. (10)
Thus:

/(E0) = f? = -^

•&»

at)

* ('4)

= r'^o

For 2:1 compression, k = -, therefore:

f(Eo) = c'Eo

—
Eo

0-170

(12)

References

1. R. O. Drew and E. W. Kellogg, "Start-
ing characteristic of speech sounds,"
Jour. SMPE, 34: 43-58, Jan. 1940.

2. Donald E. Maxwell, "Dynamic per-
formance of peak-limiting amplifiers,"
Proc. IRE, 35: 1349-1356, Nov. 1947.

3. F. E. Terman, Radio Engineer's Hand-
book, 1st ed., McGraw-Hill, New York,
1943, pp. 22, 23

144

February 1952 Journal of the SMPTE Vol. 58

Prints From 16mm Originals

By R. L. BUTTON, K. B. CURTIS and LLOYD THOMPSON

The introduction of reversal film — both black-and-white and color — made
16mm photography very acceptable for commercial use. Release prints in
quantity were a problem. New printing equipment had to be designed and
built, and new materials and techniques had to be improved. This paper
will describe the methods used by The Calvin Company for producing high-
quality release prints in quantity at the present time.

JL HE INTRODUCTION of black-and-white
reversal film and, later, the introduction
of color reversal film made 16mm
photography practical. Through the
use of these 16mm reversal materials,
the field of motion picture photography
was extended to many potential pro-
ducers. Originally, it was thought that
16mm film would be for amateur use
only. It was but a short time, however,
until it was also being used for pro-
fessional purposes. The amateur motion
picture field has continued to grow and
change, and today we find wide pro-
fessional use of 16mm materials while
the majority of amateur use is 8mm
film. With professional use of 16mm
materials, duplicate prints naturally
were desired. This involved extensive
work in printer design and also in
photographic research to obtain better
materials and methods for making
duplicate prints from originals, both in
black-and-white and in color.

Presented on October 17, 1951, at the
Society's Convention at Hollywood, Calif.,
by R. L. Sutton, K. B. Curtis and Lloyd
Thompson, The Calvin Company, 1105
Truman Road, Kansas City 6, Mo.

Much of the research and design of
printers has been accomplished by the
pioneers of 16mm film because they
were intimately aware of the changes
resulting from the growth and improve-
ments in this field. Two important
factors resulting from the improvement
of 16mm materials vitally affected the
design of printers for 16mm films:

(1) the changes made in the physical
characteristics of the original film; and

(2) the fact that more and more printing
exposure light has been necessary to
print the new products as they were
developed. Generally speaking, a
printer designed to print one type of
16mm film often was obsolete almost
overnight when another type of film
showing new, different and improved
characteristics was introduced; there-
fore, frequent changes in printer design
were necessary in the early days of the
industry. The problem was not so
much in using the newer materials as in
the difficulties of printing films as they
aged. The problem of shrinkage and
the effect of aging on film splices was
often pronounced in older originals
and it was difficult to find a printer

February 1952 Journal of the SMPTE Vol. 58

145

that would accommodate both normal
and shrunken or aged originals. With
the introduction of newer materials, the
necessity for increasing the exposure
light often demanded changes in printer
design.

In setting up specifications for a
printer, or a system of printing, it would
appear that the problem would be
simple. The problem is to make sound
prints, in either black-and-white or
color, that have consistently good quality
at a speed that is economically feasible
when using 16mm reversal originals.
However, to accomplish these things, we
believe a printer, or system of printing,
should have the following characteristics:

1. The printer must give good contact
and produce prints having good steadi-
ness.

2. Sufficient uniform illumination
must be available to print on such
Eastman stock film as Type 5504, reversal
duplicating; Type 7302, fine-grain release
positive stock; Type 5365, black-and-
white fine-grain duplicating positive;
and Type 5265, color stock. If possible,
additional illumination should be avail-
able to provide for future needs.

3. The printer must handle originals
so that they are not scratched or
damaged, even though large numbers of
prints are made from the same original.

4. The printer must be able to make
satisfactory prints from originals with
normal and abnormal shrinkages. Our
definition of these would be a shrinkage
of from 0.01 to 1.5%.

5. A system of light changes for density
corrections in the original must be pro-
vided, and this should be as foolproof
as possible.

6. A minimum of maintenance should
be required on the machine, and this
must be done with a minimum of ex-
pense.

7. The printer, or system of printing,
should provide for optical effects, in
each individual print, as well as straight
printing.

8. Provisions should be made for
adding a filter pack to the light system.
This should be far enough away from
the light source so that the heat does not
damage it over a long period of time.

9. Future requirements should be
anticipated, insofar as possible, for such
things as the color correction of indi-
vidual scenes, when such methods
become practical.

10. Means should be provided for an
accurate measurement of illumination,
both as to quantity and color quality
of the printing light.

11. The power supply should be kept
simple and, if possible, should operate
directly from standard 60-cycle a-c
current.

12. In designing, we feel that wherever
possible standard parts, available on
the open market, should be used to keep
the original cost down, but more im-
portant, to allow for repairs with a
minimum of trouble and expense.

13. The operation of all printers
should be as simple and as automatic
as possible so as to require a minimum
of training for new operators, and thus
reduce errors.

14. The printer should accommodate
a minimum original print footage of
1 200 ft and, if possible, it should handle
2000-ft rolls of originals and raw stock.

15. The take-up mechanism should
handle both short and long lengths of
film without trouble.

16. If the design of the printer is
such that there is any tendency for the
printing aperture to collect dirt, lint
or hairs, an air blast should be provided
to keep it clean at all times.

17. It would be highly desirable for
the light-change cuing device to be
standard. A notchless system is pref-
erable. However, inasmuch as there
is no standard for this, each laboratory
has set its own standard as to where
film should be notched. While it
seems to be impossible to standardize
the number of frames between the scene
changes and the notch, it is in most

146

February 1952 Journal of the SMPTE VoL 58

cases possible to standardize the type
of notch. We have chosen the Bell &
Howell narrow notcher for this purpose.

18. In making black-and-white re-
versal prints from black-and-white or
color originals, reversal color prints
from color originals, dupe negatives
from either black-and-white or color
originals, it is necessary that the light
change between each step be greater
than for printing positives from original
negatives, as is customarily done in
35mm film. The design of a light-
change system must take this into con-
sideration.

1 9. The claw which moves the original
and raw stock in a step printer should
be exactly opposite the picture aperture
in order that the framelines in the print
be as nearly like those in the original as
possible.

No doubt other specifications could
be added and probably will be, as tech-
niques and materials change.

We originally tried to solve our first
printer problem by converting a Bell &
Howell projector head into a printer.
In many ways this did a good job, but
it was not too long before it was obsolete.
Several different printers were built
and tried, but each had its limitations.
We finally reached the conclusion that,
for a system of printing suitable for our
use, it would be necessary to have three
types of printers:

1. A step-type picture printer.

2. A continuous-type picture printer.

3. One or more types of sound printers
to add sound to the picture prints, made
on the step- and continuous-type printers.

The Step-Type Printer

First, let us describe the step- type
picture printer and see how it meets the
specifications. By looking at Fig. 1,
you will immediately recognize that a
number of standard parts have been
used to build this printer. Some of
these parts, such as gears, which cannot
be seen in Fig. 1, are also standard. An

inspection of the printing gate will show
that the raw stock and the original film
are handled separately so that tension
is applied to each of the films. They
are also edge-guided, separately. This
was done in order to assure a steady
print and to eliminate side weave.
The printing gate is curved in order to
remove the curl in the original film, so
that good contact could be made with
the raw stock. By using this method to
flatten the original film for good contact,
it has been possible to relieve the print-
ing gate at all points where the original
film would touch metal. Experience
has shown that the only way to keep
from scratching film is not to let it drag
on anything, regardless of how highly
polished it may be. Such a surface
will eventually cause trouble.

Certain types of duplicating film,
especially Kodachrome, have a tendency
to curl or cup at low relative humidities,
which means that good contact is not
always possible in the middle of the
picture. This tendency can be mini-
mized greatly by maintaining the rela-
tive humidity in the printing room at
about 50%. To eliminate this difficulty,
a special pressure shoe was designed to
hold the raw stock against the original
in the center of the film. Thus, the
problem of contact and steadiness was
solved in this particular printer. This
type of gate is very easy on the original
film. Damaged sprocket holes, or other
defects in the original, may cause it to
lose a loop, but the original is not
damaged. Examination of the printing
gate will show that the pulldown claw
is exactly opposite the picture frame, thus
making it possible to keep the frameline
as nearly like the original as possible.
Productions photographed with several
different cameras having widely differ-
ent framelines will cause trouble. About
the only way to minimize such trouble
is to use an especially wide frameline
in the printer at the time prints are
made from such originals.

In order to secure optical effects in

Sutton, Curtis and Thompson: 16mm Prints

147

Fig. 1. The step picture printer with light-change board.

the prints, The Calvin Company has for
a number of years printed from A and
B rolls in combination with A and B
rolls of optical-effects mattes (Fig. 2).
This means that our printers must be
able to run the optical-effects mattes.
This system has been described before,1
and since most people are acquainted
with such mattes only a few things need
be said here. The success of the system
depends upon the optical-effects mattes
being projected onto the back of the
original film as it is printed on raw stock.
Such mattes cannot be run in contact
with the original, as has been done in
35mm, for several reasons. Trying to
run three pieces of film through one
film gate causes a lot of trouble and in
addition, any dirt, scratches or slight
defect in the matte will be printed into
the final print quite easily. By project-
ing these mattes and throwing them

slightly out of focus, nearly all of these
difficulties have been eliminated, and
doing so completely avoids the trouble
of trying to run three pieces of film
through one gate at the same time.

A standard Bell & Howell silent
projector with certain modifications was
used as a matte runner, and as a light
source for printing. In order to use a
projector for this purpose it was necessary
to disconnect the regular projector motor
and use an external constant-speed
motor to drive the ventilation fan,
because the regular projector motor
would not stand up under such long,
hard service. It was also necessary to
construct a special tube which would
fit very close to the aperture of the
projector in order to eliminate the
majority of the stray light which escapes
from the projector in the printing room.
A method had to be provided for carry-

148

February 1952 Journal of the SMPTE Vol. 58

Fig. 2. Optical-effects and density mattes, A and B rolls, etc.,
are checked on this multiple synchronizing equipment.

ing the heat from the lamp out of the
printing room and, in doing this, we
have also trapped the stray light at
the top of the lamphouse to prevent its
entering the printing room. The optical
system for projecting the effects mattes
is a standard projection lens. We use
the largest-aperture lens available for
the focal length involved to gain efficient
light transmission. The color-correction
filters used in the filter pack are placed
in the printer head of the mechanism
as far away from the light source as
possible, and there has been practically
no damage caused by heat from the
lamp affecting these filters. They are
stable over a period of many months.

A special device has been made which
can be locked over the printing aperture
to hold a photoelectric cell for measuring

illumination. This photoelectric cell is
connected to a galvanometer. Further-
more, the device is fitted with a multiple
slide carrier which holds color filters
as well as neutral density filters. When
the illumination level is to be measured
the neutral density position on the carrier
is used. The other three positions are
used for checking the quality of the
printing light for color printing and for
matching individual printers. This
device has previously been described by
P. S. Aex.2 We have added the fourth
slide with the neutral density filter for
checking the quantity of light, and have
found this to be a useful addition.

Take-ups on all machines handling
film have always been a problem. At
the time our present printers were
built we were more satisfied with the

Sutton, Curtis and Thompson: 16mm Prints

149

constant tension cloth belt type of take-
up, as used on several projectors, than
any other we had ever used. In general,
these have been very satisfactory. How-
ever, torque motors have been used on a
number of take-up mechanisms in the
last few years, and new printers are
now being built with torque motors as
take-ups. Results indicate that these
will be even more satisfactory.

Light-Change Devices

Light changes are made in this printer
by a resistance type of board (Fig. 1).
We realize that, theoretically, a re-
sistance board should not be used in
making density correction in color
prints, but we have both types of light-
change devices in our laboratory, as
will be described later. Experience has
shown time and time again on tests we
have conducted that, for all practical
purposes, there is no difference between
the changes made with a resistance board
and those made with a neutral density
type of correction. For that reason,
we have continued to use this type of
system on the step printers. A change of
illumination, or a change of materials
or processing in the future may make
this statement void.

Originally, drop-type light boards
were used with all the difficulties en-
countered with such boards. When it
was necessary to replace these we thought
it desirable to make a number of changes,
and so another type of light board was
built. The idea is not new; but, on
the other hand, we do not believe that
these boards are generally available on
the open market. A piece of 35mm
positive film is punched and used in the
mechanism to actuate the light changes
(Fig. 3). Since there is not enough
room on a piece of 35mm film to make
enough punches to allow for 1 8 different
light changes — which is the standard
we use — it was necessary that we make
a punching machine that would punch
these light changes in code. By looking
at one of these pieces of film it can be

seen that the first six light changes are
made by simply punching a hole in the
proper place for numbers one to six,
but number seven light change is one
and two, number eight is one and three,
etc. By using this code system, it has
been possible to get all the light changes
on the strip of 35mm film. In addition,
we have room left over for several holes
which can be used to add automatically
the corrective filters to the light beam
at the same time the light changes are
being made, if such changes are de-
sirable. Such a system of making color
correction is not in general use in the
field as yet, although most laboratories
have some method of doing this if the
occasion demands it. As color process-
ing becomes more refined and as other
new materials are added for duplicating
purposes, we feel the time will come when
color corrections will be desirable and
probably necessary.

Such a light board has a number of
advantages over the conventional drop-
type board. There is no limit to the
number of light changes that can be
made in one reel of film. In other
words, a hundred changes can be placed
in one 400-ft film if necessary. This
system is highly desirable when printing
long lengths of film. Such a system
also means that once the film has been
correctly cued, it is impossible for the
operator to set up the board incorrectly.
Furthermore, it eliminates hours of
wasted setup time. With this system
it is only necessary to thread the strip
of film into the light-change mechanism,
turn it up to a point where a signal light
comes on, showing that it is in proper
position to print, and then proceed to
print. This cuing strip is kept with the
original film at all times and, in the
future, when a print is wanted, all that
is necessary to set up the board is to
thread in the strip and proceed to print.
Built into the light board is a voltage
regulator which automatically keeps
the voltage level constant. A variac is
also included in the circuit so that small

150

February 1952 Journal of the SMPTE Vol. 58

Fig. 3. This punch is for making the light-change cue strip and also color-correction
changes. The upper row of keys is for correction.

variations in normal light can be made
in order to correct small changes in the
filter pack, etc. By measuring this
normal light with the photoelectric cell
circuit, previously described, it is possible
to keep the printing normal quite con-
stant.

Built into the printer head is an air-
blast mechanism which constantly blows
against the printing aperture, thus
keeping it free of dirt and lint which
might otherwise accumulate.

This step printer, we believe, meets
the specifications we outlined previously
and can be used for making Kodachrome
prints, reversal prints, dupe negatives
and, with the proper aperture, black-
and-white positive prints. Such a
printer is necessary for a small quantity

of prints from one original and for special
purposes, such as the making of dupe
negatives.

The Multimatic Printer

There is, however, another problem
that we do not feel the step printer
answers as it should. This is the problem
of large-quantity print orders from the
same original. For this purpose we have
designed a continuous-type printer which
is known as the multimatic printer (Fig.
4). This is a three-headed printer
which was originally designed for making
color sound prints with optical effects
and light changes, automatically. The
machine runs in both directions and once
it is threaded with the proper optical-
effects mattes and density-change mattes

Sutton, Curtis and Thompson: 16mm Prints

151

Fig. 4. The Multimatic printer threaded for making sound Kodachrome prints with

optical effects and light changes. The optical-effects and density mattes

are on separate rolls on this setup.

it is not unthreaded again until the
prints are finished or the originals are
taken off for cleaning. The operator
simply stops the machine at the end of
each print, threads on more raw stock,
and makes another print going back in
the opposite direction. This way, there
is never any rewinding of originals. In
addition, the machine has the advantage
of being able to use odd lengths of film
which are a problem in Kodachrome
printing. The printer may be backed
up at any point, utilizing odd lengths of
raw stock. Once these have been re-
turned from processing they can be cut
in at the proper point and spliced to-

gether, thus using raw stock with a
minimum of waste.

This machine has been built to handle
1200-ft rolls of original and raw stock,
and runs at 72 ft/min in either direction.
Light-change boards for such a machine
would complicate the job and probably
give a considerable amount of trouble.
For this reason, we made a special
density matte containing the light
changes which are run along with the
optical-effects mattes, thus producing
the desired effects and light changes in
the print. These density mattes are
made on a Bell & Howell Model J
printer which has been remodeled for

152

February 1952 Journal of the SMPTE Vol.58

Fig. 5. Modified Bell & Howell Model J printer for making density mattes.
The cue film is threaded around the gate.

this work (Fig. 5) . Such a matte system, of
course, does not change the color tem-
perature of the lamp. As we have said
previously, prints made by this method
do not show any particular difference
from those made with the resistance type
of board on Kodachrome film as we
know the process today. The printing
on this machine is done on a 40-tooth
sprocket which has been designed to
accommodate film shrinkages up to
1.5%. Contact is maintained with a
rubber roller at the printing aperture.
Here again the problem of curl in Koda-
chrome Duplicating Film made it

necessary to provide this rubber roller
for consistent operation. The shoe type
of contact which is generally used on
continuous printers was too critical in
adjustment and too hard to keep in
adjustment to be satisfactory.

Printing is done by contact on the
40-tooth sprocket with the mattes
printed optically from the opposite side
of the sprocket below. The filter pack
is placed between the objective lens
and the printing aperture. Beneath
the matte aperture, and enclosed in the
lamphouse, is a right-angle prism which
turns the light up from a horizontal

Sutton, Curtis and Thompson: 16mm Prints

153

source. Thus, the lamp occupies a
normal upright position. Beside the
objective lens, there are three condensing
lenses. As in the step printer, the origi-
nal film does not touch metal at any
point, so the chance of scratching the
original is at a minimum. Many
scratches or cinch marks are caused in
rewinding original material. Since the
originals are not rewound between
prints when they are made on this
printer, the danger is largely eliminated.
The printer was designed to make a
large number of Kodachrome prints
from a single original, so that second-
generation prints would not have to
be used as originals. We do not know
how many prints can be made from one
original on this machine, because we
have never made a large enough number
to find out. We have printed over
600 Kodachrome prints from one origi-
nal, and from all appearances a good
many hundreds more could be made
from it. This does not, of course, mean
that that many prints ( could be made
from any original, because we frequently
receive originals which are in bad shape
before we ever start printing them.
However, when the originals received
for printing are in good shape and good
splices have been made, we have had
almost no trouble in making as many
prints from them as any customer might
want.

The multimatic printer is also suitable
for making prints from dupe negatives
and sound tracks. When a printer is
used for this purpose only two heads are
used — one picture head and the sound
head. Of course, the optical-effects
mattes are not used because both the
optical effects and the light changes
have been incorporated into the dupe
negative. Nearly all black-and-white
release prints are now made by using a
dupe negative from original reversal
black-and-white or color, and then
printing on fine-grain positive. Black-
and-white reversals are used on only a
few special orders.

The third type of printer which we
must use is the sound printer. Prints
made on contact step printers have the
sound added from a Maurer optical
printer. Before the dimensional charac-
teristics of sound-film base were
stabilized, this type of printer was
highly desirable as it would handle
originals with various degrees of shrink-
age. When the multimatic printers were
built, provisions were made for printing
the sound optically. However, tests
at that time did not indicate any ad-
vantage would be gained by this method,
and still other tests made over a period
of years have confirmed this point.
These tests were made in our own
laboratory and in other laboratories,
using the optical system of sound
printing. Therefore, the sound on the
multimatic is printed by contact.

An Optical-Effects Printer

At the present time, we are putting
into operation a new machine which
was designed to be used with the multimatic
and step printers. This is known as the
Curtis Automatic Effects Printer (Fig. 6).
This machine is an optical-effects printer
to be used for printing the optical-effects
mattes. Up until the present time all
these mattes have been edited and spliced
to the picture by splicing together optical
effects with black and clear film.
Instead of making up a matte in this
manner, we now punch both edges of
the workprint with cue marks and the
workprint is used to cue the optical-
effects printer. The printer is then
loaded with positive film, turned on, and
it automatically prints the mattes with
wipes, fades and dissolves, all in one
piece of film. As soon as this film has
been developed it is then ready for
checking and printing. In addition
to printing the optical effects onto this
piece of film, we can add a density
wherever necessary so that when we
have a final matte for printing it will
make both the optical effects and the
light changes.

154

February 1952 Journal of the SMPTE Vol 58

Fig. 6. Optical-effects — density-matte printer. The light-change board is threaded
with a punched 35mm strip for selecting the proper printing exposure for the density
matte. A second 35mm punched strip (hardly visible in photograph on right top of
printer) is used as a selector for the proper optical effect.

How does the machine operate?
Basically, it is built around a peculiar
arrangement of movements that control
the effects blades. Each effect, with
the exception of the straight wipes, has
a set of these movements which combine
the four standard effects of that type
through selective triggers into a single
unit with only one set of movements.
Since the straight cut is not considered
an effect, the omission here of a set of
movements is not another exception,
even though the dowser, responsible for
the straight cut, is also used as an integral
part in every effect.

First among the parts involved in
each set of movements is a planetary-

gear type of clutch which is geared
direct to the main driving shaft of the
machine. Upon a given signal it is
released for a single cycle at a rate
proportional to the number of film
frames involved in the effect being
driven. In turn, an eccentric pin,
geared to the clutch, drives a movement
known as the Walschaert gear. The
object of the Walschaert gear is to allow
admission to a new source of driving
power, emanating from a double-acting,
rotary- type solenoid. This quickly re-
verses the effect back to its normal
position or, if the reverse operation is
involved, out of its normal position
before or after the planetary-gear clutch

Sutton, Curtis and Thompson: 16mm Prints

155

has released its cycle of motion, let us
say, causing the effect to close. The
solenoid, acting through the Walschaert
gear, quickly opens the effect back to
its original position. But, before the
solenoid can work, the dowser chops
off the light, holding its closed position
until the next signal reverses the con-
tinuity of operation and opens it. Each
set of movements is extremely versatile.

During the closed position of the
dowser, or any other blade for that
matter, no light can reach the film.
Hence, the film will be transparent dur-
ing this period following its development
and will allow printing to be done
through it. Therefore, the image of a
closing blade actually opens a scene,
while the image of an opening blade
closes the scene. This commonly under-
stood inversion is only one among many
encountered in the machine. The
opening and closing of each sequence
is continuous throughout a film and is
the means by which release printing
can be done, subsequently, at first one
gate and then another without the show
of a splice. The dowser is the only
automatic instrument in the machine
that needs to be positioned before a run,
and that is done in the course of pre-
selection. All other effects hold a
normal open position when not in use.

The mechanical movements just de-
scribed entail considerable electrical
equipment which includes a signaling
system. Its manipulation throughout
one run is made easy, however, by simply
edge-notching the workprint for each
effect and/or timing change as desired
and by perforating two 35mm films as
selector strips. Since two combination
mattes (timing and effects) are required
for each show, both edges of the work-
print are notched as A and B, respec-
tively, and two more selector strips
perforated to match for the second run.
The use of both edges of the workprint
avoids making an extra cue film and
retains the advantages of notching to an
actual picture continuity.

Preselection is not relegated to any
one department or person. It is an
accumulative process, developed over
the preparatory printing route. The
signal originates with the workprint
because, when edited, the splices between
scenes represent the absolute, with
reference to the desired effect, if any,
penciled on the film in code. This
eliminates the script from further use
in finishing the picture. When the
originals are being critically scanned for
timing in the laboratory, the estimate of
correction needed for each scene is
recorded on a cue sheet in terms of
light-change numbers for future use
in punching the selector strips. The
workprint is also included on the
same gang synchronizer, and is marked
for edge notches which coincide with
the timing tabulations entered on the
cue sheet. Actual notching, however, is
done later when the workprint is returned
to editing with the cue sheet. Here the
effects and timing continuities are
matched so that a single notch will
accommodate both as often as possible.
Also, the effects cues are tabulated on
the same cue sheet which is then sent
back to the laboratory where it is used
in punching the selector strips. Finally,
the strips are brought together with the
workprint at the machine for the run.

To insure proper placement of the
notches, two standard Bell & Howell
notchers have been cut down and
mounted side by side into a single unit
(Fig. 7). The two blades, marked A
and B, respectively, face one another
across one film path with edge guides
intact. In this path a single pilot pin
is mounted between the blades to insure
positive registration to the notch in
relation to the sprocket holes. A gradu-
ated scale is mounted off to the left
and points out where the splice between
scenes shall be placed when notching for
each type of effect. Furthermore, the
scale translates the effects code, penciled
on the film at the splices, into the par-
ticular selective station numbers con-

156

February 1952 Journal of the SMPTE Vol.58

. *>. ;+ ' -*

Fig. 7. Special punch for notching cue film, or workprint,
for Curtis effects— density-matte printer.

cerned. Index buttons on the A and B
edges of the notching block are manually
pushed back and forth to remind the
operator of the particular track (A or B)
a scene or sequence is relegated to as the
alternations proceed, otherwise the pic-
ture continuity of the workprint is
obviously without this information. The
location of timing notches depends only
upon the marks penciled on the film in
timing, which may be at the splice with
certain opticals or apart within the
scene proper. Because timing is included
in each combination matte, there are no
edge noches in the originals, which in-
advertently serves to circumvent the lack
of a notching standard as discussed
previously.

In this respect, it is well to note that,
while the entire process is really quite
simple, there is a definite technique in-
volved in both preselection and machine
operation that readily lends itself to a
minimum of schooling among the person-
nel taking part.

The size and arrangement of* the per-
forations in both selector films provide
room at each step for ten selective
stations, positioning being equal to the
spacing of the sprocket holes. This
gives the workprint notch control over
20 selective stations. Of the ten timing

stations only six are used to control a
wide range of light changes, while of the
ten effects stations only eight are at
present in use, leaving an ideal situation
for future development. Preselection
is done in the laboratory, as previously
stated, on a punching machine, the key-
board of which resembles a typewriter.

Timing is arrived at through coupled
resistors in series with the lamp, selection
for intensity being directed through a
relay system interlocked with the various
taps in the resistor. There are 18
different light-intensity levels available,
each arrived at as a plus or minus rela-
tive of the ninth level, which is manually
preset through a variac. Since timing
requires a finer gradation of light, the
lamp is located on the emulsion side of
the raw stock which is separated from
the constant-burning effects lamp,
located on the base side. However,
the effects light also requires accurate
setting because of the fade effect which
is a form of the photographic wedge.
Both lamps converge their rays along
the axis through separate optical systems
onto the same gate, which is so con-
structed that it has an aperture on each
side of the film. Since certain effects
are made optically, such as the wipes,
there is an objective lens on the effects

Sutton, Curtis and Thompson: 16mm Prints

157

side which makes a camera of that part
of the printer. This cameralike head
has a dissolving shutter for the fades
and a dowser for both the straight cuts
and effects auxiliary. Beyond this head,
large condensing lenses spread the light
field for the effects. The timing system
has only a small condenser lens, but
there is a density filter pack included.
Both lighting systems have separate
voltage regulators, transformers for low-
voltage lamps, manually controlled
variacs, voltmeters and ammeters.

The machine is of the step type and
its product will eliminate the separate
light-change matte which was previously
used on the multimatic printer.

Experience has shown that most of
the optical effects wanted today are
fades, dissolves, right and left wipes,
and up-or-down curtain wipes. The
printer has been built to provide these,
automatically, on signal. Any special
wipe can, of course, be cut into a printed
matte if necessary. But, if this is done,
a separate density matte will be required. *
We will be able to use this matte with
the light changes on step printers. Once
a picture has been set up for printing,
in this manner, it can be printed on

* Since this paper was originally written,
a method has been discovered which per-
mits other types of wipes to be printed
directly.

either type of printer with the same matte
producing the opticals and light changes
in the final print. When printed optical
effects are used with the step printers,
density light-change boards are un-
necessary. This, we feel, will be an
advantage because it will eliminate any
mis-lights. We also feel that the elimi-
nation of splices in the mattes will be a
distinct advantage.

We do not necessarily believe that this
system of printing is suitable for every
1 6mm laboratory, but it has been success-
ful for us. Neither do we think it is the
final answer, because new products will
probably change some of our methods.
However, we believe that we know how
we can make conversions on present
printers. And, where conversions will
not work, we have ideas on how different
types of printers can be built for new
processes which have not yet been
developed to where they may be intro-
duced commercially.

References

1. L. Sherwood, "Editing and photo-
graphic embellishments as applied to
16-mm industrial and educational mo-
tion pictures," Jour. SMPE, 41: 476-
493, Dec. 1943.

2. P. S. Aex, "A photoelectric method for
determining color balance of 16-mm
Kodachrome duplicating printers," Jour.
SMPE, 49: 425-430, Nov. 1947.

158

February 1952 Journal of the SMPTE Vol. 58

High-Constant-Speed
Rotating Mirror

By J. W. BEAMS, E. C. SMITH and J. M. WATKINS

The rotating mirror is magnetically suspended in a high vacuum and spun
by a rotating magnetic field. The mirror is accelerated to full speed in a way
similar to that of the armature in an induction motor, but at running speed
it performs as an armature of a synchronous motor. The frequency of the
rotating field is determined by a piezoelectrically controlled circuit. Also it
is free of hunting. The maximum rotational speed of the mirror is determined
only by the strength of the mirror. Mirrors are described which rotate at
20,000 rps.

I

N A GREAT MANY problems, where it
is necessary to study accurately phenom-
ena which occur in very short intervals
of time, it is desirable to have a high-
constant-speed rotating mirror.1-2 It
is particularly important that not only
the number of revolutions per second of
the mirror must be known with high
precision, but the mirror must be free
of so-called hunting or rapid variations
in speed. This latter requirement of
freedom from hunting is usually almost
impossible to attain in practice, es-
pecially where the friction on the mirror
or bearings requires that the drive
deliver considerable power, i.e., when
the frictional torques and the driving

Presented on May 2, 1951, at the Society's
Convention at New York, by J. W. Beams,
E. G. Smith and J. M. Watkins, Rouss
Physical Laboratory, University of Vir-
ginia, Charlottesville, Va. The research
project upon which this paper is based
was supported by Contract NOrd-7873
with the Navy Bureau of Ordnance.

torques are large, small asymmetries
in either give rise to hunting of the rotor.
In the rotating mirror arrangement
described in this paper, the total fric-
tional torque is very small with the result
that the speed can be made extremely
constant and hunting, if present, is too
small to be observable.

Experimental Arrangement

Figure 1 is a schematic diagram of
the apparatus, while Fig. 2 is a photo-
graph of the suspended mirror with the
vacuum chamber and one drive coil
removed. This arrangement is the
outgrowth of a series of experiments,
using magnetically suspended rotors or
centrifuges in a vacuum, carried out at
the University of Virginia over a number
of years.3"7 The mirror R made of
high-strength ferromagnetic material is
suspended inside a glass vacuum cham-
ber by the axial magnetic field of the
solenoid S situated above the chamber.
The vertical position of the rotor is

February 1952 Journal of the SMPTE Vol. 58

159

PUMP

w

Fig. 1. Schematic diagram of high-constant-
speed rotating mirror arrangement.

maintained by the automatic regulation
of the current through the solenoid S,
while its horizontal position is deter-
mined by the symmetrically diverging
magnetic field. The mirror R is spun
by two pairs of coils K which produce a
rotating magnetic field. The small coil
Q is part of a tuned grid-tuned plate
radiofrequency oscillator (Fig. 3) which
regulates the current through S. It is
so arranged that when the rotating
mirror rises, the current through S
decreases, while when it falls, the current
in S increases in such a way as to main-
tain the mirror at the desired height
without observable hunting. The steel
cylindrical core C of the solenoid S is
suspended by a small wire W from the
adjustable support P. The core C is
surrounded by a damping fluid as shown
and serves to damp any horizontal
motion of the rotor.

Suspending Circuit

The circuit, which automatically regu-
lates the current through the solenoid S
in such a way as to maintain the rotor

Fig. 2. Suspended mirror, with

vacuum chamber and one drive coil

removed.

at the desired vertical position, is shown
in Fig. 3. The pickup coil Q is in the
grid circuit of a 5-mc partially neu-
tralized tuned grid-tuned plate oscillator.
If the rotating mirror R moves down-
ward and approaches the coil Q, the
latter's impedance, with the proper
setting of the oscillator, is changed in
such a way as to lower the amplitude
of oscillation in the circuit. This gives
rise to a so-called error signal which is
detected by a cathode-follower detector
and appears as a reduction in potential
across the resistance Ri2. A portion of
this potential change appears on the
grid of a 6SJ7 which is one-half of a
two-pentode mixer. Subsequently, this
signal increases the potential on the
grids of the three 6L6's in parallel,
which increases the current through the
solenoid S and in turn raises the rotating
mirror R.

In order to prevent vertical oscillation
of the rotor R the "error" signal is
differentiated by the resistance RH-
capacity Q combination and mixed
with the original error signal. Also,

160

February 1952 Journal of the SMPTE Vol. 58

0
lO

0 1
lO

M
tO

-/VVW— —

»

-CD-

1

01
(0

If)
lO

MMA,

L-vwv-l

IO

IT)

->
CD

N-
C/)

H

o
o

CO
CD

H

o

CD

5 =

u

r -

£

oe

r-.00j

Cl

-^

I

I

^

^

i: -s

Beams, Smith and Watkins: Rotating Mirror

161

the use of two 6SJ7 pentodes as mixers
together with the negative feedback
through resistor Ri3 and condenser C9
produces increased stability. The power
supplies were of the conventional regu-
lated degenerative type.8 The regula-
tion of the 500-v supply is less critical
than that of the 300-v supply so the
latter is stabilized from the former. A
variation of from 100 v to 135 v in the
line voltage produces less than a 10-v
change in the 500-v supply which in
turn produces less than 0.02 v in the
300-v supply. The — 375-v supply is
obtained from a conventional trans-
former rectifier with condenser input
filter system and stabilized with two
VR 150's and one VR 75 in series.

The solenoid S consists of 25,000 turns
of No. 28 insulated copper wire wound
on a bakelite frame. Its inductance is
19.5 h (henrys) and its resistance, 1010
ohms. The cold-rolled steel, cylindrical
core C of the solenoid (y^- in. in diameter
and 3^ in. long) is suspended by a
-g^-in. length of hardened 0.018-in.
diameter piano wire W. The height
of the core C is adjusted with a brass
plunger P which fits into a brass disk A.
The disk, which slides on the frame F,
is adjusted by setscrews to the proper
axial position so that the core remains
approximately on the axis of the solenoid
S when the current is raised to maximum
value. The length of the core C and
wire W are adjusted so that the period
of the pendulum so formed is approxi-
mately that of the rotor S when given a
horizontal displacement. The core hangs
in a "dash pot" (a glass test tube flat-
tened at the lower end and filled with
SAE No. 10 motor oil) and damps any
horizontal motion of the rotating mirror
R. No motion of the rotor either in a
horizontal or vertical direction can be
detected by a SOX microscope focused
on the scratches of the suspended mirror.

Rotating Mirror

For greatest stability of magnetic
support it is desirable (although not

absolutely necessary) to make the rotor
as long or longer in the direction of the
axis of spin than in the radial direction.
On the other hand, for rotational sta-
bility, the moment of inertia around the
axis of spin should be larger than that
around the radial or perpendicular
direction. Added to this, the rotor
should be symmetrical around the axis
of spin. It was found that a sharp cone
on top of a short cylinder proved to be
a very stable configuration. The faces
of the mirror were ground on the
cylindrical surface and the sharp cone
concentrated the magnetic flux in the
proper way to give stability. The
edges of the top and bottom of the
cylindrical portion were slightly beveled
to prevent discontinuities (resulting from
the mirror faces) from affecting the
pickup coil Q.

The first mirrors were made of mag-
netic stainless steel (Carpenter 2B
stainless 400). They were machined to
shape and then heat-treated by the
standard procedure to give good mirror
surfaces and high strength. They were
next ground to exact shape and the
mirror surfaces lapped and polished.
They were flat to roughly 0.2 wave-
length of sodium light. Rotors of 0.5-
in. diameter with mirror faces -|- in. X
i in. were used successfully for long
periods at 16,000 rps, but exploded at
1 8,500 rps. As a result the stainless steel
has been replaced by hard high-strength
alloy steel with the mirror faces covered
with a very thin coating of aluminum.
Ball bearings ground to the proper
shape were found to be satisfactory when
care was taken not to remove the temper
during the grinding process. The mirror
used at 20,000 rps was 0.5 in. from the
bottom to tip of the cone and each of the
six mirror faces was 0.25 in. in diam-
eter. The first type of mirror is
shown in Fig. 2. The rotating mirror
was surrounded by an all-glass vacuum
chamber with an optically flat glass
window, through which the light passes,
sealed on with low-vapor-pressure vac-

162

February 1952 Journal of the SMPTE Vol. 58

uum cement or wax. The chamber was
evacuated by a standard forepump,
diffusion-pump, cold-trap arrangement.

Driving Circuits

A schematic diagram of the drive
circuit is shown in Figs. 4 and 5. The
drive frequency is determined by a
piezoelectric crystal-controlled electron-
coupled oscillator operating at a fre-
quency of 100,000 cycle/sec (Fig. 4).
The crystal operates in a thermostat-
controlled oven to improve stability.
The oscillator is calibrated by zero-
beating the 100th harmonic with the
10-mc wave broadcast by radio station
WWV of the National Bureau of
Standards. The oscillator may be
tuned over a very narrow range and, in
practice, set to give the lowest practical
beat frequency. This procedure allows
the oscillator frequency to be determined
to about one part in 108. However, the
published precision of WWV is only
five parts in 108, so that when radio
transmission irregularities are con-
sidered, the precision of the oscillator
is not known to perhaps better than one
part in 107. In practice, the oscillator
circuit is operated for long periods of
lime and the drift is extremely small.
If it becomes necessary to determine the
frequency to better than one part in
107, it will be necessary to have a
laboratory standard.

The output of the buffer amplifier of
the oscillator is fed to a multivibrator
frequency divider. The output of the
multivibrator is a square wave of fre-
quency \/n X 105 cycle/sec, where n
is an integer. The divider was designed
for n = 5 or 6, i.e., frequencies of 20
kc or 16f kc, but other division ratios
are easily obtained. This square wave
is fed through an amplifier which serves
as a filter. The resultant sine wave is
passed through a phase-splitter and
buffer-amplifier. The output (Fig. 5)
is then amplified and transformer-
coupled to the power tubes which operate
as class C amplifiers with the drive coils

resonant with the proper capacitors as
the plate load.

The speed is measured by a method
shown schematically in Fig. 6. Light is
reflected from the mirror faces into a
photomultiplier cell. This signal is
amplified and applied to one pair of
plates of an oscilloscope. The com-
parison frequency is applied to the other
pair of oscilloscope plates so that the
resultant Lissajous figure gives the fre-
quency relationship. The comparison
frequency was usually a standard audio-
frequency oscillator except at operating
speed, where the drive frequency or
WWV was used as a comparison.

Operation

The procedure in starting the rotating
mirror is to turn on the crystal oscillator
in the drive circuit several hours before
operation so that it will have sufficient
time to reach thermal equilibrium. In
the meantime, the pumps are started
and the chamber surrounding the rotor
evacuated to 10~6 mm Hg pressure or
below. The mirror is then supported
and the power applied to the driving
circuit. In practice the support circuit
approaches equilibrium in a relatively
short time. The rotating field produced
by the two pairs of coils K (Fig. 1)
induces eddy currents in the mirror and
it starts spinning. Consequently, the
mirror acts as a high-resistance armature
of an induction motor and continues to
accelerate.

When the mirror speed approaches
within about 40 rps of the frequency
in the coils K, the rate of acceleration
falls off, but if the pressure in the vacuum
chamber is below 10~6 mm Hg the rotat-
ing mirror will continue to accelerate
until its rotational speed approaches
closely enough to the frequency of the
rotating magnetic field to "lock in."
When this occurs, the rotating mirror
operates as an armature of a synchronous
motor and spins without observable
hunting at a rotational speed equal to
the drive frequency. Consequently,

Beams, Smith and Watkins: Rotating Mirror

163

164

February 1952 Journal of the SMPTE Vol.58

270 V

250 V 500 V

6F6

6F6

- 105 V

Fig. 5. Drive amplifier.

LIGHT

Fig. 6. Scheme of speed-measuring method.

the rotational speed of the mirror is
known with the same precision as that
of the master driving oscillator. Usually
it requires more time to accelerate the
rotating mirror the last 40 rps than to
bring it up to this speed since the torque
falls off very rapidly as the "slip"

becomes small. As a result, it is usually
advantageous to disconnect the crystal
oscillator from the phase-inverter and
substitute an audio oscillator during the
acceleration period. In this way the
drive frequency is set at 50 or 60 cycles
above the desired running speed. When

Beams, Smith and Wat kins: Rotating Mirror

165

the speed of the mirror slightly exceeds
the desired running speed, the audio
oscillator is disconnected and the crystal
control substituted. The mirror then
decelerates slowly and "locks in."
When the mirror first "locks in" it
hunts with a considerable amplitude,
but in a few minutes this damps out and
becomes too small to observe (less than
10~8 radian/sec). Since the rotor speed
is over 105 radian/sec the error intro-
duced by hunting is less than one part
in 108.

With the circuit of Fig. 5 and a power
input to the coils K of 1 50 w or 1.6 amp
in the coils, the mirror accelerated at
the rate of approximately 1 000 rps/min
until the "slip" frequency became about
50. However, with this much power
input it is necessary to cool the coils
with a small fan. On the other hand,
when running speed is obtained, the
power in the drive coils should be con-
siderably reduced. The temperature of
the mirror increases a few degrees
during the acceleration period if the
power input is not greater than indicated
above. At running speed the rotor
temperature decreases slowly to prac-
tically that of the surrounding walls.
By removing the driving torque and
permitting the mirror to "coast" freely,
the deceleration is found to be extra-
ordinarily small. As a matter of fact,
the measured deceleration can be ac-
counted for as due only to the friction
of the residual gases surrounding the
rotor. As a result, in order to bring
the mirror to rest, it is necessary to
reverse the direction of the rotating
magnetic field and drive it down,
otherwise it would take a very long time
for the rotor to come to rest.

The above rotating-mirror arrange-
ment is especially useful when phenom-
ena which occur in very short periods
of time must be studied with precision.
It was developed for photographing the
successive stages of sparks in different
gases and the various stages of vacuum
sparks. Also, it is being applied in a

study of the velocity of light through
liquids as a function of the wavelength
of the light. Due to the high precision
with which the rotational speed is
known (one part in 107) and its freedom
from hunting, the arrangement is almost
ideally suited to the measurement of the
velocity of light in a vacuum. However,
for highest precision, the light path
should be of the order of a mile in length
and this distance is very difficult to
measure and maintain with a precision
of one part in 107. The maximum
rotational speed of the mirror is limited
only by the mechanical strength of the
mirror. Consequently, by reducing the
size of the rotating mirror higher speeds
can be obtained. At the present time,
a rotating mirror which spins at 105 rps
is under development.

Acknowledgment: It is indeed a pleasure
to acknowledge the valuable help of
Dr. P. B. Buck during the initial stages
of this work.

References

1. J. W. Beams, "Spectral phenomena in
spark discharges," Phys. Rev., 35: 24-33,
Jan. 1930; "The propagation of
luminosity in discharge tubes," ibid.,
36: 997-1001, Sept. 1930; "High
rotational speeds," /. Applied Phys., 8:
795-806, Dec. 1937.

2. G. D. Miller, "The optical system of the
NACA 400,000-frame-per-second motion
picture camera," U.S. NACA ARC No.
E6C25, 1946; U.S. NACA TN No. 1405,
Aug. 1947.

3. F. T. Holmes and J. W. Beams, "Fric-
tional torque of an axial magnetic
suspension," (letter to the Editor),
Nature, 140: 30-31, July 1937.

4. F. T. Holmes, "Axial magnetic sus-
pensions," Rev. Sci. Instruments, 8:
444-447, Nov. 1937.

5. L. E. MacHattie, "Production of high
rotational speed," Rev. Sci. Instruments,
12: 429-435, Sept. 1941.

6. J. W. Beams, J. L. Young and J. W.
Moore, "The production of high
centrifugal fields," /. Applied Phys., 17:
886-890, Nov. 1946.

166

February 1952 Journal of the SMPTE Vol. 58

7. J. W. Beams, "High centrifugal fields,"
/. Wash. Acad. Sci., 37: 221-244, July
1947.

8. W. C. Elmore, "Electronics for the
nuclear physicist — IV," Nucleonics, 2:
50-58, May 1948.

Discussion

M. L. Sandell: If you wanted to slow
up the rotor faster than happens directly
from friction, could you do it by reversing
the field?

Dr. J. W. Beams: Yes, that is the best
way of doing it.

E. Salzberg: I would like to know whether
the techniques you have developed in
supporting a rotating object have found
any application in industry or commerce?

Dr. Beams: Well, I don't know. This
is a research tool as far as I know. Besides,
in the spinning of mirrors, I think probably
it will be very useful as an aid in producing
a new type of centrifuge. I believe that
it is going to allow us to increase the pre-
cision of the measurement of molecular
weights, especially of the proteins.

The magnetic suspension used for
supporting the mirror in these experiments
may be slightly modified to make it into
an excellent magnetic balance. We have
succeeded in weighing weights of one
milligram with a precision of about one
billionth of a gram. This, of course, may
find considerable use in industry.

Mr. Salzberg: Would it be possible to
eliminate the use of the vacuum in rotating
at relatively low speed?

Dr. Beams: Yes. However, the air
friction goes up pretty rapidly with rotor
speed.

Kenneth Shajtan: What material do you
use?

Dr. Beams: We are using steel mostly.
The rotor is made of the best steel we can
get. We made some experiments on the
bursting of different steels and we ran a
long series on ordinary commercial ball
bearings and on selected ball bearings.
It turned out the ball bearings burst at
the same peripheral speed if made of the
same material. There were a great
many flaws in the larger ball bearings.
The probability of the rotor going up to
full speed was roughly inversely propor-
tional to the diameter of the rotor. I

think that this result can be explained
metallurgically.

Anon: What was the measurement be-
tween the solenoid field and the rotor
itself?

Dr. Beams: Do you mean what distance?

Anon: Yes.

Dr. Beams: All the way from a few
millimeters to 6 or 8. It is a variable
thing, depending upon the field in the
solenoid and its gradient at the rotor
position.

Anon: What order of magnitude of power
inversely is required to spin the bearing
rotor?

Dr. Beams: Now, this is a relative
matter, of course. I had one that was
small, near •£$ in. in diameter, which
started spinning slowly when the light from
a Western Union electric arc was focused
on its periphery. In other words, the
light pressure was sufficient to spin it.

In this rotating mirror we had 1.6 amp
to the coil and it accelerated at the rate
of 1000 rpm. We try in most of our
experiments to bring the rotor up as slowly
as we can ; by accelerating it faster, more
heat is generated in the rotor. But under
about one ampere in the coil the rotor
increases in temperature less than 10°.

A. W. Carpenter: In bursting ball bear-
ings could you tell me offhand within
what angle it proved to be splaying or
clipping?

Dr. Beams: Well, they sort of powdered
and completely disintegrated. One also
notices a little yellow light, like on a
grinding wheel. You, of course, look
through a right-angle mirror to see the
yellow light.

E. A. Andres, Sr.: If I understood you
correctly, you said you had 1.6 amp
accelerated at 1000 rps...

Dr. Beams: No, 1000 rpm.

Mr. Andres: I would like to know how
you made the measurement.

Dr. Beams: By photoelectron multiplier
tube and a light-beam arrangement.

C. D. Miller: Dr. Beams, as you know
at NACA we used a system similar to the
one developed by you for supporting and
driving a rotor used in a camera with
which we took pictures at speeds up to
800,000 frame/sec. We used a rotor
weighing about two-thirds of a pound,

Beams, Smith and Watkins: Rotating Mirror

167

about three inches long and about an
inch in diameter.

I was interested in your remarks about
the heating effect. We were not able to
get an extremely good vacuum, as you
have, because of certain mechanical
limitations involved in our optical system.
Because of the consequent high slip and
resulting eddy currents, we ran into very
serious heating of the rotor.

We eliminated the heating by resorting
to what I call a self-synchronous motor.
We cross-magnetized the rotor and drove
it up to a few revolutions per second as
an induction motor. Then, with two
small coils alongside the lower end of the
rotor, 90° apart, we picked up a four-
phase voltage induced by the cross magne-
tization. We amplified this four-phase
pickup, through both voltage and power
amplifiers, and fed the output into the
driving coils. We adjusted the positions
of the pickup coils so that the rotating
field was a little ahead of the cross magne-
tization of the rotor. The rotor then
accelerated as a synchronous motor, and
we avoided the heating altogether.

Dr. Beams: Yes. Yours was a very
beautiful experiment. The method you
used was certainly a good one. We have
had to use a similar sort of scheme where
we cannot have any temperature rise.
The only reason we did not do it here is
that the small mirrors do not get too hot.
On the other hand, for larger rotors this
is necessary.

Mr. Miller: I was wondering whether

you found that the cross magnetization of
the rotor would cause any undesirable
effects in your experiments.

Dr. Beams: No, the cross magnetization
seems not to upset anything else.

Anon: Mr. Miller, how much tempera-
ture rise did you encounter in the rotor
when attempting to drive it up to full speed
as an induction motor?

Mr. Miller: I did not measure the
temperature rise except by touching the
rotor with the hand. It was obviously
excessive.

R. O. Painter: I wonder why the sup-
porting field does not introduce eddy current
flow. As I have it, there would be eddy
current loss caused by this field since it fans
out in the rotor.

Dr. Beams: Well, you see the magnetic
field comes down uniformly across the
rotor since the latter has a high permea-
bility. Hence, there is no current flow.

Mr. Painter: Is it not generating eddy
currents in the rotor periphery? You have
a radial magnetic field.

Dr. Beams: You have a radial electrical
field as it works out in practice. On the
other hand, you have no closed circuit for
the current unless the spin axis of the rotor
makes a sizable angle with the direction
of the magnetic field.

Mr. Painter: Between the center and the
outside?

Dr. Beams: There is an electrical po-
tential between the center and periphery
of the rotor, but no current can flow.

168

February 1952 Journal of the SMPTE Vol.58

Report of SMPTE
Standards Committee

By FRANK E. CARLSON, Committee Chairman

JL HE STANDARDS COMMITTEE has con-
tinued to function with the type of
organization and in accordance with the
policies described in the preceding
report.1 This, the final report of the
present Committee, includes not only a
review of the work of the past two years,
but also observations regarding the
organization and policies of the Com-
mittee in the light of past experience.

Organization

The current practice of naming the
Chairmen of the several Engineering
Committees as members of the Standards
Committee has proven quite satisfactory
and it is recommended that this be con-
tinued. The objectives sought in ap-
pointing to this Committee the Chairmen
of ASA sectional committees having
interests closely related to the motion
picture industry have not been realized,
possibly because the activities of those
committees during this period did not
happen to bear on subjects of interest
to motion pictures. In any event, since
an important part of the related fields
is represented in the Photographic
Standards (Correlating) Committee, it
seems desirable to reconsider the im-
portance of such appointments. Par-

Submitted as of December 27, 1951, by the
Society's Standards Committee Chairman,
Frank E. Carlson, General Electric Co.,
Nela Park, Cleveland 12, Ohio.

ticipation by the Motion Picture Re-
search Council and the few members-at-
large has been commendable although,
in the case of the MPRC, it was some-
times felt that the Committee would
benefit if it were better informed of the
Council's standards activities and in-
terests.

Policies

The present practice of publication
for trial and criticism, reviews, approvals,
and reapprovals of proposed standards
is different from the practices in many
other and related fields. Unquestionably
such thoroughness serves a useful pur-
pose, but it must also be conceded that
it adds to the Society's cost for processing
standards and, in large measure, dupli-
cates work which is the logical assignment
of Sectional Committee PH22 of ASA.
Since this Sectional Committee is spon-
sored by the SMPTE, and its member-
ship is reviewed and approved by the
Board of this Society, it is suggested
that this present duplication of re-
sponsibility and effort be studied.

Coordination of Photographic
Standards Work in ASA

Early in 1950 the Standards Council
of ASA authorized the formation of a
Photographic Standards (Correlating)
Committee and, in accordance with ASA
procedure, delegated to that Committee

February 1952 Journal of the SMPTE Vol. 58

169

general administrative and supervisory
responsibilities in this field. Prior to
this time all proposed photographic
standards were classified in the "miscel-
laneous" group and, like other miscel-
laneous standards, were identified by
numbers which included the prefix letter
Z. Formerly a standard approved by
the old Sectional Committee Z22 had
to be referred to the ASA Board of
Examination which in turn had to refer
it to the full Standards Council con-
sisting of over 70 members. Since the
formation of the Correlating Committee
all proposed photographic standards
(as well as revisions of old standards)
are identified by the prefix PH. These
proposals from one or another of the
new Sectional Committee for photog-
raphy go directly to the Correlating
Committee and then to a six-man Board
of Review for final approval. Thus,
the formation of the Photographic
Standards (Correlating) Committee has
made possible substantial savings in
both time and money.

It will be noted that, in subsequent
sections of this report, standards or
proposals are identified by Z22 numbers
in some cases and PH22 numbers in
others. This obviously reflects the
change in organization just described.
In the future all motion picture standards
will be identified by the prefix PH22
as new standards are completed and old
ones reviewed. Other photographic
standards (formerly identified by Z38
numbers) will, in the future, be identified
by the prefix PHI, PH2, PH3, or PH4,
depending upon which of the four other
new Sectional Committees for photog-
raphy sponsored the proposal.

Standards Completed in 1950-1951

The following ten standards have
been processed since the last report and
have been adopted by ASA:

Z22. 7-1950: Location and Size of Pic-
ture Aperture of 16mm Motion Picture
Cameras2

Z22.8-1950: Location and Size of Pic-

ture Aperture of 16mm Motion Picture
Projectors2

Z22. 19-1950: Location and Size of
Picture Aperture of 8mm Motion Picture
Cameras2

Z22. 20-1950: Location and Size of
Picture Aperture of 8mm Motion Picture
Projectors2

PH22.7 1-1950: Cutting and Perforat-
ing Dimensions for 32mm Sound Motion
Picture Negative and Positive Raw Stock3

PH22.72-1950: Cutting and Perforat-
ing Dimensions for 32mm Silent Motion
Picture Negative and Positive Raw Stock3

PH22.73-1951: Cutting and Perforat-
ing Dimensions for 32mm on 35mm Motion
Picture Negative Raw Stock4

PH22. 74-1951: Zero Point for Focusing
Scales on 16mm and 8mm Motion Picture
Cameras4

PH22. 76-1951: Mounting Threads and
Flange Focal Distances on 16mm and 8mm
Motion Picture Cameras4

PH22.82-1951: Sound Transmission of
Perforated Projection Screens6

Additionally, Z22. 78-1 950, Mounting
Frames for Theater Projection Screens,2
was adopted by ASA but not processed
by the Standards Committee. This
standard was developed by a subcom-
mittee of ASA Sectional Committee
Z22.

Similarly, the following three stand-
ards adopted by ASA were developed
by the Joint SMPTE-MPRC Committee
on Test Films:

Z22. 79-1950: 16mm Sound Projector
Test Film2

Z22. 80-1950: Scanning-Beam Uni-
formity Test Film for 16mm Motion
Picture Sound Reproducers (Laboratory
Type)'

Z22.8 1-1950: Scanning-Beam Uni-
formity Test Film for 16mm Motion
Picture Sound Reproducers (Service Type)*

The difficulties encountered in at-
tempting to process a standard for 16mm
and 8mm sprockets were described in
the preceding report.1 Accordingly, the
Committee has published7 an SMPTE
Recommendation for 16mm and 8mm
Sprocket Design for the guidance of
sprocket designers. This material is in

170

February 1952 Journal of the SMPTE Vol. 58

a format such that it can be included in
the Society's Standards Binder.

The Standards Committee has also
completed its work on the following four
proposals which have been submitted
to ASA with the recommendation that
they be adopted as American Standards:

PH22.11: 16mm Motion Picture Pro-
jection Reels8

PH22.83: Edge Numbering of 16mm
Motion Picture Film9

PH22.24: Splices for 16mm Motion
Picture Films for Projection10

PH22.77: Splices for 8mm Motion
Picture Film10

Standards Currently in Process

PH22.15: Emulsion and Sound Record
Positions in Camera for 16mm Sound
Motion Picture Film11

PH22.16: Emulsion and Sound Record
Positions in Projector for Direct Front
Projection of 16mm Sound Motion Picture
Film11

Both of the above are proposed re-
visions of Z22.15-1946 and Z22.17-1947,
the most important detail of which is
elimination of reference to the "guided
edge." As sometimes happens in a
case such as this, additional suggestions
for improvement of the revision have
been received with the result that a
revised revision of the proposal is
scheduled for republication shortly.

PH22.86: Dimensions for Magnetic
Sound Tracks on 35mm and 17£mm
Motion Picture Film12

PH22.87: Dimensions for Magnetic
Sound Track on 16mm Motion Picture
Film12

PH 22.88: Dimensions for Magnetic
Sound Track on 8mm Motion Picture
Film12

These badly needed proposals are
the work of the Subcommittee on Mag-
netic Recording of the Sound Committee
and the comments which have resulted
from preliminary publication are now
being reviewed by that Committee.

Z22.75: A and B Windings of 16mm
Raw Stock Film With Perforations Along
One Edge1 »

This proposal, originally an SMPE
Recommendation adopted in 1941, has
given the 16mm and 8mm Motion Pic-
tures Committee a great deal of trouble.
It was first published as a proposed
standard in September, 1949; the pres-
ent revision of the proposal, which
appeared in January, 1951, has brought
forth suggestions for further changes,
with the result that it has been again
referred to the sponsoring Committee.

PH22.84: Dimensions for Projection
Lamps, Medium Prefocus Ring Double-
Contact Base-Up Type3

PH22.85: Dimensions for Projection
Lamps, Medium Prefocus Base-Down
Type3

These two proposals, developed by the
16mm and 8mm Motion Pictures Com-
mittee, seem to be about ready for final
action by the Standards Committee on
the question of submittal to ASA.

PH22.1: Cutting and Perforating Di-
mensions for 35mm Motion Picture Film —
Alternate Standards for Either Positive or
Negative Raw Stock14

The history of this proposal since 1 932
is briefly set forth in the Journal and is
an example of the complexity of the
problems that frequently confront the
Film Dimensions Committee.

Other proposals on the Agenda of the
Standards Committee which have not
yet reached the stage of publication for
trial and comment include the following:

Revision of Z22.41-1946, Sound Rec-
ords and Scanning Area of 16mm
Sound Motion Picture Prints, again
with particular reference to the question
of "guided edge." The Sound Com-
mittee is considering this question to
determine what can be done to establish
consistency with related standards with-
out degradation of 16mm sound quality.

Aperture Calibration of Motion Pic-
ture Lenses, a proposal developed by the
Optics Committee, has encountered
strong criticism in the initial ballot of
the Standards Committee on the question
of preliminary publication.

F. £. Carlson: Standards Committee Report

171

Enlargement Ratio for 16mm to
35mm Optical Printing is a new pro-
posal developed by the Laboratory
Practice Committee and will probably
appear in an early issue of the Journal.
[See the Jan. 1952 Journal.]

16mm Motion Picture Projector for
Use With Television Film Chains
Operating on Full-Storage Basis. This
is a proposal developed by the joint
RTMA-SMPTE Committee on Tele-
vision Film Equipment and is encounter-
ing opposition on the Standards Com-
mittee initial ballot.

Under ASA procedure, existing stand-
ards are re-examined periodically for
the purpose of determining whether the
Standard should be reaffirmed in its
present form, be revised in the light
of new developments or changing prac-
tices, or rescinded because it is no longer
of value. The responsibility for such
review is delegated to the several
Engineering Committees and two of
them have recently submitted recom-
mendations to the Standards Committee
for further action. These include:

Nomenclature for Electrical Filters,
Z22.33-1941. The Sound Committee
has recommended that this Standard
be discontinued because it finds that the
method described for designating elec-
trical filters has had very little use and
has no further value.

Emulsion Position in Projector for
Direct Front Projection of 16mm Silent
Motion Picture Film, Z22. 10-1 947.

Emulsion Position in Projector for
Direct Front Projection of 8mm Silent
Motion Picture Film, Z22.22-1947.
Here too, the Committee on 16mm and
8mm Motion Pictures recommends that
these standards be discontinued because
films for projection are produced by such
a variety of processes that the Standards
are no longer of value.

Finally, the Engineering Vice-Presi-
dent has asked the Standards Committee
to assist in another effort to develop a
glossary. As a first step in this program,,
work done in the early 1940's on this
subject is being subdivided by the
Society's Staff Engineer into parts
corresponding to the scopes of the several
Engineering Committees. Thus, work
already done need not be duplicated by
those Committees. It is expected that
this initial work will be completed in time
for the several Committees to begin
work early in their new terms.

References

1. "Report of the Standards Committee,""

Jour. SMPTE, 54: 102-105, Jan.
1950.

2. "New American Standards," Jour*

SMPTE, 54: 494-507, Apr. 1950.

3. "Standards," Jour. SMPTE, 56: 235-

246, Feb. 1951.

4. "Three New Standards," Jour*

SMPTE, 56: 684-689, June 1951.

5. "New American Standards," Jour*

SMPTE, 57: 170-171, Aug. 1951.

6. "New American Standards," Jour*

SMPTE, 55: 117-119, July 1950.

7. "Recommendations for 16mm and

8mm Sprocket Design," Jour*
SMPTE, 54: 219-228, Feb. 1950.

8. "Proposed American Standard, 16mm

Projection Reels," Jour. SMPTE*
54: 229-231, Feb. 1950.

9. "Edge Numbering of 16mm Motion

Picture Film," Jour. SMPTE, 56:
115, Jan. 1951.

10. "Splices for 16mm and 8mm Film,'*

Jour. SMPTE, 56: 354-361, Mar,
1951.

11. "Revision of PH22.15 and PH22.16,"

Jour. SMPTE, 56: 559-561, May
1951.

12. "Proposed American Standards," Jour.

SMPTE, 57: 71-74, July 1951.

13. "A and B Windings of 16mm Raw

Stock Film With Perforations Along
One Edge," Jour. SMPTE, 56:
112-113, Jan. 1951.

14. "Proposed American Standard," Jour*

SMPTE, 57: 275-278, Sept. 1951.

172

February 1952 Journal of the SMPTE Vol.58

71st Semiannual Convention

The Spring Convention at The Drake in Chicago on April 21-25 is well planned already
by many good hands, some of whom were noted in the report in the January Journal.
BUT now is the time for all good authors to hasten information about their papers for
the Convention to the proper authority.

If you don't have an Authors' Form or can't readily get one from one of the Papers
Committeemen listed in the January Journal, ask the Society's headquarters office for one.
AND in the meantime, not merely in posthaste but by wire, advise the 71st Convention
Program Cochairman on the spot in Chicago:

Telegraph: George W. Colburn, 164 N. Wacker Drive, Chicago 6, 111.

The sooner you send word, the easier will be the work of arranging the program in the
form of sessions, which is the job of the Program Cochairmen, R. T. Van Niman and
George Colburn.

Convention Vice-President Bill Kunzmann gave a detailed report of plans for the
Convention to the Society's Board of Governors in January, and from that report and later
information from Bill, as well as from C. E. Heppberger, we have the following roster of
the folks who will put on the Chicago Convention :

Program Cochairmen — R. T. Van Niman and George W. Colburn

Local Arrangements — C. E. Heppberger

High-Speed Photography — Richard O. Painter

Hotel Reservations and Transportation — W. C. De Vry

Luncheon and Banquet — George W. Colburn

Membership and Subscriptions — Ray Gallo and Samuel R. Todd

Motion Pictures Program — L. E. Weber, assisted by R. J. Sherry

Projection, 16mm — E. W. D'Arcy

Projection, 35mm — I. F. Jacobsen, assisted by Officers and Members of Local 110,

IATSE

Public Address and Recording — Robert P. Burns
Publicity — Harold Desfor and Leonard Bidwell
Registration and Information — James L. Wassell, assisted by E. W. D'Arcy, J. E.

Dickert, Steve Hunter, C. L. Lootens, K. M. Mason, John S. Powers and Reid H. Ray
Television — William C. Eddy
Ladies' Registration — Mrs. George W. Colburn and Mrs. C. E. Heppberger, Cohostesses,

assisted by the wives of the Central Section's Officers and Managers

Early in March, all members will receive the Advance Notice of the Convention, which
will contain a condensed schedule of the Convention sessions and will have attached the
usual tear-off postal for making hotel reservations. Bill Kunzmann has received from
John R. Bogardus, Front Office Manager, The Drake, Lake Shore Drive & Upper Michi-
gan Ave., Chicago 11, 111., the following rates:

Single room, per day $5.50; 6.00; 6.50; 7.00; 7.50.

Double room with twin beds, per day .... $9.00; 10.00; 11.00; 12.00; 14.00; 15.00.
Suite parlor and one bedroom, per day .... $16.00; 18.00; 19.00; 26.00; 33.00; & up.

173

Board of Governors Meeting

The Society's Board of Governors held its
first 1952 meeting on January 24, in New
York City. This is the meeting at which
the members of the Board examine the
previous year's operations, comparing
carefully the report of actual performance
for last year against the budget that was
set up in January 1951.

EXECUTIVE COMMITTEE

Most significant administrative develop-
ment of 1951 was the appointment by
President Peter Mole of an Executive
Committee which will assume some of the
operational advisory functions formerly
exercised directly by the Board of Gover-
nors. Growth of the Society's business
and the importance of a number of its
activities in the almost daily evolution of
television and the current technical growth
of motion pictures call for closer super-
vision of the headquarters operations than
the Board of Governors could provide
with its regular quarterly meetings. As a
consequence, the Executive Committee
will meet monthly, or more often if neces-
sary, consult with the staff, examine the
precise details of operation on a month-
to-month basis and will submit recom-
mendations for consideration by the Board
of Governors when that body meets every
third month. Under this arrangement,
matters of general policy and questions of
membership or industrial services provided
by the Society can receive proper Board
attention while details of execution will
in general be left to the discretion of the
executive body.

ENGINEERING

Test film operations were placed on a
somewhat more secure footing when Fred
Whitney joined the headquarters staff
during the early part of 1951 and took
charge of test film quality control. Pre-
cision required by American Standards
or by specifications developed by Society
committees for motion picture test films
demands careful supervision of production.
In addition, development of certain new
test films through the coming year will
necessitate not only agreement on the
manner in which the films are produced

but also agreement on standard methods
of testing. It will be Mr. Whitney's re-
sponsibility to spell out these test methods
in detail and submit them for considera-
tion by the Test Film Quality Control
Committee.

PUBLISHING

The report of the Editorial Vice-Presi-
dent pointed out Journal changes made
during 1951 which have been considered
a marked improvement. The new style
provides more words per page and the
combination of type face, line length and
line spacing contributes to easier reading
and yields more printed words per pub-
lication dollar than was possible in the
older format. With the new format now
well in hand, the major objective for 1952
is the adoption of more realistic publica-
tion dates. The first three issues for 1952
will probably be somewhat thinner than
usual and it is expected that the May
issue will be out by the 1 5th of that month.

The actual cost of producing the twelve
copies of the Journal which each member
receives annually is a figure that has been
quite difficult to arrive at, considering the
recent major changes in accounting, the
manner of operating Society Headquarters
plus the format changes of the past year.
Since things are now settling down, the
Board asks that Headquarters prepare
such a cost analysis with a view toward
determining whether or not each member's
dues pays his share of the cost of operating
the Society. Figures that result from
pro rata allocation of costs depend, of
course, on membership. Unit costs go
down as the total number of members
goes up. That brought up the question
of membership solicitation activities for
last year and also for 1951.

MEMBERSHIP

A new committee, under the Chair-
manship of Ray Gallo of Quigley Publica-
tions and with Beatrice Conlon of the So-
ciety Headquarters as full-time Secretary,
is attempting something new in the way
of membership work. Between 65 and
100 company or city member-delegates
are being selected and each will be armed

174

with advance information about conven-
tions, Society committees, Section meetings
and membership and publications am-
munition. Each of these member-
delegates will be the focal point for Society
information in his own community. It is
hoped that many questions about the
SMPTE, its engineering activities and
membership requirements can be answered
on the spot, to the benefit not only of the
inquirer but of the Society as well.
SECTIONS

Reports of the three Section Chairmen
were read into the record and it was noted
that extra effort at organizing Section
meeting programs was almost invariably
rewarded by an increase of attendance,
entirely justifying the added costs of rented
chairs or screening-room facilities. Popular
reaction to the repeating of Convention
papers at Section meetings brought an
official request for the recording on mag-
netic tape of certain papers for re-presenta-
tion with accompanying slides. It was
suggested that a small library of such papers
be assembled and made available in
appropriate batches for regional meetings
or student chapter sessions. There was

Engineering Activities

also a formal recommendation that the
Student Chapters in Hollywood and in
New York should function under the super-
vision of the local Section Chairmen and
Boards of Managers. This would probably
result in the more efficient use of funds
and perhaps encourage Chapter participa-
tion in local Section meetings.

CONVENTIONS

The report of the Convention Vice-
President concluded, with some enthusiasm,
that the two Conventions held in 1951
had drawn better attendance than any
in previous years. As a consequence,
plans for the Spring and Fall Conventions
in 1952 are being adjusted to provide
facilities for the larger registration. The
following dates were reported to and
approved by the Board of Governors:

71st Convention: The Drake, Chicago,
111., April 21-25, 1952

72nd Convention: Hotel Statler, Wash-
ington, B.C., October 6-10, 1952

73rd Convention : Hotel Statler, Los
Angeles, Calif., April 26-30, 1953

74th Convention: Hotel Statler, New
York, N.Y., October 4-9, 1953

New Chairmen Engineering Commit-
tees are appointed in
accordance with Section V of the Society
Bylaws, which states that the term of
appointment expires every two years,
along with the term of the appointing
officer (the Engineering Vice-President)
and further that Committee Chairmen are
eligible for one reappointment, or for a
total service of 4 years. (There is no limit
on the reappointment of members, except
that imposed by their degree of interest in
the work of the committee.)

The four-year limitation now requires
that Fred Bowditch as Engineering Vice-
President appoint new chairmen to six
committees, four of whom are:
Standards, Henry Hood, Eastman Kodak
16 & 8, Malcolm Townsley, Bell & Howell
Sound, John Hilliard, Altec Lansing
Motion Picture Studio Lighting Process Pho-
tography, John W. Boyle

The two committees still without new
Chairmen are Color and High-Speed
Photography. It is expected that they
can be announced in the next Journal.

ISO The Technical Committee on
Cinematography of the Inter-
national Organization for Standardization
(ISO TC/36) was canvassed as to their
interest in a meeting in New York on
June 9 and 10, 1952, as mentioned in the
previous issue. Affirmative replies have
since been received from Belgium, Canada,
France, Germany, Italy and the United
Kingdom. Based on this response, the
Chairman of the ASA Sectional Com-
mittee PH 22, Dr. D. R. White, and the
SMPTE Engineering Vice-President, Fred
Bowditch, recommended to the ASA that
the meeting be scheduled. A proposed
Agenda is now being formulated, on the
recommendations of the SMPTE Engineer-
ing Committees, secured in anticipation
of such a meeting and including items
submitted by several of the member na-
tions. ISO procedure requires the Agenda
to be circulated to all members four months
prior to the meeting, in this instance by
February 9, 1952. As Secretariat the
United States will chair the meeting. —
Henry Kogdy Staff Engineer.

Book Reviews

Fundamental Mechanisms
of Photographic Sensitivity

(Proceedings of a Symposium held at the
University of Bristol in March 1950.)
Edited by J. W. Mitchell. Published
(1951) by Butterworths Scientific Publica-
tions, London. Distributed in U.S.A. by
Academic Press, 125 E. 23 St., New York
10. i-viii + 347 pp. + 270 illus. 7 X
9f in. Price $9.50.

This represents an excellent and up-to-
date review of data and theories on the
fundamental mechanisms of photographic
sensitivity. Original papers as presented
at an International Conference in Bristol,
England, in March 1950, have been as-
sembled in book form by the editor.

By arranging the papers in groups, such
as "Photographic Sensitivity" and "Latent
Image Formation," the editor has made it
convenient for the reader to follow the latest
trends and developments in these con-
cepts. Professor N. F. Mott, in an intro-
duction to the book, outlines the latent
image theory as proposed by him and
Gurney in 1938 and gives its present
status, pointing out the problems which
still need explanation. This introduction
will be helpful to those who are not too
familiar with the subject.

The book contains contributions and
first publications of papers from a large
number of European, British and American
scientists. It is interesting to note from
these papers that their various observations
and theories about latent image formation,
photographic sensitivity and optical and
chemical sensitization begin to dovetail
with the basic concept of Gurney and
Mott and the evolution of this theory by
concepts proposed by Pohl, Stasiw and
Teltow, West, Mitchell and others.

The book also contains a series of articles
under the general headings: "Physical
Properties of Silver Halide," "Production
and Properties of Silver Halide Grains in
Photographic Emulsions" and "Nuclear
Track Emulsion." A summary prepared
by the editor after the conference gives
a critical review of the status of the theory
of the physical properties of the silver

halides and the theory of photographic
sensitivity as it appeared to him.

The book will be of interest primarily
to those working in the field of photo-
graphic research and development; how-
ever, it should also appeal to those working
on practical applications of photography
and interested in knowing what makes
photography work. The book is well
printed and illustrated. — Herman H. Duerr,
Ansco, Binghamton, N.Y.

Einfuhrung in die wissenschaftliche
Kinematographie (Introduction to Scien-
tific Motion Picture Photography)

By Dr. Werner Faasch. In German.
Published (1951) by Verlag von Wilhelm
Knapp /Halle (Saale), Germany. 76 pp. +
63 illus. 5f X 8 in. Available in U.S.A.
from Stechert-Hafner, Inc., 31 E. 10 St.,
New York 3. Price $1.30.

At this time, when photography is being
recognized more and more as an indis-
pensable tool for scientific and technical
investigation, the appearance of any book
which surveys some part of the field should
not be ignored. This book is intended
merely as an introductory survey to the
applications of motion picture photog-
raphy as a means of scientific study.

The opening chapter is concerned with
time-lapse and high-speed motion picture
studies, and in particular with German
apparatus for use in these fields.

Among the special applications of motion
picture photography dealt with in succeed-
ing chapters are motion photomicrography,
x-ray and electron microscope motion
photography, endoscopic studies and photog-
raphy of operations, astronomical and
photoelastic studies, Schlieren photography
and a number of other applications. The
final chapter treats the important subject
of evaluation of the photograph.

The book would have its greatest appeal
to the general reader. It is confined to
known practices, and is devoted almost
entirely to German equipment. It is
well illustrated. — Walter Clark, Kodak
Research Laboratories, Rochester 4, N.Y.

176

New Members

The following members have been added to the Society's rolls since those last published.
The designations of grades are the same as those used in the 1950 MEMBERSHIP DIRECTORY.

Honorary (H)

Fellow (F)

Active (M)

Associate (A)

Student (S)

Angwin, Bruce S., Regional Sales
Manager, Equipment Tubes, General
Electric Co., Electronics Div. Mail:
238 North Frederic St., Burbank, Calif.
(A)

Avil, Gordon, Free-lance Motion Picture
Cameraman. Mail: 13809 Weddington
St., Van Nuys, Calif. (A)

Baldridge, Claude C., Motion Picture
Supervisor, U.S. Air Force, Edwards
Air Force Base, Edwards, Calif. (A)

Bisno, Lou, Production Assistant, Snader
Telescriptions Corp. Mail: 530 North
Frederic St., Burbank, Calif. (M)

Bridge, Harry P., University of Maine.
Mail: 109 Commercial St., Boothbay
Harbor, Me. (S)

Bryant, Harry L., Recording Engineer,
Radio Recorders. Mail: 4350 Chevy
Chase Dr., La Canada, Calif. (M)

Buxbaum, Morton L., New Inst. for Film
and Television. Mail: 357 Milford
St., Brooklyn 8, N.Y. (S)

Cain, Donald G., University of Minne-
sota. Mail: 5125 South Washburn,
Minneapolis 10, Minn. (S)

Chaikofsky, Samuel, New Inst. of Film
and Television. Mail: 2504 Bronx
Park East, New York 67, N.Y. (S)

Crevenna, Alfredo B., Writer and Motion
Picture Director. Ultramar Films.
Mail: Parque Melchor Ocampo 28,
Dep. 5, Mexico, D.F., Mexico. (A)

Curtis, Harold K., Foreman, Release
Dept., Paramount Pictures Laboratory.
Mail: 9021 Dicks St., Los Angeles 46,
^Calif. (A)

Dixon, Herbert W., Television Motion
Picture Production, John Sutherland
Productions, Inc. Mail: 11280 Brook-
haven Ave., Los Angeles, Calif. (M)

Downey, C. E., Television Engineer,
KGO-TV. Mail: 119 Villanova Dr.,
Oakland 11, Calif. (A)

Eckes, John D., Jr., Supervisor, Camera
Dept., United Productions of America.
Mail: 362 South Myers St., Burbank,
Calif. (A)

Elliott, Richard S., Senior Photographer,
Motion Picture Div., University of
Southern California at Los Angeles.
Mail: 550f South Barrington, Los
Angeles 49, Calif. (A)

Fallis, Marne F., Projectionist, United
Productions of America. Mail: 156
South Pacific Ave., Glendale 4, Calif.
(A)

Freeman, Howard E., Owner, H. E.
Freeman Co. Mail: 4517 Sepulveda
Blvd., Sherman Oaks, Calif. (A)

Gancie, Joseph J., School of Radio
Technique, Inc. Mail: 108 Central
Ave., Brooklyn 6, N.Y. (S)

Glennan, Gordon R., General Manager,
Sound Services, Inc. Mail: 802 North
Martel Ave., Hollywood 46, Calif.
(M)

Governor, Frank, School of Radio Tech-
nique, Inc. Mail: 74 Irving PI.,
New York, N.Y. (S)

Graff, Earl F., Assistant Manager, Pem-
brex Theatre Supply Corp. Mail:
10540 Pangborn Ave., Downey, Calif.
(A)

Hall, Robert D., Manufacturer, Projection
screens and equipment, Commercial
Picture Equipment, Inc. Mail: 1567
West Homer St., Chicago, 111. (M)

Harris, Sgt. William J., AF 13234996,
Motion Picture Sound Recording and
Projection, U.S. Air Force. Mail: 731
Franklin Cir., Portsmouth, Va. (A)

Hoffman, Wendell L., Manager, Photo-
graphic Laboratory, University of Ne-
braska. Mail: 5019 Walker Ave.,
Lincoln, Nebr. (A)

Jacobs, Harry N., Television Engineer,
KGO-TV. Mail: 1600 Merced St.,
Richmond, Calif. (A)

Jones, Merwin C., Television Engineer,
KGO-TV. Mail: 270 El Bonito Way,
Millbrae, Calif. (A)

Kelly, Peter J., Motion Picture Camera-
man (Documentary), Shell Film Unit.
Mail: White Barn Hotel, Cuddington,
North Northwich, Cheshire, England.
(A)

Kenik, Marvin, SRT Television Studios.
Mail: 348 E. 19 St., New York 3, (S)

Koshlaychuck, William E., Supervising
Editor, Commercial Dept., Telenews,
Inc. Mail: 6817 Owls Head Ct.,
Brooklyn, N.Y. (M)

Lehman, John Francis, Syracuse Uni-
versity. Mail: 115 College PI., Syra-
cuse 10, N.Y. (S)

Le Vino, Richard B., Chief, Televisual
Equipment Section, Signal Corps Engi-
neering Laboratories, Coles Signal Labo-
ratory. Mail: 36 Riverside Ave., Red
Bank, N.J. (M)

Lunt, Mack G., Cinetechnician, Pembrex
Theatre Supply Corp. Mail: 636 Ditt-
maiz Dr., Whittier, Calif. (A)

177

McLaren, Norman, Animation Producer,
National Film Board of Canada. Mail:
520 St. Patrick St., Ottawa, Ont.,
Canada. (A)

Micllef, Edgard Roger, School of Radio
Technique, Inc. Mail: 492 Third St.,
Brooklyn 15, N.Y. (S)

Morgan, Kenneth, Physicist, Inter-
chemical Corp. Mail: 45-14 — 30
Ave., Long Island City 3, L.I., N.Y.
(M)

Morrison, Arnold, Self-employed, Film
Producer. Mail: 68 Fifth Ave., New
York, N.Y. (M)

Mueller, Arthur C., Design Engineer,
Bell & Howell Co. Mail: 1637 Sher-
man PI., Des Plaines, 111. (A)

Mullin, John T., Electronics Development
Engineer, Bing Crosby Enterprises, Inc.
Mail: 1351 Kelton Ave., Los Angeles
24, Calif. (A)

Nicholson, Donald S., Technical Assistant
to Director of Studio Operations,
Technicolor Motion Picture Corp.
Mail: 1216 Oak Circle Dr., Glendale 8,
Calif. (A)

Nopper, C. G., Chief Engineer, WMAR-
TV. Mail: 31 Dunkirk Rd., Baltimore
12, Md. (M)

Peterson, Richard S., School of Radio
Technique, Inc. Mail: 572 Amsterdam
Ave., New York, N.Y. (S)

Petrushansky, Yevsie S., Free-lance Pro-
ducer-Director. Mail: 5222 North
11 St., Philadelphia, Pa. (if)

Pruitt, Jerome, School of Radio Tech-
nique, Inc. Mail: 251 W. Ill St.,
Apt. 4B, New York 26, N.Y. (S)

Read, George W., Electronic Design
Engineer, Westrex Corp. Mail: 941
East Dryden, Glendale 7, Calif. (M)

Rice, John G., Electronic Engineer (Tele-
vision), Signal Corps Engineering
Laboratories. Mail: 4 William St.,
Red Bank, N.J. (M)

Rogers, John M., Laboratory Technician,
Commonwealth Film Laboratories, Pty.,
Ltd., 60 Wilton St., Sydney, N.S.W.,
Australia. (A)

Savage, Alfred D., Projectionist, Instructor
on Theater Television, Fred Wehren-
berg Theatre Circuit and Local 143.
Mail: 215 Eichelberger, St. Louis 11,
Mo. (A\

Sayers, Eric Russell, Executive Vice-
President, Agency Consultants, Inc.
Mail: 639 E. 11 St., New York 9, N.Y.
(A)

Seligman, Steven M., Film Editor, CBS-
TV. Mail: 40 W. 72 St., New York,
N.Y. (A)

Stubbs, William S., Photographer, Air
Reduction Sales Co. Mail: 556 Strat-
ford Rd., Union, NJ. (A)

Switzer, Israel, University of Alberta.
Mail: 10542-83 Ave., Edmonton, Al-
berta, Canada. (S)

Talamini, Arthur, Jr., Television Engi-
neer, A. B. Du Mont Laboratories, Inc.,
1000 Main Ave., Clifton, N.J. (M)

Thiebaux, M. L., Design Engineer, North
American Aviation. Mail: 14869
Janine Dr., Whittier, Calif. (A)

Tremblay, Louis R., Self-employed, High-
Speed Motion Pictures. Mail: 17146
Warrington Dr., Detroit 21, Mich.
(A.) ^

Waddington, Lester E., Radio-Television
Director, Miles Laboratories, Inc. Mail:
820 Edwardsburg Ave., Elkhart, Ind.
(M)

Wagg, Alfred, Self-employed, Camera-
man, Journalist and Film Director-
Producer. Mail: 3565 Martha Custis
Dr., Alexandria, Va. (M)

Walther, E. L., Sound Engineer, RCA
Photophone of Australia, Pty., Ltd.,
221 Elizabeth St., Sydney, N.S.W.,
Australia. (A)

Ward, H. Connell, Engineer, RCA Victor
Div., 1560 North Vine St., Hollywood,
Calif. (A)

Winchester, Ted, Assistant Head, Photo-
graphic Dept., RKO. Mail: 1704
South Canfield Ave., Los Angeles 34,
Calif. (A)

CHANGES IN GRADE

Conant, Russell W., Technicolor Motion
Picture Corp., 6311 Romaine St.,
Hollywood 38, Calif. (A) to (M)

Culley, Ray, President, Cinecraft Pro-
ductions, Inc. Mail: 21271 More-
wood Pkwy., Rocky River 16, Ohio.
(A) to (M)

Dickely, F. C., Sales Engineer, Altec
Service Corp., 2211 Woodward, Detroit,
Mich. (A) to (M)

Dickert, James E., Motion Picture Re-
cording and Production, Wilding Picture
Productions, Inc. Mail: 642 Ash St.,
Winnetka, 111. (A) to (M)

Eglinton, William, Head, Photographic
Dept., RKO Radio Pictures, 780 North
Gower St., Hollywood 38, Calif. (A)
to (M)

Fermaglich, Charles, Motion Picture
Producer, President, Empire Studios.
Mail: 618 Medical Arts Bldg., Houston,
Tex. (A) to (M)

Fulwider, Robert W., Patent Lawyer,
Partner, Fulwider Mattingly. Mail:
5225 Wilshire Blvd., Los Angeles 36,
Calif. (A) to (M)

Gaw, Ernest D., Service Inspector, Inter-
State Circuit, Inc. Mail: 830 Cherokee
Trace, Grand Prairie, Tex. (A) to (M)

178

Hart, William J., Motion Picture Sound dustrial Film Co. Mail: 919 M & M

Technician, Wright-Patterson Air Force Bldg., Houston, Tex. (A) to (M)

Base. Mail: 970 West Main St., Martin, Mahlon H., Jr., Owner, Audio

Wilmington, Ohio. (A) to (M) Visual Center. Mail: 1118 Lincoln

Hedden, William D., Laboratory Super- Way, East, Massillon, Ohio. (A) to (M)
intendent, The Calvin Co., 1105 Tru- Peck, Charles D., Manager-Owner, South-
man Rd., Kansas City, Mo. (A) to west Theatre Equipment Co., 118$
(M) West Douglas Ave., Wichita 1, Kan.

Landau, Alfred, Motion Picture Engineer, (A) to (M)

Columbia Pictures Corp. Mail: 344 Pope, Lucian E., Purchasing Agent,

Spalding Dr., Beverly Hills, Calif. (A) Fox Midwest Amusement Corp. Mail:

to (M) 2216 W. 49 St. Ter., Kansas City, Kan.

LaRue, M. W., Jr., Mechanical Engineer, (A) to (M)

Bell & Howell Co. Mail: 1225 Grove Sandback, Irving C., Optical Design

Ave., Park Ridge, 111. (A) to (M) Engineer, Bell & Howell Co. Mail:

Lowe, L. W., Self-employed, Producer, 3711 West Pratt, Lincolnwood, 111.

Lecturer. Mail: Box 89, Paola, Kan. (A) to (M)

(A) to (M) Souther, Howard T., Manager, Speaker

Macon, N. Donald, Industrial Motion Division, Electro-Voice, Inc., Buchanan,

Picture Producer, Owner, Texas In- Mich. (A) to (M)

Meetings

The Central Section of the SMPTE has scheduled two papers for its meeting at the
Bell & Howell Co., 7100 McCormick Blvd., Chicago, on March 27. Bruno G. Staffen,
development engineer of the Jensen Manufacturing Co., will describe a new low-cost
theater speaker system, and there will be a description of the new Bell & Howell magnetic
and optical 16mm sound projector by J. B. Weber, H. H. Brauer, F. J. Schussler and
M. G. Townsley. C. E. Heppberger is Central Section Chairman, and John S. Powers
is Program Chairman.

71st Semiannual Convention of the SMPTE, April 21-25, The Drake, Chicago

Other Societies

I.R.E. National Convention, Radio Engineering Show, Mar. 3-6, Hotel Waldorf-Astoria

and Grand Central Palace, New York

National Electrical Manufacturers Association, Mar. 10-13, Edgewater Beach Hotel,

Chicago, 111.

American Physical Society, Mar. 20-22, Columbus, Ohio
Optical Society of America, Mar. 20-22, Hotel Statler, New York
American Physical Society, May 1-3, Washington, D.C.
Acoustical Society of America, May 8-10, New York

American Institute of Electrical Engineers, Summer General Meeting, June 23-27,

Hotel Nicollet, Minneapolis, Minn.

American Physical Society, June 30-July 3, Denver, Cclo.

Photographic Society of America, Annual Convention, Aug. 12-16, Hotel New Yorker,

New York

American Institute of Electrical Engineers, Pacific General Meeting, Aug. 19-22, Hotel

Westward Ho, Phoenix, Ariz.

Illuminating Engineering Society, National Technical Conference, Aug. 27-30, Wash-
ington, D.C.

SMPTE Officers and Committees: The roster of Society Officers and the
Committee Chairmen and Members were published in the April 1951 Journal.

179

New Products

Further information about these items can be obtained direct from the addresses given.
As in the case of technical papers, the Society is not responsible for manufacturers' state-
ments, and publication of these items does not constitute endorsement of the products.

of reproduction is also excellent for musical
material. The disc drive uses no turn-
table, which, it is said, eliminates all
flutter and wow, enabling extended bass
response. The design of the feed screw
affords accuracy of pitch, it is said, as
great as that found in the most professional
recorders. The "Wagner-16" also features
"spiralling," and a "magic-eye" volume-
level indicator is an integral part of the
equipment. A recording time reference
scale operates automatically. Two input
receptacles, switch selected, are provided,
one for the microphone and the other for
"bridging." The equipment operates
from 115 volts, 60-cycle a-c, with 50-cycle
optional.

MicroDisc blanks are $2.50 for a package
of 12. The complete MicroDisc recorder
includes all styli (sapphire), microphone,
PM loudspeaker, extra input plugs and
instruction manual, at a cost of $295 from
the Audio & Video Products Corp., 730
Fifth Ave., New York 19, N.Y.

A playback recorder that puts 60 min of
speech or music on a vinylite disc 4f in.
in diameter, operating at 16 rpm and at
a pitch of 448 lines /in. is the "Wagner-16"
MicroDisc Recorder, Model PI 6-450.
The portable carrying case, weighing 28
Ib complete, contains the complete
mechanism and recording head, amplifier
and power supply, playback pickup and
loudspeaker for recording and playing
back.

Although it is recommended that the
prime application for MicroDiscs is in
reference, closed circuit and conference
work, the manufacturer claims that fidelity

Film Research Associates, located at
150 E. 52 St., New York 22, N.Y. publishes
and distributes seven guides to available
training aids. Listed are the sources of
functional films for meetings, conferences,
classes or study groups, with alphabetically
arranged descriptions of type, running
time and use. Procedures are recom-
mended for effectively using audio-visual
methods. The guides are priced sepa-
rately at $1.00 to $2.00 and the complete
set of seven publications is $9.00.

American Standards form the technical foundation for motion pictures around the
world. All current standards were listed by subject and by number in the Journal In-
dex 1946-1950. Reprint copies of this list, which includes all previous Journal refer-
ences to each standard, are available from Society Headquarters without charge.

Complete sets of all sixty current standards in a heavy three-post binder with the in-
dex are $13.50, plus 3% sales tax for purchases within New York City, and are avail-
able from Society Headquarters. Single copies of any particular standard must be
ordered from the American Standards Association, 70 East 45th St., New York 17, N.Y.

180

Image Gradation, Graininess and Sharpness
in Television and Motion Picture Systems

Part II : The Grain Structure of Motion Picture
Images— An Analysis of Deviations and
Fluctuations of the Sample Number

By OTTO H. SGHADE

CONTENTS

Symbols 182

Summary 182

A. Dynamic Fluctuations and Static Deviations 183

B. Physical Sampling Apertures for Random and Continuous Sampling Proc-
esses 184

C. Method of Evaluating the Effective Sampling Area of an Unknown Aper-
ture . 185

1. Units and Terminolc

spor
3. Evaluation of an Equivalent Passband JVe, and a Characteristic Aperture

logy
2. Aperture Response Characteristics

Dimension from the Sine-Wave Spectrum

4. Equivalent Aperture Diameters

5. Equivalent Passband and Aperture Diameter of Processes Containing a
Number of Elements in Cascade

D. Granularity and Random Fluctuations in Motion Picture Processes . ... 201

1 . Deviation Characteristics of Photographic Film

2. Sampling Apertures and Transfer of Deviations in Motion Picture Systems

3. The Signal-to-Deviation Ratio in Projected Positive Film

4. The Optical Passband of the Total Deviation in Projected Positive Film

5. Signal-to-Deviation Ratios and Gamma of Motion Picture Film for Tele-
vision Recording

6. Luminance Fluctuations and Optical Passbands of Motion Pictures

Presented by Otto H. Schade, Tube Dept., Radio Corporation of America, Harrison,
N.J., in part on April 20, 1950, at the Society's Atlantic Coast Section Meeting at New
York, and on May 3, 1951, at the Society's Convention at New York. Portions of this
paper have been made available to the Subcommittee on Distribution Facilities, Theater
Television Committee of the SMPTE, in the form of three reports: (1) "Random Fluctua-
tions in Television and Motion Pictures," June 7, 1950; (2) "Outline and Results of a
Study to Determine Television System Parameters Providing an Image Sharpness Equiva-
lent to a 35-Millimeter Motion Picture Process," May 15, 1951 ; (3) "Theater Television
Transmission Channel: Choice of Performance Factors," May 31, 1951.
Note: Part I of this paper, "Image structure and transfer characteristics," was published
in this Journal in February 1951, pp. 137-171.

March 1952 Journal of the SMPTE Vol.58 181

SYMBOLS

Note: Peak values are designated by a peak sign over a symbol,
and average or mean values by a horizontal bar, n.

a
A<z

D*
E

/
H

K

/

N

Ne

N, Ne *
_

N,

Ner

n

An

[An]
r0
r$
[R]

[R]T

[/?]*

Area of sampling aperture

Incremental portion of a

Mean value of response factor r$

in incremental section AN
Frame area
Luminance
A constant
Total density of photographic film;

D\ density of negative film, Z>2

density of positive film
Density above base density
Exposure (unit: meter candle sec-

onds)
Frequency; /(*, y) a function of x

and^y
Horizontal dimension of picture

frame
A constant
Unit of length
Line number = number of half-

wave lengths of line- or sine-wave

patterns per length unit
Limiting resolution
No Equivalent passbands (Eqs. ( 22 )

to (28))
Equivalent passband of an asym-

metric aperture (Eq. (23))
Rated resolving power of film

(r$ ^ 0.02)
Line number indicating aperture

diameter (Eq. (19))
Number of particles or samples

inside of sampling area
Deviation from average number n;

An = n — h
Rms value of deviation
Characteristic radius of an aperture
Sine-wave response factor (Eq. (18))
Signal-to-rms-deviation ratio static

value in a single image frame

(Eq. (13))
Signal-to-deviation ratio of film

transmittance (Figs. 49-52)
Signal-to-deviation ratio neglecting

lens flare and ambient light

[R]i3 etc. Signal-to-deviation ratio of a
process. The index 1 is used for
deviations originating in a nega-
tive film process; the index 2 for
a positive film process; the index
0 for a process (lens) preceding
the negative; the index 3 for a
process (lens) following the posi-
tive. An index 13 indicates
deviations originating in 1 and
observed after the processes 2
and 3.

[R]0 Luminance fluctuation ratio, dy-
namic value (Eq. (47))

s Length of side of square aperture,

or storage factor

Tf Frame time

T8 Storage time (Eq. (47))

u A characteristic length unit used

in aperture calculations

V Vertical frame dimension

x,y Coordinates, x = coordinate in the
direction of scanning

T Amplitude, intensity

y Constant gamma

7 Point gamma, definition in Part I,

p. 145. A single index number
indicates the process as stated
under [/2]is; a combination index
(713) indicates the product 717273.

5 Characteristic aperture diameter

(index system as for [R])

e Base of natural logarithm

T Transmittance

or Relative deviation (Eqs. (13) to

(17))

tf- Flux

[$] Rms value of variational (a-c) flux
(see Eq. (20))

$ Average (d-c) flux

\J/0 Zero "frequency" component of
flux (see Eqs. (18) and (21))

J/0 A light bias (see discussion of Eq.
(38))

SUMMARY OF PART II

An objective measure of the luminance
deviation caused by the random structure
in motion picture and television images
is developed, based on the distribution
and a count of the number of grains or
electron "samples" in a specified
sampling area. The "effective sampling
area" of the various components in

photographic or television systems is
determined from their response to sine-
wave test signals and specified by an
equivalent measure Ne. The accuracy
of this method is compared with a direct
evaluation from the dimensions of the
point image or resolving "aperture" for
which the geometrical properties are

182

March 1952 Journal of the SMPTE Vol. 58

known. The law of sample distribution
at the sources of the deviations in motion
picture systems is investigated and
methods are developed for computing
the relative deviation and the "fre-
quency" spectrum in the entire lumi-
nance range of the projected image as
modified by successive aperture-response
and nonlinear transfer effects. Nu-
merical evaluations of a number of
motion picture processes are carried
out and a discussion is included of
optimum conditions for video recording
and possibilities for improvements.

Part II is limited to a treatment of
aperture-response theory as applied
to the evaluation of the relative deviation
in motion picture processes. Part III,
to be published later, will treat the raster
effect, the apertures and the relative
deviation ("noise"-to-signal ratio) in
television systems, followed by an inter-
pretation of graininess which includes
the process of vision. The aperture
theory and use of the measure Ne will
be developed further in Part IV which
will deal specifically with methods and
measurements for evaluating resolution,
definition and sharpness of images.

A. DYNAMIC FLUCTUATIONS AND STATIC DEVIATIONS

Continuity of contours and uniformity
of tone values in an image are in reality
illusions created by a limited perception
of fine detail, an inability to count or
resolve the individual "samples" of
energy or matter forming the image.
This limitation can usually be removed
when a small area a is inspected under
high magnification. Suppose the num-
ber n of samples arriving in this
"sampling area" or passing through a
"sampling aperture" placed over a live
image is counted during a time unit.
The number n obtained in various
counts will be found to fluctuate around
an average value n which does not
change when the sampling area is moved
to different locations in a larger area
representing a uniform intensity. The
fluctuations in different positions of the
area a, however, are generally found to
be nonsynchronous. In a "live image"
from a lens, television tube or motion
picture film, these fluctuations in the
number of light samples are fluctuations
of luminance which give the impression
of a moving granular structure.

In most reproduction processes the
light samples in a live image are con-
verted, accumulated, and stored in
special image surfaces during a given
(exposure) time in the form, for example,
of silver grains or electron charges.

Such an image frame is, therefore, a
static record of the number and dis-
tribution of effective samples collected
during the exposure time. A single
"frame" does not show dynamic fluctua-
tions but exhibits static deviations in the
number of samples between sampling
areas.

The static deviation in a sample arrange-
ment representing a constant density
can be evaluated statistically by counting
the samples (grains, electrons) in a
sampling area a which is moved to
various positions, determining the mean
value n of the readings, and tabulating
the deviations in number Arc from the
mean value n. The rms value [An]
of the deviations is known as the standard
deviation. The ratio of the standard
deviation to the mean value is the
relative deviation <r; its reciprocal is the
"signal-to-deviation" ratio [/?]. The two
ratios are defined by the equation:

a = \I\R\ = [Aiil/n

(13)

When the samples in an image frame
are distributed with uniform probability
independently of one another, and when
the area occupied by a sample is small
compared to a, the average number of
samples counted in a sampling aperture
of uniform transmittance is proportional
to its area a

Otto H. Schade: Motion Picture Granularity

183

(14)

The rms deviation [An] is found to
follow the law

[An] = (n)»

(15)

and the relative deviation (Eq. (13)) for
this distribution is, hence,

a = («)-»

(16)

This relation changes inversely with the
square root of the sampling area:

<7<xa-i (17)

A distribution with these characteris-
tics will, henceforth, be referred to as a
random distribution. This distribution
can be verified by testing the validity of
Eqs. (15), (16) and (17) by a sampling
process. A grain structure containing
samples of several sizes has a random
distribution when the distribution for
each sample size is random as specified
above. The definition holds also for a
structure which can be considered as a
linear superimposition of a number of
sample layers, each having a random
distribution.

B. PHYSICAL SAMPLING APERTURES FOR RANDOM
AND CONTINUOUS SAMPLING PROCESSES

The random step-wise selection of
sampling positions in static-deviation
measurements (discussed in the Ap-
pendix to this Part) can be replaced by
a continuous displacement of the
sampling aperture. When, for example,
the grain structure of an illuminated
film of constant density is "scanned" by
an aperture, the light flux passing
through the sampling aperture into a
phototube is a relative measure of the
grain number n and causes a current,
fluctuating in amplitude around an
average value. The total current may
be shown by an oscillograph or recorded
by a microphotometer; its average value
(signal) as well as the rms deviation
(a-c component) can be measured
directly by suitable current meters.

In both step-wise or continuous devia-
tion measurements the total flux, i.e.,
the number of samples integrated or
"counted" within a sampling aperture
of uniform transmittance, is proportional to
the area of the aperture. The aperture
shape is, in principle, unimportant
(barring diffraction or diffusion) because
any odd-shaped aperture of uniform
transmittance can be assembled from a
number (£) of small apertures Aa of
equal size, which may be arranged in

any desired shape, f When the grain
distribution is random, as specified
above, there is no correlation or pre-
ferred sequence of grain occurrences
when the grain pattern is sampled or
scanned. A round sampling aperture,
however, is advantageous because it
reduces undesired optical effects (diffrac-
tion) to a minimum.

In contrast to the above, the shape of
a scanning aperture analyzing or forming
pictorial images or images of test charts
can be of importance, because these
images may consist of sample arrange-
ments with definite preferred locations
(straight or curved contours), according
to the subject material. Sampling
apertures with circular symmetry are
preferred for simultaneous imaging and
are used also as scanning apertures in
television systems.

Comparison of a graph of sample
readings taken at random from a grain
pattern with a microphotometer trace
shows that there is a considerable
difference in the sequence of values ob-

fThe relative deviation Jor an area a =
%Aa is simply a = <ri/\/£, with <n = rela-
tive deviation of Aa.

184

March 1952 Journal of the SMPTE Vol. 58

tained by random step-wise sampling
as compared to the conditioned sequence
of amplitudes in a waveform obtained
by a continuous displacement of the
same sampling aperture over the same
grain pattern.

The average value (d-c signal) of the

graph or waveform and the rms devia-
tion, however, are unchanged when a
sufficient number of sampling points
has been taken. The signal-to-deviation
ratio [R] can, hence, be determined
accurately by random or continuous
sampling with an aperture.

C. METHOD OF EVALUATING THE EFFECTIVE SAMPLING AREA
OF AN UNKNOWN APERTURE

The standard deviation has obviously
no meaning when the size of the sampling
aperture remains unknown or unspeci-
fied. When deviations are imaged,
i.e., transferred, through an imaging
system, the components (lens, film) of
the system cause an integration of
samples within the area of their point
image (star image). The point image
which has been identified as the "re-
solving aperture" of the device, repre-
sents, therefore, the sampling aperture
of the device referred to the image plane.
The equivalent area a of this aperture
must, hence, be known for all system
components in order to evaluate signal-
to-deviation ratios.

It is difficult and in many cases im-
possible to measure the sampling aper-
ture of a system component directly,
but it is relatively simple to analyze the
aperture effect in images made of suit-
able test objects. According to a Fourier
theorem the complex waveform obtained
by the scanning of a random structure
can be broken down into a continuous
spectrum of constant-amplitude sine-
wave components which, for this purpose,
can be regarded as having equal ampli-
tudes and random phase relation. This
sine-wave spectrum does not extend to
infinity, but the sine-wave amplitudes
may be assumed as constant over a wide
range of wavelengths usually extending
far beyond the limiting resolution of the
first sampling aperture. It is thus
expected that the integrating effect of
a sampling aperture on fluctuations

can be synthesized from the aperture
effect on single sine-wave components
determined with a series of test patterns,
each representing a sinusoidal flux pat-
tern of one constant optical sine-wave
length in the image field. When the
shape and passband of the sine-wave
response characteristic of the unknown
aperture are compared with those of
known aperture types (round, cosine
squared, exponential, etc.), an exact
or an equivalent aperture area can be
established for the particular device or
for an entire process. It appears ad-
visable at this point to define units and
terms, particularly the meaning of
optical sine waves, line numbers, and
passbands.

1. Units and Terminology

The conventional optical test pattern
consists of groups of sharply defined
adjacent bars, alternately black and
white and of equal width. The width
of the bars decreases from group to
group. These patterns are optical
square-wave flux patterns having decreas-
ing square-wave lengths. Variable-
density recordings of electrical sine-
wave signals on the sound track of a
motion-picture film accordingly repre-
sent optical sine-wave flux patterns (see
Fig. 40). Optical patterns always have
and must have two dimensions to be
useful for measurements with two-
dimensional apertures. For a study of
the sine-wave response of an aperture
in one direction, the test pattern must

Otto H. Schade: Motion Picture Granularity

185

o
o

10

CVJ

186

March 1952 Journal of the SMPTE Vol. 58

give a flux distribution varying sinu-
soidally in one direction of the field
but remain uniform in the perpendicular
direction, appearing to the eye as a
series of parallel dark and light bands
or lines (Fig. 40). When imaged by an
aperture (lens, film) or scanned by an
aperture (perpendicular to the lines)
the reproduced flux pattern or the flux
signal from the aperture is again a
pure sine-wave pattern or signal, but
with reduced amplitude which may be
computed or measured as a function
of the sine-wave length in the optical
test pattern. When measuring optical
wave patterns the unit is obviously a
length. Its reciprocal in analogy with
electrical terms, is an "optical frequency"
stating the number of waves per length
unit (not time). In television termi-
nology one full wave in the test field
consists of two half-waves, the positive
half-wave being identified as a light
"line" and the negative half-wave as a
dark "line," thus leading to the
definition: the line number JV specifies
the number of half -waves per reference
length. The television unit will be used
throughout this paper unless stated
otherwise. The television unit is smaller
by a factor of two than the photographic
unit which identifies one line with one
complete wavelength, f

The sine-wave flux response is specified
in relative units by the sine-wave flux
response factor r$, defined as the ratio of
the sinusoidal aperture flux $N at a
line number JV to the sinusoidal aperture
flux i£0 at a line number JV approaching
zero as a limit.

The response factor n£ is single valued
and independent of the test-pattern contrast

f When the number of lines in a back-
ground other than black or white is speci-
fied, the photographic definition of count-
ing only one of the two distinct lines in the
pattern appears less descriptive than the
television definition which applies to all
cases.

provided waveform distortion by non-
linear transfer characteristics is avoided
by restricting the "signal" amplitudes
(contrast) from the test pattern to
appropriate values. The optical pass-
band of the aperture is the range of line
numbers from JV = 0 to JV = JVC in
which the aperture response r$ decreases
from unity to zero. It is emphasized
that, strictly speaking, this passband
describes the aperture response in one
direction only, the direction of scanning.
Apertures of circular symmetry have but
one response characteristic and passband
for all directions, while asymmetric
apertures (squares, slits, etc.) have
many response characteristics and optical
passbands depending on the aperture
orientation relative to the direction of
displacement. Description of an asym-
metric aperture by its aperture response
requires at least two characteristic
passbands (vertical and horizontal)
which for some purposes can be replaced
by the passband of an aperture with
circular symmetry and equivalent area
a (see Sec. 3 below).

2. Aperture Response Characteristics

The "aperture response" of optical
devices is usually observed on square-
wave line patterns. The sine-wave
response can be derived from the square-
wave flux response or may be measured
directly as will be described in Part IV
which will deal specifically with the
subjects of resolution and aperture re-
sponse characteristics and a new system
of rating based on the measure of equiva-
lence developed in the following section.
The sine-wave response factor r$ has
been computed for various sampling
apertures and is shown in Figs. 41 to
46a.

The line number for these aperture
types is expressed in relative units
N/NI which refer to a characteristic
aperture diameter

d = //Ai (19)

where / = unit of length (/ = 1 milli-

Otto H. Schade: Motion Picture Granularity

187

meter, or / = V = vertical dimension
of frame). The line number JV$ specifies
the condition at which the length of
one half-wave in an optical sine-wave
pattern equals the characteristic aper-
ture diameter 6. Sharply defined
apertures have response characteristics
(Figs. 41 and 42) exhibiting an oscilla-
tory decrease of the response factor with

several zero values and 180° phase
change at every zero value, indicated
in the drawings by a change of direction.
Grain structures such as in photo-
graphic film or kinescope phosphors do
not, in general, form an infinitesimal
image of a mathematical point of light
and, therefore, have an aperture effect.
When a grain layer of finite thickness is

2 3 4

RELATIVE LINE NUMBER (N/NS)

Fig. 41. Sine-wave response characteristics of square aperture.

2345
RELATIVE LINE NUMBER (N/N§)

Fig. 42. Sine-wave response characteristics of round aperture (r = const.).
188 March 1952 Journal of the SMPTE Vol.58

exposed to an infinitesimal pencil of
light (or electrons), diffraction, diffusion,
and progressive absorption of light cause
an exponential spreading of light in the
grain layer, leading to the hypothesis
that the point image of the structure
has the form of a round "aperture" with
exponential transmittance T = e~r/r<>.
The sine-wave response computed for

this theoretical equivalent, is shown in
Fig. 46a in relative units N/N& for a
diameter d = 6r0.

Photographic film is exposed to light
twice, once during exposure, and then
again upon projection of the developed
image. The main aperture effect of
its grain structure occurs during ex-
posure in the undeveloped semitransparent

0 1234

RELATIVE LINE NUMBER (N/N§)

Fig. 43. Sine-wave response characteristics of round aperture (r = cos2r).

I 2 3 4 5

RELATIVE LINE NUMBER (N/N$)

Fig. 44. Sine-wave response characteristics of round aperture (T = C~ (r/ro)1).

Otto H. Schade: Motion Picture Granularity 189

state of the structure. The second
aperture effect occurs in the transmission
of light through the developed grain struc-
ture in printing or projection processes
and is of much smaller magnitude be-
cause of the high absorption of light in
the "black" silver grain structure. The
total aperture effect of photographic
film is, therefore, caused by large and

small exponential aperture in cascade,,
dominated by the first aperture effect
during exposure. It is thus found that
the sine -wave response measured with
an electronic microphotometer1 on test
patterns photographed on a variety of
films, or optically projected onto kine-
scope phosphors with various grain
sizes, mixtures and thicknesses, is closely

1.5 '• THEORETICAL LENS (WHITE LIGHT) 4^
: N§ = 820/-P'

12345
RELATIVE LINE NUMBER (N/N§)

Fig. 45. Sine-wave response characteristics of theoretical lens (white light).

2 3 , 4

RELATIVE LINE NUMBER (N/N§)

Fig. 46a. Sine-wave response characteristics of mathematical equivalent
aperture of grain structures.

190

March 1952 Journal of the SMPTE Vol. 58

represented by the normalized response
characteristic Fig. 46b. Exposures rep-
resenting large signals may lead to
waveform distortion which depends on
the linearity of the transfer characteristic
of the process. These effects can be
determined separately by the same
methods used in electron-tube evalua-
tions. The rated resolving power Ncr
of film is found to correspond to a re-
sponse factor r$ of roughly 2%. For
want of a better reference value, the
rated resolving power jVcr has been
chosen as the reference unit. Com-
parison of the theoretical and measured
characteristics (Figs. 46a and b) shows
them to be almost perfect duplicates
when Ncr is placed at JV/JV5 = 5.17
in Fig. 46a. It is emphasized that the
sine-wave response characteristic of a
grain structure is not a measure of its
particle size or its granularity and is
single-valued in a linear system. Small
signals, therefore, should be used when
photographic film is measured as pointed
out above.

The sine-wave response characteristic
of a number of aperture processes in
cascade can be computed accurately.

The overall response factor at any given line
number N is the product of the response
factors of the system components at that line
number. The use of sine-wave test pat-
terns avoids the indirect quadratic addi-
tion of line numbers required for
cascading square-wave characteristics1
because, unlike the square-wave flux
response which contains harmonic fre-
quencies, the sine-wave flux response
retains a pure sine-wave form through-
out the system.

3. Evaluation of an Equivalent Passband
N, and a Characteristic Aperture Dimen-
sion From the Sine-Wave Spectrum

The sine-wave components in a
source of deviations, such as an illumi-
nated random grain structure, may be
determined by a Fourier analysis of the
complex waveform obtained by scanning
the source with an infinitesimal aperture.
The amplitude and flux of the various
sine-wave components can be considered
alike and constant, the components
filling a continuous spectrum up to a
very high line number, as illustrated by
Fig. 47a. Integration of random devia-
tions by a scanning aperture of finite

GRAIN STRUCTURES. (.PHOTOGRAPHIC FILM,

0.5 1.0

RELATIVE LINE NUMBER (N/Ncr)

Fig. 46b. Sine-wave response characteristic structures of grain structures

(measured).

Otto H. Schade: Motion Picture Granularity

191

IDEAL RANDOM
GRAIN STRUCTURE

SCANNING

APERTURE COMPLEX

5=0 APERTURE OUTPUT

NUMBER OF HALF WAVES
(N) PER UNIT LENGTH

SPECTRUM OF SINE -WAVE
COMPONENTS IN COMPLEX WAVE

Fig. 47a. Fourier com-
ponents of ideal ran-
dom grain structure.

IDEAL RANDOM
GRAIN STRUCTURE

SCANNING

APERTURE COMPLEX

§>0 APERTURE OUTPUT

LINE NUMBER (N )

SPECTRUM OF SINE -WAVE
COMPONENTS IN COMPLEX WAVE

Fig. 47b. Fourier com-
ponents of random
grain structure after
an aperture process.

size results in a distribution of flux
components and a limited sine-wave
spectrum of the type shown in Fig. 47b.
The relative deviation resulting from
the scanning of an ideal random struc-
ture with an aperture is related to the
total response of the aperture by the
equation

= toy*0

(20)

The value if represents the average value
of the total flux; i.e., its d-c component.
The value [$] is the rms value of the total
(a-c) flux variation given by the £-

power of the integral of squared sine-
wave flux deviation components.

When the sine-wave response is nor-
malized according to Eq. (18) so that
$N = I at JV = 0, the mean squared
flux variation or deviation is expressed
by

(21)

where r$N is the sine-wave response
factor and i£0 measures the magnitude of
the (a-c) flux component passing through
the aperture with a line number N
approaching zero as stated by Eq. (18).
A hypothetical aperture having a con-

192

March 1952 Journal of the SMPTE Vol. 58

stant response (r$ = 1) from N = 0 to
a line number Ne* where the response
drops abruptly to zero, would give a
mean squared deviation

The integral of squared response factors
in Eq. (21) may hence be interpreted
as a normalized mean squared deviation
or an equivalent passband of constant
amplitude extending to the line number
Ne* as defined by

The measure N6* has the dimension:
length"1. Its reciprocal value expresses
an equivalent length or diameter of the
aperture in the scanning direction. Like
the aperture response, Ne* depends, in
general, on the aperture orientation
relative to the direction x of aperture
displacement. Apertures with circular
symmetry have a single effective length
proportional to their diameter 5 and a
single value Ne*. Elliptical or rec-
tangular apertures can be specified by
two values JV,*(0) and Ne*v>) obtained
by orienting their major or minor
dimensions (a or b} in the direction of
scanning. These two values can be
combined into a single value

representing an equivalent symmetric
aperture.

The direct evaluation of the measure
Ne* for an unknown aperture requires
a calibrated random grain pattern which
must be tested by a harmonic analysis
of the complex aperture output. A
practical alternative is a synthesis of
the sine -wave characteristic from the
aperture response to constant amplitude
sine-wave patterns of various wave-
lengths and an evaluation of Ne by Eq.

* The asterisk on the value Ne* is used to
indicate that this value is obtained when
a random grain structure is scanned.
Other values will be introduced subse-
quently.

(22). Optical sine-wave patterns con-
sisting of parallel "lines," however, do
not duplicate exactly the sine-wave
components in a random flux pattern,
but rather in a pattern which is random
only in the direction x of scanning and
uniform in the direction^, perpendicular
to the scanning direction. Figure 48a
illustrates the difference in cross sections
through a random grain structure and
a synthetic structure representing an
addition of sine-wave test patterns with
random phase relation. The differences
resulting from scanning a random grain
structure or sine-wave test patterns and
the suitability of jVe-values, in general,
for the purpose of indicating an equiva-
lent aperture area can be determined by
a comparison with an equivalent N0
based on the sampling of a normalized
random structure. The various equiva-
lents Ne*, N« and N0 can be computed
without recourse to response charac-
teristics when the geometrical properties
of the aperture are known.

The effective sampling area of an aperture
(pictured as a three-dimensional body,
the aperture transmittance r represent-
ing height) may be determined by sub-
dividing the aperture into differential
columns (see Fig. 48b) with a base
area Aa = AxAy and constant or varying
height representing the transmittance
T = f(X) y). The relative deviation
obtained by taking a large number of
samples from a random grain pattern
with one differential column is
A<r = (r2n0Aa)i/™°Aa

where n0 = average number of grains
per unit area. For a normalized grain
density n0 = 1, the above relation be-
comes

A<r0 = (T2Aa)i/rAa

Integration over the aperture area
yields the normalized relative deviation:

_ [lim S

, .

im SrAa

ff

Otto H. Schade: Motion Picture Granularity

193

FLUX PATTERN
RANDOM IN X AND y

0 LINE NUMBER (N)

SYNTHETIC FLUX PATTERN
RANDOM IN X, UNIFORM IN

V

Fig. 48a. Normal and
synthetic grain pat-
terns.

APERTURE SCANNING FLUX PATTERN RANDOM IN X AND

SCANNING DIRECTION

APERTURE SCANNING SYNTHETIC FLUX PATTERN RANDOM IN X, UNIFORM IN y

/dy[/f(x,y)dx]2

I IN X, UNIFORM IN ;
X2_/dx[/f(x.y)dy]2

SCANNING DIRECTION

APERTURE SCANNING FLUX PATTERN1" RANDOM IN X AND

//fO,y)dxdy

("^PATTERN FLUX NORMALIZED SO THAT HQ = I PER UNIT AREA WITH T * l)

Fig. 48b. Subdivision of apertures for integration of flux values.

The normalized relative deviation <r0
has the dimension length"1. The length
may be regarded as the geometric mean
of the sides of an equivalent rectangular
sampling area ae having constant trans-
mittance r = 1. According to Eq.
(19) the relative deviation a0 = l/(ae)*
can also _be interpreted as the line
number N0 of an equivalent square

sampling aperture and Eq. (24) may,
hence, be stated in the form:

, , , ,.

N0 = r r f(x \dxf (25)

'

The measure N0 is independent of the
aperture position for both symmetric
and asymmetric apertures and can,
hence, be used as a standard for com-

194

March 1952 Journal of the SMPTE Vol. 58

parison. The equivalent passband JVe* of
an aperture scanning a grain structure
random in x and y directions can be
computed by subdividing the aperture
into incremental sections parallel to the
direction of scanning (see Fig. 48b).
The mean squared flux obtained is the
same as that obtained when the aperture
is sampling. The flux $o(y) at JV = 0
contributed by each section to $0 is
represented by the areas J* rdx of the
sections, and because the flux is random
(out of phase) in y, the total flux i£02 is
obtained by the sum of the squares i£02 =
2[y* rdx]z. The measure JVe* obtained
when a random grain pattern is scanned
is, therefore,

f dy[f f(X,y}dx}*

(26)

The asterisk is used to distinguish JVe*
from the value JV, which will henceforth
be used to indicate a sine-wave synthesis.
Evaluation of the equivalent passband Ne
from a response characteristic obtained by the
method of scanning sine-wave test patterns
represents the case in which a synthetic
structure random in the x direction but
uniform (in phase) in the y direction is
scanned. The aperture is subdivided
into sections parallel to y. The mean-
squared flux [$]* is the sum of the
squares of the section flux values
2[y* rdy\\ and the flux ^02 is the squared
sum of the section flux values, furnishing
the ratio

All measures JV0, JVe* and JV, represent
dimensionally a length"1, but the formu-
lations appear to have little resemblance
to one another. Because the measures
JV«* and JV« depend on the direction of
scanning, asymmetric apertures require
evaluation of two JVe-values as stated by
Eq. (23). For apertures having circular
symmetry, however, the sampling
equivalent JV0 is seen to equal the geo-

metric mean (JVe* JVe)*. To evaluate
the relative accuracy of the three
measures it is of interest to determine
how closely the values computed with
Eqs. (25), (26) and (27) compare in a
number of representative cases. To
provide JV« in relative units JVe/JVa, the
above equations must be multiplied by
the ratio of the characteristic lengths
8/u when the length u chosen for com-
puting the measure JVe differs from the
length expressing a characteristic diam-
eter of the aperture. In relative units
Eqs. (25), (26) and (27) can be written:

u fff(*,y}dxdy

5 fdx[ff(x,y)dy}*

It must be kept in mind that the length
u in Eq. (25a) is the square root of an
area and, therefore, independent of
the aperture orientation. The length
ux in Eqs. (26a) and (27 a), however, is
always the characteristic aperture length
in the direction x of scanning.

The measure JVe for a round aperture
with T = e~(r/ro)2, for example, may be
computed in terms of a radius length
r0 = ux; the corresponding relative
line number unit JV5 in Fig. 44 represents
a length"1 measured by the diameter
5 = 4r0. The ratio S/ux in this case is,
therefore, four. The relative values
in Table IV show that the sine-wave
equivalent JV, is as good an equivalent
as the value JV«* obtained by the scan-
ning of a random grain structure. Both
values are somewhat in error for a round
aperture with r = \ and for a square
scanned diagonally. Practical apertures
such as lenses, grain structures or electron
beams have nonuniform transmittances
similar to the aperture types 4 to 6 in
Table IV, for which the error is neg-
ligible or zero. The definition of JV,* as
the integral of squared response factors

Otto H. Schade: Motion Picture Granularity

195

V

a
•<

I

I
J

D

196

March 1952 Journal of the SMPTE VoL 58

I

given by Eq. (22) applies also to the
measure JV., which is obtained from a
sine-wave synthesis; i.e.,

JV. =

(28)

The results obtained by a numerical
integration of the squared aperture re-
sponse according to Eq. (28) are illus-
trated by the curves JVe(o_*Jv) in Figs.
41 to 46 which show the growth of the
partial integral when the limit is in-
creased from JV = 0 towards JV = oo.
The accurate agreement of the values
obtained by this method is a check on
the accuracy of the sine-wave response
characteristics as well as the formulation
of Eq. (27). The e~r/ro aperture is of
interest as a mathematical equivalent
to grain structures with finite thickness.
The line-number scale of this aperture
is referred to a diameter 8 = 6r0 which
for identical values JVe places the rated
resolution JV^ of film at the value
NjNs = 1.245/0.241 = 5.17 of the
theoretical characteristic. A comparison
of Figs. 46a and 46b shows an almost
perfect agreement of the sine-wave re-
sponse characteristics. The resolving
or sampling aperture of grain structures
is, therefore, well represented f by a
round aperture with a transmittance
T = e"~r/ro.

The value JV, of an asymmetric aper-
ture of length a and width b can be
determined accurately when the de-
formation of the dimensions a or b from
circular symmetry does not alter the
relative aperture transmittance in the
b or a dimension, respectively. In this
case the sine-wave measure JV^,,) or
JV«(6) obtained with Eq. (27a) is deter-
mined by the dimension of the aperture
(a or b} which is oriented parallel to
the direction x of scanning, the measure
being independent of the aperture scale

f A finite grain size removes the pointed
tip of the aperture transmittance. The
effect, however, is negligible because the
flux contributed by a transmittance ex-
ceeding the value r = 0.65, (r = 0.6 r0)
is only 2.5% of the total flux.

Otto H. Schade: Motion Picture Granularity

197

Table V. Evaluation of JVe for 40-mm
Cine Ektar Lens at f/1.6 (5°).

Table VI. Evaluation of Ne for Grain
Structures.

JV/mm

r*

a

a^N

^(a}^N

10

0.98

0.99

9.8

20

0.94

0.96

9.2

30

0.90

0.92

8.5

40

0.85

0.88

7.75

50

0.79

0.82

6.7

41.95

60

0.74

0.765

5.85

70

0.67

0.70

4.9

80

0.62

0.65

4.22

90

0.57

0.59

3.5

100

0.53

0.55

3.0

63.42

120

0.46

0.49

4.8

140

0.42

0.44

3.88

160

0.39

0.40

3.2

180

0.36

0.37

2.76

200

0.33

0.345

2.38

80.44

250

0.27

0.30

4.5

300

0.20

0.23

2.65

350

0.14

0.17

1.45

400

0.08

0.11

0.61

450

0.03

0.05

0.13

89.78

A/

'Ncr

r^

a^N WAX

0.05

0,

97

0.

985

0

.049

0.

10

0.

91

0.

95

0

.045

0.

094

0,

15

0

835

0.

88

0

.0385

0.

20

0,

740

0.

79

0

.031

0.

1635

0.

25

0,

67

0.

685

0

.0235

0.

30

0,

53

0.

585

0

.0171

0,

2041

0.

,35

0.

44

0.

50

0

.0125

0,

,40

0,

,37

0.

41

0

.0084

0,

,225

0,

45

0,

30

0.

335

0

.0055

0.

,50

0

,245

0.

275

0

.0038

0

,2343

0

.55

0

,20

0.

22

0

.0024

0

60

0,

,16

0.

18

0

.0016

0

,2383

o

65

0

,125

0.

14

0

.0010

0

,70

0

,10

0.

11

0

.0006

0

.240

0

,75

0

.075

0.

085

0

.00035

0

,80

0

.058

0.

065

0

.0002

0.2405

0

,85

0

.04

0.

045

0

.0001

0

,90

0

.03

0.

04

0

.0001

o

.2407

o

.95

0

.02

0.

03

1

.0

0

.018

0.

02

0

.241

N€ = 90 Lines/mm

N. = 0.241

factor in the y direction. The aperture
is thus simply considered first as an
aperture with circular symmetry and a
diameter 5 = a, furnishing the value
JV«(o), and second as an aperture with
the diameter 5 = b, furnishing the value
Ne(b). The geometric mean of these
values (Eq. (23)) furnishes the sym-
metric equivalent JV«. The correspond-
ing procedure when the sine-wave re-
sponse of an astigmatic lens is measured,
for example, requires orientation of the
sine-wave pattern and scanning direc-
tion parallel or perpendicular to the
direction of astigmatism. The values
W«(«) and JV«(6) are then determined by
numerical integration from the two
corresponding sine-wave response
characteristics (Eq. (28)). The evalua-
tion of Ne is illustrated by two examples
in Table IV.

The numerical evaluation of the meas-
ure JV, from a sine-wave response charac-
teristic by means of Eq. (28) is illustrated
by Table V for a 40-mm f/1.6 Cin6
Ektar lens measured at //1 .6 and 5° off
axis. The value a is the mean response

factor within the increment AJV. The
equivalent passband jVe is obtained di-
rectly in television lines per millimeter;
JVa = 90 lines/mm. Table VI illustrates
the evaluation of JVe for grain structures
from Fig. 46b in relative units. With
reference to the rated resolving power
Ncr of film, Ne = 0.241 Ncr. Hence,
for fine-grain positive film (Type 5302)
with Ncr =180 television lines per milli-
meter, N, = 43.4 lines/mm.

4. Equivalent Aperture Diameters

The line number for known round
apertures is expressed in relative units
JV/JVj which refer to the aperture diam-
eter

5 = l/Jf9

where / is the unit of length (/ = 1
millimeter, or / = V — vertical picture
dimension). Relative to the equivalent
passband JV« = kN&, the diameter of these
apertures is expressed by the relation
5 = lk/N. given in Table VII.

An equivalent aperture or point image of
specified characteristics can thus be
obtained for a system element by the

198

March 1952 Journal of the SMPTE Vol. 58

Table VII. Diameter 5 and Equivalent Passband Ne of Various Aperture Types.

Aperture type

Relative
transmittance Diameter (S)

Relation of 5 to Ne

Square

T

= I

S

s = l/N.

Round

T

= 1

2r0

d0 = 1.08 l/N.

Round

T

= cos2 r

2r0

5C0. = 1.59 l/N.

Round

T

— e~r/ro

6r0

Sf = 1.245//.W,

Round

T

= 6~(r/ro)2

4r0

8el = 1.6 l/N.

insertion of its JVe-value into the relations
given in Table VII.

The equivalent passband Nt (tele-
vision lines) and the equivalent aperture
sizes of a number of system elements
used in photographic processes are
summarized in Table VIII.

5. Equivalent Passband and Aperture
Diameter of Processes Containing a
Number of Elements in Cascade

The sine-wave response characteristic
of a number of system elements in
cascade, including the eye if desired,
can be computed accurately by forming
the products of the response factors
r$i r$2 • • • T$n °f actual response charac-
teristics at corresponding line numbers.
The equivalent passband JV"e(p) of the
process is thus given accurately by the
integral

Because of the nature of the response
characteristics of lenses, films and tele-
vision tubes it has been found that the
equivalent sampling area of a combina-
tion of such "apertures" can be evaluated
with usually less than 5% error by simply
adding the equivalent aperture areas of
the components or, as expressed in
terms of equivalent aperture diameters :

*(rt - (5i2 + 522 + . . . + «,»)» (30a)

(30b)

It thus becomes a simple matter to
compute the equivalent passband Nt(p)

and the aperture diameter of photo-
graphic systems by the use of Eq. (30)
in conjunction with Table VIII. Equa-
tion (30) is exact for exponential
apertures T = e~(r/ro)2 because the re-
sponse characteristic (Fig. 44) has the
form r\j/(N) = e~KN*. The response
characteristic of a system of two-
dimensional apertures tends to approach
this form (Fig. 44), which may therefore
be used as an equivalent response charac-
teristic with a line number scale JVfi =

jVe/1.6.

The simplified method will lead to
larger errors and should not be used
when electrical components of a tele-
vision system such as amplifiers or filters
with sharp cutoff or a rising frequency
characteristic are included. Although
equivalent passbands (JV.) for such
components have a significance, they
cannot be treated as normal optical
apertures. It is well known, for example,
that any number of amplifier stages of
limited range can be combined and
corrected to have an overall "flat"
response characteristic. It must also
be kept in mind that the measure N,
of an optical aperture is used as a sub-
stitute for N0 and normally refers to
the property of an area. JV, differs,
therefore, from the "noise equivalent
bandwidth" A/(eq) of an electrical fre-
quency-response characteristic which di-
mensionally refers to a length. Combi-
nations of electrical components with
two-dimensional apertures will be dis-
cussed in Part III.

Otto H. Schade: Motion Picture Granularity

199

a

f

O,

<

bo

I

"3

200

March 1952 Journal of the SMPTE Vol. 58

D. GRANULARITY AND RANDOM FLUCTUATIONS
IN MOTION PICTURE PROCESSES

An objective analysis of the luminance
fluctuations caused by the grain structure
in motion pictures requires evaluation
of three factors: (1) the law of sample
distribution at the source of deviations,
(2) the effect of nonlinear transfer
characteristics of system components on
the signal-to-deviation ratio, and (3)
the effect of the system apertures on the
total luminance fluctuation and the sine-
wave spectrum in the final image.

Subjective effects such as the appear-
ance and threshold visibility of fluctua-
tions (graininess) depend on the fluctua-
tion level, gamma and aperture effect
of the process of vision and can be
evaluated by including the characteristics
of the eye in the imaging process.

1. Deviation Characteristics
of Photographic Film

The sensitometric curves of film
showing the grain density D as a function
of exposure E are a graph of the relative
sample number as a function of exposure
because

If a random distribution is assumed and
n' is made equal to the average grain
number (in a) at D = 1, the rms devia-
tion becomes

[AD] = (n'D)i (31)

Density and light transmittance of the
film are connected by the relation
D = -log r

Differentiation of this function

dD/dr = -(loge)A = -0.4343 /r (32)

establishes the fact that small deviations
AD in density or grain number cause
a reciprocal deviation AT in transmit-
tance. The relative deviation in trans-
mittance is therefore

<rT = [Ar]/r = 2.31 [AD] (33)
and with (31)

«TT = 2.31 (Dn')* (34)

The effective grain number n' in actual
films may, hence, be determined from
measurements of o> and D with

and

<TD = 0.4343 <rr

n' = (2.31 /ar

(35)

This simple relation is accurate for
small deviations aT < 0.1.

Random deviations [A]/ caused by
sources other than exposure to light, such
as fog, dust and other irregularities, may
superimpose a constant deviation level
which remains at zero exposure. The
total deviation is then increased to the
value

[A]' = ([A]' + [A]/)i

and

<rr' = 2.31([ADP+ [AD]/)* (36)

When both density and [R]T are
plotted in logarithmic units (see Fig. 49)
the curves showing [R]T as a function of
density become straight lines having a
slope of minus one-half for random
distributions. In this case the effect of
an unknown constant deviation causes
at low densities a departure of measured
values from a straight line. When the
addition of a constant value Df to the
measured values brings the points into
line, the deviations have a random
distribution. The results of deviation
measurements (see Appendix) made
with round sampling apertures on a
number of film types are shown in Figs.
49, 50, 51 and 52. The diameter 5 of
the sampling aperture is indicated.

Various causes introducing extraneous
deviations were noted. Variations or
irregularities in the transmission of the
film base, thickness of the emulsion,
photosensitivity, uniformity of develop-
ment or physical defects introduce
deviations similar to the various defects
which can occur and modify the fre-
quency distribution and consistency of

Otto H. Schade: Motion Picture Granularity

201

1 o

100
80

60
40

10

. PLUS

X

1 1 1 1 1 1 1 1 1 1
SAMPLING APERTURE 5 = 30ji
X D -• = O.I NEW FILM

. Df = 0.2 OLDER FILM
(30 TO 60 READINGS PER POINT)

^

X

'^

hh

*x

•
V

X

X

*^>sv^».

O.I 2 A 6 8 1.0 2 4 6 a
DENSITY (D) ABOVE BASE

Fig. 49. Deviation charac-
teristic of Plus X Negative
Film.

Z 60

Ox-s
$_^40

- '! ,o

SUPER XX

SAMPLING APERTURE & = 30^4
Df=O.I5 :
(30 TO 60 READINGS PER POINT)'

x^

pS,r

^

^

IGNAL-1
RATIO

6 !

•v,

X

X

N

^.

••<;

8
6
A

^— -H

^**5

O.I

2 4 6 8 |.0 2

DENSITY (D) ABOVE BASE

Fig. 50. Deviation charac-
teristic of Super XX Negative
Film.

2^
|<ioo

1 1
SAMPLING

APERTURE 8
1 POS. 1302
5 NEC. 1203
READINGS PE

= 15/t
R POI

NH

)

* C

• c

[3C

>f

Jf

) 1

•£.
•C.

•o

0.0
0.0
6C

- '^80
0S 60

^° 40

^ 30

>^

^v

•

^N^

•

S

X

0 oc 30
to

'fc.

v^

* "S^ .

10

\

O.I 2 4 6 6 1.0 2 468

Fig. 51. Deviation charac-
teristic of fine-grain film.

DENSITY (D) ABOVE BASE

202

March 1952 Journal of the SMPTE Vol. 58

)- DEVIATION
MT=^T)

5 S§ §

5A

UF

rr

>LI

TT
NG

~i —

APERTL

1 —

RE

1

B =

1 —

3(

)M

\

^»»s>

^"

&

sP

T o

_J 1- 40

e ?

^

s,

S?- v

< <

g * 30
«/>
20

10
<

^

\f<./.

^£c^

*0

^N

§

fl

^

Jf

<*,

Ns

3.1 z 4 6 • 1.0 2 4 e

DENSITY (D) ABOVE BASE

electrical fluctuations in television sys-
tems. In comparison with the immense
number of samples involved in a dynamic
measurement of electrical fluctuations
it can be expected that static sample
measurements on a small film area will
show a greater spread of values.

The measured characteristics (Figs.
49, 50 and 51) substantiate the variation
of [R\T with the one-half power of the
grain density predicted by Eq. (34) for
a random distribution. Additional
proofs can be obtained directly from
any one set of sample readings taken to
determine a point on the [R]T charac-
teristics. When the readings are plotted
on probability paper (arranged in order
according to value), a gaussian dis-
tribution of values is indicated by a
grouping around a straight line. Figure
53 shows the distribution of readings
determining three points on the [R]T
characteristic of Plus X film and one
point on the characteristic of fine-grain
positive film.

These plots may be compared with
Fig. 54 which shows two sets of sample
readings taken on photographs of elec-
trically produced "grain" structures in
a single television frame. The photo-
graphs were made on 4 X 5 in. film,
a positive of which was sampled with a
round aperture. The electrical fluctua-
tions so recorded are random fluctuations

Fig. 52. Deviation charac-
teristics of several film types.

occurring in the current of a multiplier
phototube exposed to a "d-c" light source
and passed through electrical amplifiers
having a constant or a rising (peaked)
frequency response characteristic. The
agreement with the theory is sufficiently
good to justify the assumption that the
deviations in motion picture film are sub-
stantially gaussian and certainly random
as defined in Sec. Cl.

A comparison of deviation ratios in
film types is made on the basis of
sampling apertures having equal areas
or equivalent optical passbands. For a
random distribution of samples, the
deviation ratio [R]T is proportional to
the diameter 5 of a round sampling
aperture. The curve for fine-grain
film in Fig. 52 is, therefore, obtained
from Fig. 51 by multiplying the [R]r-
scale by the ratio of the aperture diam-
eters. The validity of this process
can be demonstrated visually by photo-
micrographs of equal densities on Plus
X and 1302 positive film. For equal
magnification the grain structures are
shown by the prints 0 and la in Fig. 55.
Magnifications in the ratio of their
[/?]-values results in equal deviations
per unit area in the prints 1 and la.
Out-of-focus projections are a convenient
means for reducing the optical passband
and thus visually checking the relative
"frequency" distribution of the devia-

Otto H. Schade: Motion Picture Granularity

203

120

> 110-

o

z

O

<

u
oe

§90

80

_2
120^

110-

100-

90-

130

120

no

100

4

130

120

I-PLUS X 6=1.18 5 = 30ji
2 -PLUS X D = 1.6 8 = 30M

3-PLUS X D= 0.55 S=30/i
4-1302 F.G. POS. D = l.22 S=I5/LI

^ ,

I

2

Z

Jf

2

^ /

7

^

y^

>

T

^

-

/

vvx

f

>

F

«vi

•^»

v^

/

ftflX

L-

^

^ ^

/

1

2

4

i«

X*

S

IOO-

/*

X

-

'

/

S

<

/

2.

fl

«

«K^

.X

j

* \

>/

^

v/

*s

^

m

**

•*4

r

/

s*

rf*

90

z

^

2

^

'/

v>

Vx

J(

4

/

0.05

0.2

1

5

20

40

60

80

95

99

99.8

99.

0.01 O.I 0.5 2 10 30 50 70 "" 90 98 "W.5 99.9

READING NUMBER- PER CENT

Fig. 53. Sample readings on Plus X and fine-grain films
plotted on probability paper.

± J.

112-
co

|,,o

o:

t 106-

?«4H

«)

Z I02H

100-

96-

94-

120

A

' *S

. i

tf -

9

M

c

/

/

>

/

/

~7_

y

/

y

?_

7

/

RESPONSE
PQRTIONAL TC

/

/

IOO

PRO

f _,

y

>

>

r

j

r

^

^

/y,

V

/

*

\

y

\

90
0.

)

^

r

ONSTANT AMPLITUDE
RESPONSE

1 /

.

*

/

/

/

2

2

/ (

/

/

/

/

0.05

31 0

0.2

.1 0

5

;

> 5 *

> 2°3

40
0 5

06°7

0 9

O95 9

8"«

.5

£

9

READING NUMBER — PER CENT

Fig. 54. Sample readings of television "noise" photographed on
4X5 in. film plotted on probability paper.

204

March 1952 Journal of the SMPTE Vol. 58

be 2 S

'

•sod soei

sod zoei

x snid

Otto H. Schade: Motion Picture Granularity

205

~©

-it-

=>.

206

© © ©

March 1952 Journal of the SMPTE Vol. 58

tions, because the optical passband
decreases in proportion to the diameter
of the disc of confusion (sampling
aperture) in the out-of-focus projection.
The integrated grain structures 2, 2a,
3, 3a, etc., of the two film types have a
substantially identical appearance for
passbands in the ratio of their [R] -values,
confirming a substantially identical dis-
tribution. The size of the integrating
aperture of the projection is shown by
pinhole images under the grain images.
The aperture sizes are indicated in
microns (;u) .

2. Sampling Apertures and Transfer of
Deviations in Motion Picture Systems

The four dominant sampling apertures
and intermediate images of a motion
picture system are shown in Fig. 56a.
The corresponding aperture response
characteristics r$ = /(JV) of the system
elements, their transfer characteristics
lAout = /('Ain) and the points where
random deviations ($) are introduced
into the system are indicated in Fig. 56b.
It is assumed that no aperture effects
are introduced by camera, printing and
projector mechanisms.

A number of minor apertures have
been omitted in the diagram because
they are of much smaller magnitude
than the main sampling apertures of the
system. The developed grain structures
of the negative and positive films, for
example, introduce a small aperture
effect in printing and projection proc-
esses in addition to their main aperture
effect which occurs during exposure in
the undeveloped grain structure. The
diffusion of light and, consequently,
the aperture effect of the black silver
grain structure of the developed films,
however, is of much smaller magnitude
than in the undeveloped transparent
state of the structure and can be neg-
lected.

For an analysis of deviations the block
diagram contains, therefore, only the
elements shown in Fig. 56c indicating
the dependence of the relative deviations

a\ and <r2 on the densities D\ and Z)2 of
the respective film processes. The
method of measuring the relative devia-
tion (0-1) in the negative film with a
known sampling aperture (5) is indicated
by an alternate path of $\ over the
switch S.

The value of the relative deviation has
been determined by measurement for
several motion picture film types with
the use of a round sampling aperture of
specified diameter 5. In the actual
process (Fig. 56c), the deviation flux
$1 from the negative is "filtered" by the
response characteristics of the apertures
52 and 53 and passed through the transfer
characteristics 2 and 3 to the projection
screen. The sampling aperture 5 used
in the measurement of a\ is, hence, re-
placed by the cascaded value 8<& =
(522 + 332)*. According to Eq. (17)
the deviation is changed in inverse
proportion with the aperture diameter;
it is further changed by the transfer
ratios g/G,\ i.e., the point gammas
72 and 73 of the two transfer charac-
teristics. The relative deviation a\ from
the negative is, therefore, changed after
transfer to the projection screen to the
value

273 (37a)

Similarly deviations originating in the
positive film are changed from the
measured value <r2 to the value

<T23 = l/[/2]28 = ^2(5/53)73 (37b)

The total relative deviation <rp at the
projection screen is then obtained by a
quadrature addition of the relative
deviations <r13 and o-23 as illustrated by
numerical examples in the following
section.

3. The Signal-to-Deviation Ratio
in Projected Positive Film

The signal-to-deviation ratio [R]P
in a standard 35mm motion picture
process is determined by the following

fSee Part I, p. 145.

Otto H. Schade: Motion Picture Granularity

207

components: Plus X negative film,
1302 fine-grain positive film, and a 4-in.
J/2 Super Cinephor projection lens.
The sampling aperture 323 for deviations
originating in the negative film is the cas-
caded value of the equivalent aperture
of the positive film (S2 = 25ju) and the
equivalent aperture of the Super Cine-
phor projection lens (53 = 39. SM)

$M = (252 + 39.52)i = 46.7 M
A second value

'» = (252

39

transfer characteristic with a constant
gamma of unity (73 = 1) requires a
complete absence of lens flare and a
totally dark and nonreflecting projection
room. In all practical cases lens flare
and ambient light superimpose a light
flux "bias" ^0 on the projection screen.
The signal flux is, hence, increased by
the ratio (if + <£o)/i/' which, as easily
shown, reduces the constant value 73 = 1
as a function of \js to the values 73 =
$3/($a + ifo). Expressed in units of
film transmittance

representing a combination of the posi-
tive film and a physical 30-M aperture
will be used for test purposes. A tabula-
tion of the response factors of motion
picture components and combinations
(products) is given in Table IX. (Note
the close agreement of the equivalent
apertures computed from the actual
response characteristics with those ob-
tained above.)

Representative transfer characteristics
of the negative and positive films are
shown in Figs. 14 and 15 of Part I.
The total density values Z>i and D2,
the densities above base D±* and Z)2*,
and the point gamma of the positive
film (72) are listed in Table X.

A projection lens having a linear

73 = T2/(T2 + T6)

(38)

For a normal high light transmittance
T2max = 0.5, a 1.5% light bias i£0 =
0.015 i/'smax (compare Fig. 18, Part I)
corresponds to the value rb = 0.015 X
r2max = 0.0075 and 73 = r2/(r2 +
0.0075) which is listed in the lower
portion of Table X for various values of
Do and r2 + 0.0075, the latter being
proportional to the total screen
luminance B -f- #0. The effect of the
light bias is computed separately by
first letting 73 equal unity. The ratio
[R]i of the Plus X negative film measured
with 5 = 30ju is taken from Fig. 52.
The transferred values [^]*ia, indicated
by an asterisk, are computed with Eq.
(37a) for the cascaded sampling apertures.

Table IX. Sine-Wave Response Factors of Motion Picture Components.

.V/mm

Baltar

at//2.8

Neg.
film
Plus X

Fine-
grain Cinephor
pos. 5203 at//2

Aperture

30 M

3 + 4

3 + 5

1 to 4

10

20
30
40
50
60
70
80
90
100

0.99
0.95
0.90
0.82
0.74
0.66
0.58
0.50
0.45
0.40

0.93
0.78
0.60
0.43
0.30
0.205
0.14
0.09
0.07
0.03

0.97
0.90
0.815
0.70
0.59
0.48
0.39
0.31
0.25
0.20

0.94
0.80
0.60
0.45
0.33
0.25
0.18
0.13
0.10
0.08

0.97
0.91
0.80
0.66
0.51
0.35
0.22
0.09
0

0.91
0.72
0.49
0.315
0.195
0.12
0.07
0.04
0.025
0.016

0.94
0.82
0.652
0.462
0.30
0.168
0.084
0.028

0.84
0.54
0.26
0.111
0.044
0.016
0.006

No.

1

2

3

4

5

3 + 4

3 + 5

1 to 4

^rm

64

17

26.5
40.8

43.4
25
Computed

27.3
39.5
with Eq.

36

30
(30): 8PO

23.5
46
= 46.7

27.8
39
39

15.8
68
64.5

208

March 1952 Journal of the SMPTE Vol. 58

J

<£

c^

V

a

CS

be
5 B

CL g
S .§

^^^^^^^^.(NrO^

CM CN oj -r-i m

V oo d <N vo o <N oo r- CN m

?.

un LO m in CN oo m

CO CM CN (N

O O O O ^- ^ -TH T-. ^H T-H

8

CO fM fN CN fN CN

CM oo CN oo oo h- in

m T-I r-~ cv

II

Otto H. Schade: Motion Picture Granularity

209

5'23 = 39/z and S23 = 46. ?M letting 73
equal unity. The signal-to-deviation
ratios [R]*& of the positive film alone
are tabulated likewise for the density
values Z)2* (from Fig. 52) and the
sampling apertures S'3 = 30> and
53 = 39.5/t (cinephor lens). The total
signal-to-deviation ratio [/?]% in the
projection is then computed with

Division of the values [/?]% by corre

sponding values -ys furnishes the values
[R\p containing the effect of the light
bias fa.

A comparison of the values [/?]*i3
from the negative alone with the values
[R] *p of the process shows that the fine
grain of the positive film contributes
little to the total deviation. As illus-
trated by the block diagram Fig. 56c,
the total deviation is a composite value
of deviations from two unequal pass-
bands and requires further discussion.

120

2
O

S'oL

u^-, I0°
0 & 80

Po 6<>

< a. 40

o
«o

20

10

lb- STD. M. P.

WITH Tb = 0.0075
PLUS X NEC., 1302
SUPER CINEPHOR LENS
FILM PROCESS
BY 30-M APERTURE
PROCESS 5203,
.ENS = 16. SM

1

- STD. M.P.
F.G. POS.
- STD. M.P.
SAMPLED
- F. G. M. P.
5302, 8 I

2

lb

\

3

Xy

\^

^

s

s^

S

S

j

f

s

,

^

s

s

>

^ .

r '

"--

-^

^

\

^,-

^

2

T

"=

•*••, ,

"

1^

*••
^

--

rfc_

--*.

-=
)

^

^ II •

v=

•— .

•• ^
X

—

•-^
=.
~

Sr-

•^m

E=3

^=
— —

?-

^

^

=?.

^

^ ^

2.5

2.0 1.5

DENSITY (D£)

1.0

0.5

Fig. 57 a. Deviation characteristic of projected motion picture
without ambient light.

0001

0.01 0.1

TRANSMITTANCE (TZ + 0.0075)

1.0

Fig. 57b. Deviation characteristic of projected motion picture
with ambient light.

210

March 1952 Journal of the SMPTE Vol. 58

V) 120

no

90-

no

PLUS X COPIED ON 1203
SAMPLING APERTURE $

F.G. POS.

FILM

y

/

/

j

)

/

f

z

>

4

/

,X

90

2

%/

?

/

2 = l-

y1

»n

A

L>

2

y

u

/ *

y

/

1

2

-n

2

Y

/

A

/^

2

r

*~D2=2.I

x

/

/

0

/

)

^

/

/

J

?

• /

/

0.05
01 0

,°-2o

5 ' j

\ 5 l<

) 2°3

40
0 5

06°7

O95 9

8

99

99.8
).5 9<

99.
(.9

READING NUMBER — PER CENT

Fig. 58. Sample readings on motion picture positives
plotted on probability paper.

STD. M.P. PLUS X, 5302,
_ 4" SUPER CINEPHOR LENS
D2 = 0.9I

28

>

^4

Fig. 59. Sine-wave spec- 2
trum of deviations in
motion picture process.

6 8 10 20

LINE NUMBER

40 60 80 100

200

Otto H. Schade: Motion Picture Granularity

211

212

3d

Fig. 60. Composite grain structures in motion picture positives.
March 1952 Journal of the SMPTE Vol. 58

A plot of the signal-to-deviation ratios
[R]*p and [R]p from Table X is shown
in Figs. 57a and 57b, illustrating the
considerable reduction of deviations
and corresponding improvement of [R]p
in the shadow tones by ambient light.
The broken line, curve 2 in Fig. 57a,
for a 30-ju aperture was computed for
comparison with a direct measurement.
The measured values (#) were obtained
by sampling a contact print (on 5302
stock) of a I IB sensitometer exposure on
Plus X film with a 30-^ aperture (spray
process on negative, deep tank on
positive by De Luxe Laboratories). The
agreement with computed values is
good, particularly when it is considered
that optical effects and film slippage in
printing cause additional integration
and increase the sampling aperture and
[.ft] *-values. Figure 58 shows that the
gaussian distribution is maintained in the
composite structure sampled by a 30-ju
aperture.

4. The Optical Passband of the Total
Deviation in Projected Positive Film

The sine-wave components of different
passbands are combined into a single
passband by a geometric addition of like
sine-wave components. To perform this
addition the relative values of corre-
sponding Fourier components must be
known. The flux $0 representing the
flux or amplitude T0 of the Fourier
component approaching zero line
number is obtained from Eqs. (27) and
(20) ^0 = #<r/JV.*. Substitution of
I/A; = c*d from Table VII yields

T0 = c$8*<r = c$8i/[R] (40)

in which c is a numerical constant.

Because the relative deviation (<r)
itself is proportional to 1/5 and $ is pro-
portional to 52, Eq. (40) reveals the
interesting fact that the amplitude of
the zero-line number component (T^)
or (\J/0) increases with the 3/2 power
of the diameter of the sampling aperture;
a fact having no parallel in electrical
filter circuits. The amplitudes of the

zero line number components in the
projected relative deviation <r13 =
Vigils originating in the negative, and
the projected relative deviation <r23 =
l/Ma« originating in the positive follow
from Eq. (40).

ru
r23

and their ratio is:

(41)

(42) f

The constants c and the signal flux ratio
cancel out because they refer to the
common signal flux from the projection
lens (see Fig. 56). The geometric sum
Tp = (r]32 + r232)* can be expressed
by the ratio

(43) t

Analogous to Eq. (41), Yv can also be
expressed by

Tp = cfJJ/W, (44)

Forming the ratio TP/T23 with Eqs.
(44) and (41) eliminates the constants
c and the identical flux values \J/P — fa,
and yields an expression for the equivalent
sampling aperture dPo for the total relative
deviation in known quantities:

(45) f

The corresponding equivalent passband is:
JV«(p) = 1.08/5Po. The amplitude ratios
and the equivalent aperture of the stand-
ard motion picture process are listed in
Table X. In photographic processes
the values N or Ne are usually measured
in lines per millimeter whereas lengths
or aperture diameters (5) are measured
in microns. A conversion factor of 103
appears, therefore, in the quantities in
Table X.

It is seen from Table X that the rela-
tive amplitudes and equivalent pass-
bands vary somewhat with the film

fThe [R] values in these equations may
be replaced by the corresponding [R]*
values computed for 73 = 1 which cancels
out in these ratios.

Otto H. Schade: Motion Picture Granularity

213

density. A set of conditions is shown
in Fig. 59 for D2 = 0.91 . The amplitude
characteristic of the deviations from the
negative film (curve N) is given by the
combination of 3 + 4 in Table IX.
The relative amplitude scale is indicated
by the value Yn = (4.4/4.5) Tp at
N = 0. The response characteristic
(P) for the deviation from the positive
film is determined by the projection
lens and copied from column 4 of Table
IX. Its initial amplitude is T2Z =
Tp/4.5. The geometric addition JV + P
of the composite deviations from both
films shows the negligible effect of the
fine-grain positive on the total deviation
at the density D = 0.91. The aperture
effect of the positive film can be demon-
strated visually by photomicrographs of
the grain structure in 5302 positive
film containing a print of the grain
structure of Plus X negative film.
Print 1 of Fig. 60 is a sharp copy of the
composite grain structure. The large
white patches represent integrated grains
from the negative film which average in
diameter the equivalent resolving aper-
ture (52 = 25/i) of the positive film.
Prints 2 and 3 are out-of-focus projections
with lens apertures 53 = 5/z and 83 =
10/x, respectively, demonstrating the
integration of the fine-grain structure
of the positive film by an excellent pro-

jection lens. The negative grain in
print 3 is, hence, integrated by an
equivalent aperture of 27-ju diameter
(Eq. (30)) while the positive grain is
integrated by a 10-M aperture. A com-
posite print 3a was made from a positive
copy of the negative plate projected
with the equivalent aperture 5J3 =
26,6/x and the fine-grain positive placed
over and spaced from the copy of the
negative so that it was projected simul-
taneously with an equivalent aperture
53 = 13ju to approximate artificially
the conditions of print 3. The similarity
of the grain structures in prints 3 and
3a is apparent.

The use of positive film with a resolving
power and grain structure equal to that of the
negative film has a more noticeable effect
on the total deviation in the projected
positive print. Table XI was computed
for the same relations of densities and
gammas as listed in Table X, but the
Plus X film was replaced by 5203 fine-
grain negative film and a better pro-
jection lens was used. This combination
of two fine-grain films represents a
condition used for television recording
on 16mm film. The signal-to-deviation
ratios [R]*P computed for a projection
lens with an equivalent aperture 63 =
16.5 fj. are only slightly better in the
highlight range (see curve 3 in Fig.

Table XI. Signal-to-Deviation Ratios for 5203 Fine-Grain Negative Film Copied on
5302 Fine-Grain Positive Film and Sampled by a 16.5-Micron Aperture (73 = 1).

30M

30M* 16.5M 30&16.5//

[R]*p r13/r23 rp/ru

2.77

100

60

14.3

13.9

0.32

1.05

17.2

63

2.3

62

33.5

16

14.4

0.645

1.19

18.8

57.5

2.1

58

30

16.8

14.7

0.77

1.26

20

54

1.9

52

26

17.6

14.5

0.91

1.35

20.3

53.2

1.6

48

24

19.3

15

1.09

1.48

21.8

49.5

1.35

44.4

23.5

20.9

15.6

1.2

1.56

22.5

48

0.91

40

25.5

25.8

18.2

1.365

1.69

23.4

46.2

0.6

36.6

30

32.5

22.0

1.46

1.77

23.8

45.5

0.4

34

38

40.6

27.8

1.44

1.75

23.5

46

0.27

32.4

51

51.0

35.4

1.35

1.68

22.5

48

0.19

32

80

65

50.6

1.1

1.48

22

49

* Negative sampled by $2 = 25/i in cascade with 83 = 16.5/i, equalling £23 = 30/i.

214

March 1952 Journal of the SMPTE Vol. 58

57a) because they refer to a higher
equivalent passband JVe(p).f Amplitude
distribution and equivalent passbands
are shown in Fig. 61. It is pointed out
that this comparison is made on the
basis of unit areas and does not refer to
actual frame sizes!

A reduction print of a 35mm Plus X nega-
tive on 16mm 5302 positive film has
substantially the same relative deviation
per unit area as the above fine-grain
process. The number of grains per
unit area in the reduced negative image
is increased by the ratio of the frame
areas, and [R] *i3 increases, therefore, by
the linear reduction factor 15.7/7 =
2.25. In comparison, the change from
Plus X to 5203 fine-grain negative film
increases [#ji3 2.6 times (see Fig. 52).
In the reduction print the sampling
aperture of the negative film, however,
is increased by the printer lens which
reduces the passband and decreases the
difference in [/?]-values.

The effect of the quality of the projection
lens on the total relative deviation is readily
computed by the above method and is

fit is cautioned that equal [/?]p-values
do not necessarily indicate equal visibility
of the grain structure.

illustrated in Fig. 62 for three values of
the sampling aperture 33 of the lens.
When 63 is large compared to the
effective aperture 52 of the positive film,
the deviations from the negative film
are predominant ([R]w<^. [R]zz)- When
the lens quality is increased (S3 < 52)
the deviations (l/[R]n) from the positive
film increase because of the increased
passband and exceed the deviations
from the negative film which approach
a fixed value determined by the pass-
band of 62. The relative amplitudes
and line number spectra of i^i3 and ^3
for D2 = 0.91 are approximated by
exponential aperture characteristics (Fig.
44, Part I).f

5. Signal-to-Deviation Ratios and Gamma
of Motion Picture Film for
Television Recording

Television images are recorded on
film for the purpose of storing video
signals for use at a later time. To
obtain a perfect duplicate of the original
signals, the overall transfer characteristic
of the system components involved in

f The line number scale is established with
Eq. (19) and Table VII.

F.C

. N

EG

t

F.G

P

DS.

; L

EN

S I

3 =

16

5P

Ne

v

\N +

3

~T

s

Np

i

— 1>
SNN

s1

i

i

p

\

\

s

N

e

•— •*

"*»>,

s^

s

\

v

>»,

^

N

\J

HC

^s

5

\

0~

Ss

5

"^a
^*

^N.

•*<.
1C

- — .

— .
)0

r-fc.
*• •«

••^i.
-— — ,

• — *,

" •

— —

i
If

~ ,

— -
>0

— • —
•

1

r^ .

2C

• — .

)0~

— —

*- -•

•- —

• .

~2.

—

^

LINE NUMBER (N/mm)

Fig. 61. Sine-wave spectrum of deviations in
fine-grain motion picture process.

Otto H. Sehade: Motion Picture Granularity

215

the recording and reproducing process
from signal to duplicate signal must be
linear. This requirement implies that
the product of the point gammas of the
system components must equal unity at
any signal level as stated by

(7^7*,)

(46)

The first product contains the gammas
of video amplifier (7ri) and associated
recording kinescope (7*1), the second
term contains the gammas of the
negative and positive films used in the
photographic process (neglecting lens
flare), and the third term contains the
gammas of the reproducing camera
tube (ye) and its associated video and
correction amplifier (7v2). The combi-
nation of a television process of constant
overall gamma with a motion picture
process of constant overall gamma has
many advantages as pointed out in
Part I.

The overall transfer characteristic
of the motion picture process most
suitable for video recording has a con-
stant gamma (7172) and a relatively
short density range AZ)2 (see Table III,
Part I). It is of interest to determine

the effects of varying gamma and
density range of the negative film on the
signal-to-deviation ratio [K]p, when the
product 7i72, the exposure range of the
negative, and consequently the density
range AZ)2 in the positive film are main-
tained constant. To prevent distortion,
it is required to operate on film charac-
teristics having adequately long constant-
gamma sections. For a numerical
evaluation, the values given in Fig.
52 for fine grain 5203 and 5302 film
will be assumed.

Given an exposure range A log EI =
1.3 of the negative and a density range
AZ)2 = 1 in the positive, the product of
the film gammas must have the value
7i72 = 1/1-3, which can be obtained,
for example, with the values 71 = 0.77,
72 = 1, or 71 = 1.54, 72 = 0.5. Densi-
ties and signal-to-deviation ratios for
these two conditions are listed in Table
XII. The signal-to-deviation ratios
[•^]*23 °f the positive film remain the
same for both conditions because they
are determined by the fixed density
range AZ)2. The values [-ft]*is trans-
ferred from the negative to the positive,
however, change in proportion to l/72
and as the square root of the negative

I

2

3

.

?

'X,

fp

23
)

5203 NEC.

5302 POS.

TOTAL
PROCESS

H.

It
Ne,3

[R]23
§3

Ne23
Wp
Ne(p)

35.2
25
33
26
51.5
33
32.7
29
28.2

25.5
25
16.5
36
25.8
16.5
65.5
18.2
46.2

22.4
25/z

41 L/mm
12.5
8/A

i35L/mf

10.9

88 L/mr

-

0 5

S

N^H

•^

n_

n

\

\

X

2"

>»s

\

\*

'

X3

2^^

S^

\

JX

5s>

~~~*

^

t\i

•-*.

—

•**^

0

r>

"s,

ft

-^

\,

^

***^

-^.

"^

\l

\

s

^^

•w^

"*-

"**^,

•**^1B

s

\

^.

^^

*x.

«

-^.

•— -.

—

*s

^s!}^

°;

,

k*v>

—

-~^

^^-^^

•*~~.

-— ^

=i

n

±

±3

If

•irT

-=—

~\*

-^^
n

•~— .

-*^

— — i

B

?r

)O

2!

^r

LINE NUMBER (N/mm)

Fig. 62. Effect of projection lens quality on sine-wave spectrum
of deviations in fine-grain motion picture process.

216

March 1952 Journal of the SMPTE Vol. 58

II

ON

D.
*

CM so 00

ON

e.
*

oo oo so m TJ-

II

O

g

oo o\ T-H co m

T-H TH CM CM CM

0

g

•^- so ON CO O
CM CM CM CO Tf

***

CO

CO

:£

T-t

***

so CM

m

ON

. 1

oo T-H m CM so

*»

00

1 '

T-H CN CM CO in

2

5

in f- CM ON sO
CM CM CO CO m

d

#**

m

3

%

oo m co co CM

o

•Jtr^

JTT1

m o oo so oo
t*^ o h~ so in

"0

ct;

T-H T-I

8

^

CM

O

<*-

1-1 II

•5

<N

m m m m in

£

#

Q
m m m m m <i

d d o d d ^

«

CM t^~ CM t^ CM ^

II

c7

H

CM T-H T-H ^^ O "^

"

d

*

to in in m m "
CM r-- CM h- CM ^

5f

m m m Q
CM o r~- m CM <j

T-' ,-.' O O O ^

m

rt

^

O O T-H T-H CM '

0

M

ON

*_,

in in in in in

CM h- CM r- CM «

*O

T-H

a

.

Q

•a

O

*

"Vt^oo

^H

O O T-H' T-' CM "^

i

CO

5

e.

CO T-H

.s

ON

*"

oo T-I m CM so

o

CO

5

rj- m r- ON CM

CN CM CM CM CO

^

—

T-H CM CM co m

J

ss

oo

£

8

*""

CO T-H u-> CM ON
•^- CO CM CM T-H

ON

k

m h~ CM ON so
CM CM CO CO m

o

—

m

*~*

m

1

8

g

r-~ CM O rf ON

oo so m -^ co

%

r- CM O •<*• ON

oo so m "t co

d

CM

$

*

in in m m m „
CM r^ CM r- CM Q

o

r^.

m

<*H

c

Q

CM ,-•' T-H O O <

CO

S

oo so m ^ co

.2

2

1

i

m

«

<o

bo

CM CM CM CM CM

m ' m in
CM o r- m CM

2

<*-

*

II

m m m Q
CM o r-~ m CM <]

9

3

-H'-H'OO'O'P

X

^

r^

,2

d

o

S>

#

m o m m "
CM m r» o CM _

*

in o m m I'
CM m r- o CM _

^

Q

d d d T-.' T-4* ^

^

O O O T-H T- v^

H

tf

CO

C£

CO

1

be

,3

0 T-H

JT

O T-H

Otto H. Schade: Motion Picture Granularity

217

density DI*. The values [R]*n and
[/?]% increase, therefore, when the
negative gamma increases. This relation
is equally true for other values of the
density range AZ)2* in the positive print
as shown by Table XII and the graphs
in Fig. 63 for AZ)2 = 2 for two conditions
7i = 0.77, 72 = 2, and 71 = 1.54,
7j = 1. The contribution by the
positive film deviations becomes rela-
tively larger when the negative gamma
is increased, and the curve of [R]*P for
the process cannot improve beyond the
limit set by [#]*2s. After conversion
into video signals the signal-to-deviation
ratios [R] *p of the photographic process
are modified by the gamma of the
television camera chain to the value

[/?]*/ = [*]%/(7c7,2) (47a)

It is emphasized that this expression
does not take into account the aperture
effects of television components and is,
therefore, not an electrical signal-to-noise
ratio. For a given camera chain, how-
ever, the values [/?]%' have a direct
relation to the electrical fluctuation
signals caused by the photographic
process and Eq. (47a) can, hence, be
used to indicate the relative performance
of the photographic link in the recording
system. According to Eq. (46) the
product l/(ycyvj may be replaced by
^(7172) because the product (7^7^)
associated with the value [R]*p at any
one value of density DI will be left
unchanged and may, hence, be replaced
by a factor K. This substitution results

in the more useful expression

(47b)

which shows the influence of varying the
product 7172 of the photographic process
on the signal-to-deviation ratio obtained
in the video channel. Table XIII lists
the values obtained by Eq. (47 b) for
the four conditions shown as curves 1
to 4 in Fig. 63.

Inspection of Table XIII shows that
curves 1 and 3 give higher signal-to-
deviation ratios [.ft]*?' at highlight
signal levels than curves 2 and 4 indicat-
ing preference for a high negative gamma
(71 = 1.54). At the black level the
longer range positive films 3 and 4 are
preferred because the value [R]*p on
curves 1 and 2 are seriously limited by
deviations [/?]23 contributed by the
positive film (see Table XII). A 16mm
positive film having a finer grain than
the 5203 negative ([ft]*23 increased by
at least a factor of two) is, therefore,
desirable for video recording because it
practically eliminates the limitation by
[#]*23. Good ratios [/?]% in the high-
and medium-transmittance range of
the motion picture film are most im-
portant to reduce the visibility of film
grain.

When a sharp kinescope image with
separated raster lines is recorded on
film the equivalent rectangular cross
section s = V/Ne(a) of the lines in the
negative film may be smaller than the
raster line distance V/Nr, where Ne(a) —
cascaded aperture passband of kinescope,

Table XIII. Signal-to-Deviation Ratios of Photographic Constant Gamma Processes

for Video Recording.

[*]%

[*]%'(

!Eq. (47b))

Curve
No. in
Fig. 63

A* = 0.25, 1.25,

2.25

Highlight level
(D2* = 0.25) (1

Black level
>i* = 1.25, 2.25)

7i7z

AD,*

1

2
3

4

40.4 24.8
32 24
25.8
18.4

18.2
17.1

31
24.6
39.8
28.4

19.1
18.5
28
26.2

0.77
0.77
1.54
1.54

1
1
2
2

218

March 1952 Journal of the SMPTE Vol. 58

camera lens and negative film; V =
vertical frame dimension, and JVr =
number of raster lines in V (see Part I,
Sec. B8, p. 160). In this case the
exposed frame area and the number of
utilized grains in the negative film are
reduced by the factor K\ = Nr/Ne[a)
resulting in a reduction of the normal
signal-to-deviation ratio [R]i from the
negative to [R]i\/Ki. After transfer
of the image to the positive film the
cascaded value JV«(0) includes the aper-
ture of the positive film. The new
factor K2 changes, therefore, the ratio
[/2]2 from the positive which becomes
[K\j\/Kto but it does not alter the
above value \K\\\/K\. The signal-to-
deviation ratios in a video film-recording
process have normal values when K\
and Kz are equal or greater than unity,
but they are reduced to lower values
than given in the previous discussion
when the factors K± and Kz are smaller
than unity.

The conditions for optimum signal-
to-deviation ratios may be summarized
as follows:

(a) In both negative and positive
films the minimum densities should be
as low as possible and the density ranges
(A/)) should be as large as possible
without conflicting with the operating

NEC. EXPOSURE RANGE
A LOG E| = l.3
F.G. NEC. & POS. FILM
(SEE TABLE XII)

10

2.0

1.5 l.O 0.5

DENSITY (02) ABOVE BASE

requirements of the television system
which limits the maximum density range
A/)2 in the positive and makes a constant
product 7i72 desirable for adequate
exposure latitude. The value AZ)2 may
be varied within wide limits when the
positive film has a substantially finer
grain than the negative.

(b) The negative gamma should be
as high as possible, a high gamma being
obtained by selection of a film type with
a larger grain number and not by over-
development of a low-gamma film
which may give a higher gamma by
increasing the grain size. In practice
the requirement for high gamma is
tempered by the decreasing exposure
lattitude, a short range A log EI per-
mitting, in general, a higher negative
gamma.

(c) The positive film should have a
finer grain than the negative film (by
a factor of two or more).

6. Luminance Fluctuations and Optical
Passbands of Motion Pictures

Twenty-four different phases of the
deviations in the positive film are shown
every second in a motion picture pro-
jection. The static deviations in the
film frames are transformed into lumi-
nance fluctuations and because of the
persistence of vision,
the grain structures
in successive fields are
integrated to some
extent by the eye.
The deviation ratio
[R]p of the process
changes to the optical
luminance fluctuation
ratio

[R]0 = s[R]P (48)

Fig. 63. Deviation
characteristics of con-
stant-gamma film proc-
ess for video record-
ing.

Otto H. Schade: Motion Picture Granularity

219

I
a

J

II

S< OT CO M M M

DOS C C c G

C C o V 4) CJ l)

uux S'JS'J

SlSLl 1111

5 5 J J J JT J

C d bo bo bo bo bo

3 4 £ £ 3 S £

ili iiii

d d d d d d d

fa fa fa PL* fa fa __'

bb bb bb bb bb

s^ s^ ^ 2 c c c

J J d (d (d d id

P | Q, r~T. [j. FT . PT | rT .

ON CN ^* m m oo m

in in to NO ^ co ^-H

NO NO

CO CO*
CN1 <N

000

^-H NO T->

m T*- CN

°i

o

m ob m _

•r-i -rH <N ON

ON ON

$

T-H T-H m

"^••^•oo P. oo co ON co

T-I^H^H bO^,CN)CSlCO

The storage factor s depends on the ratio
of the effective visual storage time Ts
to the frame time Tf and is given in
first approximation by

s ^ (TJTf)\

The value of Tt is left open for a
later discussion, but it can be stated
that a value s slightly larger than unity
is indicated for the conditions in motion
pictures.

The foregoing evaluation of deviations
in motion picture film has furnished
values which refer to an area a specified
by the effective sampling aperture of
the process. The effective optical pass-
band has been referred to a unit film
area (1 sq mm) as expressed by JV. in
lines per millimeter. The [R] -values
for the film and lens combinations shown
in Figs. 57 and 63 apply to all frame
sizes of motion picture film. The optical
"frequency" characteristic and the
equivalent passband JV«, however, must
be referred to the particular frame size
and are obtained from Figs. 59, 61 and
62 by multiplying the unit line number
by the vertical frame dimension V in
millimeters. (V = 15.7mm for 35mm
film and V = 7mm for 16mm film.)
The granularity in a motion picture frame
is determined by the square root of the
total grain number in the picture frame
area (see Part I). Expansion of the
round sampling area a to the frame area
A = VH furnishes the fluctuation ratio
with respect to the film frame

[R]f = s[R]9(A/a)\

= j[/2]p(F///0.257r302)i (49)

and with 50 from Table VII:

[R]f = s[R]pNe(P)(H/V)^ (50)

The signal-to-deviation ratio [R]p in

f The numerical value computed with the
value d0 from Table VII differs by a few
per cent from this value because of the
synthesis of the deviation spectrum from
the sine-wave response characteristic (see
Sec. 3). Equation (49) is exact when the
factor given for aperture #3 in Table IV
derived for the sampling equivalent JV0
is used.

220

March 1952 Journal of the SMPTE Vol. 58

this equation still refers to the equivalent
sampling aperture (5Po) of the process
(see Eq. (45)), or its equivalent passband
JV«(P). The ratio H/V is the aspect
ratio of the film frame. Significant
quantities for comparing granularity in
35mm and 16mm pictures are the ratio
[R]P and the line number spectrum
(see Figs. 59, 61 and 62), which is
indicated by the equivalent passband
JV.. Both are needed to define image
quality and to predict the appearance
and relative visibility of fluctuations to
the eye. The product [R]PNc(P) combines
this information into a single objective
figure of merit for the granularity of the
process. The squared value [R]2PNze(P)
expresses by definition the number of
samples of energy or matter in the
equivalent sampling area a (indicated
by its reciprocal JV«2) and is therefore in

agreement with fundamental principles.
The significant quantities for a number
of motion picture processes are sum-
marized in Table XIV.

When the sampling apertures of the
various photographic processes are ad-
justed to have the same value, indicated
by equal equivalent passbands JV«(P),
the products [R]PNe(P) remain sub-
stantially unaffected. This adjustment
can be made, for example, by a change of
the projection lens quality (S3) or by
adding an additional aperture process
(64) in cascade such as the process of
vision. Leaving a discussion of the
subjective impression of graininess to a
later publication, it can be expected that
the objective measure of granularity
[R]PNe(P) will place photographic proc-
esses in an order which is in agreement
with visual observations.

APPENDIX
STATIC DEVIATION MEASUREMENTS ON PHOTOGRAPHIC FILM

The optical arrangement shown in
Fig. 64 is similar, in principle, but not
nearly as elaborate as the one used by
Jones and Higgins.2 The sampling

aperture is the image of aperture A
formed in the film plane by a coated
14-mm objective I. The film sample is
mounted on the stage of a Leitz metal-

APERTURE "A

LIGHT SOURCE

MICROSCOPE
STAGE

±=T

OBJECTIVE IT

MULTIPLIER
PHOTOTUBE
TYPE 5819

i ! i

M

'i/

~^r

i— i

/

/

R '
>E
9

-RETRACTABLE
EYE PIECE

II

APERTURE
EFFECT OF

GRAIN
STRUCTURE

Fig. 64. Apparatus for deviation measurements on photographic film.

Otto H. Schade: Motion Picture Granularity

221

lurgical microscope. A second objective
II (8mm) below the stage is focused on
the aperture image with the film moved
out of focus but in the light beam. The
film grain is then brought into focus by
vertical stage adjustment as observed
through the observation eye piece.
When the eye piece is retracted the light
strikes the photocathode of a multiplier
phototube giving an electric current
proportional to the total light flux trans-
mitted through the sampling area a on
the film. The sampling area a is
adjusted by the aperture size A and dis-
tance from objective I and measured
by removing the phototube and inserting
a second ocular to project a magnified
image (1000 times) on a ground glass
screen (not shown). At any given film
density 30 to 60 flux readings are taken
along arbitrary cross sections of the
film in groups of 10. The readings are
averaged and the film is moved to an
unexposed area to measure the trans-
mittance ratio T/TO. The deviations
Ar from the mean values are tabulated
to determine the rms value [Ar] and
the ratio <rT (Eqs. (13) and (33)).

An optical observation of the aperture
image with the film in place must be
made to check the effective sampling
area a. Diffraction or diffusion effects
in the developed emulsion introduce an
exponential aperture effect which can
introduce considerable errors when the
sampling aperture is small because it
changes the actual sampling aperture
to a cascaded value with an effective
area larger than the optical image
obtained without the test specimen (see
insert drawing in Fig. 64). The equiva-
lent aperture diameter of the developed
film is considerably smaller than that of
the undeveloped film (see Table VIII)
because of the much higher light ab-
sorption by the developed silver grains.
The equivalent film aperture increases
with emulsion thickness for objectives
(I) of shorter focal length and for aper-
ture shapes other than round, and is, in
general, proportional to the resolution

of the film type. To prevent the
aperture error and to satisfy the re-
quirement <rr < 0.1 (see Eq. (35)) the
diameter of the sampling aperture a
must be at least three times larger than
the equivalent aperture diameter of the
developed film type. The sampling
apertures used in the measurements
reported in this paper fulfill this re-
quirement. Many discrepancies re-
ported in the literature for small sam-
pling apertures are readily explained by
the aperture error.

Acknowledgments

The author wishes to acknowledge
the helpful criticism and contributions
of W. A. Harris of the RCA Tube Dept.,
Harrison, N.J., and Dr. D. O. North
of the RCA Laboratories, Princeton,
N.J., in the analytical evaluation of
equivalent aperture passbands (A"«).
He also wishes to thank W. H. Rivers
of the Eastman Kodak Company for
supplying representative film samples for
deviation measurements.

References

Part I of this paper, "Image structure and
transfer characteristics," Jour. SMPTE.
56: 131-177, Feb. 1951.

1. O. H. Schade, "Electro-optical charac-
teristics of television systems: Intro-
duction," RCA Rev., 9: 5-13, Mar.
1948.

"Part I — Characteristics of vision and
visual systems," ibid., 13-37, Mar. 1948.
"Part II — Electro-optical specifications
for television systems," ibid., 245-286,
June 1948.

"Part III — Electro-optical characteris-
tics of camera systems," ibid., 490-530,
Sept. 1948.

"Part IV — Correlation and evaluation
of electro-optical characteristics of imag-
ing systems," ibid., 653-686, Dec. 1948.

2. L. A. Jones and G. C. Higgins, "Photo-
graphic granularity and graininess.
VII. A microphotometer for the meas-
urement of granularity," /. Opt. Soc.
Am., 41: 192-200, Mar. 1951. (This
article also gives a list of earlier papers
on this subject.}

222

March 1952 Journal of the SMPTE Vol. 58

Color Negative and Color Positive
Film for Motion Picture Use

By W. T. HANSON, Jr.

A color film for use in an ordinary 35mm motion picture camera is described.
This film contains colored couplers which, upon development to a negative,
lead to three-color negative records which are almost fully corrected. The
development procedures and the sensitometric characteristics are described.
The spectral-density characteristics of the individual images are included.
This film is printed on a color positive film. The spectral-density charac-
teristics of the dye images obtained in the positive, the development conditions
and the sensitometric characteristics of the positive are given. The printing
may be done on an ordinary continuous contact printer. However, scene-
to-scene color-balance changes require more complicated equipment. The
sensitometric characteristics of the sound-track image and the method of
developing this image are described. The color positive film may also be
used for making prints from black-and-white color-separation negatives.

I

NTEGRAL TRIPAGK three-color
subtractive films have been in use in
the motion picture industry for a good
many years. These films include Mono-
pack, Ansco Color, and 16mm Koda-
chrome in this country, and the Agfa-
color negative-positive film in Germany.
The present paper describes a new
negative color film and a new positive
color film for use in making 35mm
motion pictures. The negative film
has certain features which have not

Communication No. 1457 from the Kodak
Research Laboratories, a paper presented
on April 27, 1950, at the Society's Con-
vention at Chicago, 111., by W. T. Hanson,
Jr., Eastman Kodak Co., Kodak Park
Works, Rochester 4, N.Y.

previously been used in the motion
picture field.

The Negative Color Film. The negative
film is called Eastman Color Negative
Safety Film, Type 5247. It contains
couplers similar to those used in the
Kodacolor process described in 1942.1
Each coupler is dissolved in an oily
liquid which is, in turn, dispersed in an
emulsion. The structure of the film is
shown in Fig. 1. It can be exposed in
an ordinary 35mm motion picture
camera. The ASA speed rating of the
film is 16 and it is balanced for exposure
in daylight or with high-intensity arcs
with the Brigham Y-l filter. Being a
negative film, it has somewhat more

March 1952 Journal of the SMPTE Vol. 58

223

A O A o A° A O A O AOA A O A O A O A
O AO AO AOOA AO AO A°°o A AOA O

A»A«A«AAA»A»A»AA»A« A
• A« A »A» A»AA« A*A« A»A» A*

A»A A»A*A«A« A A*A«A» A • A •
• A»A» A* A A«0A«A«A* A»«A

/ / / ,

Gelatin Overcoating

Blue-Sensitive Emulsion and Uncolored
Yellow-Dye Coupler

Yellow Filter Layer

BJue-and Green-Sensitiv* Emulsion and

Yellow Colored Magenta-Dye Coupler
Gelatin Interlayer
Blue-and Red-Sensitive Emulsion and

ReddisT^Orange Colored Cyan-Dye Coupler
Sub-Stratum

Safety Support

Anti -Halation Backing

Fig. 1. Schematic cross section of Eastman Color Negative
Safety Film, Type 5247.

exposure latitude than most reversal
films. One stop overexposure or under-
exposure can be tolerated with no
significant degradation of quality, and
two stops overexposure will give a fairly
satisfactory result. However, as is the
case for practically all color films, the
lighting contrast ratio should be from
one to two or three and should seldom
exceed one to four except where a
special effect is desired.

The negative film can be processed
in any conventional type of processing
machine which has a sufficient number
of tanks for the steps that are required,
and which has certain tanks that are
resistant to the bleaching solution. The
processing steps are shown in Table I.
The developing agent in the color
developer solution is a derivative of
/>-phenylenediamine which does not
normally produce "sensitization" in
human skin. Its properties in this respect
are similar to the well-known Kodak
Elon Developing Agent.*

The exact formulas for the processing
solutions must be adjusted for the various
processing machines of different design
and cannot be specified quantitatively.
Information based on the most recent
experience is available through the Motion
Picture Film Division of the Eastman
Kodak Company.

Color Correction With Colored Couplers.
The dyes used to form the images in
subtractive color processes have absorp-
tion characteristics which lead to un-
desirable results when a color trans-
parency is duplicated or when a color
negative is printed to a color positive.
The cyan-dye image, for example, which
is supposed to absorb only red light,
absorbs some of the blue and the green
light. The effects of such overlapping
absorptions can be minimized or possibly
eliminated by the use of separate masks,
as described by Miller.2 However, the
procedures involved in using separate
masks are extremely tedious. A much
more direct and simple method of
eliminating the effects of the overlapping
absorptions of the dyes is the use of
colored couplers.2"4 Couplers of this
type are used in the red- and green-
sensitive layers of Eastman Color Nega-
tive Film.

The coupler in the red-sensitive layer
forms the cyan dye and is colored orange.
It has some absorption in the blue and
green regions of the spectrum but trans-
mits freely in the red region where the
cyan dye absorbs. When this coupler
is converted to the cyan dye, the orange
color is destroyed. Thus, when a film
is exposed and developed, the areas
which receive exposure are developed

224

March 1952 Journal of the SMPTE Vol. 58

to a cyan dye, the orange-colored coupler
being destroyed. The unexposed areas,
in which no development takes place,
retain their orange color. Areas of
intermediate exposure contain some
cyan dye and some residual orange
coupler. This results in a cyan nega-
tive image and, in the same layer, an
orange-colored positive image composed
of residual coupler.

The spectral characteristics of various
density levels of such an image are
shown in Fig. 2. In the red region of
the spectrum, successive areas have
increasing amounts of density, owing to
the increasing amount of cyan dye.
In the green and blue regions of the
spectrum, successive areas have essen-
tially the same density. Here the
increasing densities caused by the in-
creasing amounts of cyan dye are just
canceled by the decreasing densities of
the decreasing amounts of residual
orange-colored coupler. This series of
images is expressed in the form of
H & D curves in Fig. 3. The densities
measured with red light increase with
the logarithm of the exposure and give
the normal H & D curve of the cyan
image. The densities measured with
blue and green light, however, are
essentially constant at all levels of
exposure. This is the desired charac-
teristic of the cyan-dye image.

The green-sensitive layer of Eastman
Color Negative Film contains a yellow-
colored coupler, which, on development,
forms a magenta dye. Here again,
exposure and development lead to two
images in the layer — a magenta nega-
tive image resulting from the destruc-
tion of the yellow color in the regions in
which exposure and development take
place, and a yellow positive image
formed by the residual coupler in the
unexposed regions.

The spectral-density characteristics
of a series of levels of this image are
shown in Fig. 4. In the green region of
the spectrum, successive areas have
increasing green density, owing to the

Table I. Processing Steps for Eastman
Color Negative Film.

Step

Time

Temperature
70 F

Carbonate pre-
bath

1-2 min

Not critical

Negative color
development
Stop bath
Water wash

24-27 min
4 min
4 min

Critical
Not critical
Not critical

Bleach

8 min

Not critical

Water wash

4 min

Not critical

Fixing bath
Water wash

4 min
8 min

Not critical
Not critical

Wetting agent
Drying

1 min
15-20 min

Not critical

Lacquering — bead application (both sides)

Note: Proper development time is a
function of agitation conditions and of
the particular machine being used. De-
veloper temperature should be controlled
to ±0.25 F; other solutions, ±2 F.

increasing amounts of magenta image
dye. In the short-wavelength blue
region of the spectrum, successive areas
have decreasing density, owing to the
decreasing amounts of the residual
yellow-colored coupler. However, in
the middle of the blue region, at ap-
proximately 460 m/i, the decreasing
densities of the yellow coupler image are
just equal to the increasing densities of
the magenta-dye image so that the two
images cancel. In the red region of
the spectrum, there is practically no
density.

These images are expressed in terms
of H & D curves in Fig. 5. The densities
measured with green light increase with
increasing log exposures to form the
normal H & D curve. Densities meas-
ured with blue light are essentially
constant at all levels of exposure. A
glance at Fig. 4 will indicate that these
blue densities are measured with a
filter which has a narrow band of trans-
mission at around 460 m^t.

Spectral-density curves of a series of

W. T. Hanson, Jr.: Color Negative and Positive

225

2.0

a 1.0
O

400 500 600

WrfVelength in Millimicrons

700

Fig. 2. Spectral-density curves for a series of concentra-
tions of the cyan image of Eastman Color Negative Film.

2.0

1.0

0.0

red

°blue ond Dgreen

Log E

Fig. 3. H &D curves for the cyan image of
Eastman Color Negative Film.

226

March 1952 Journal of the SMPTE Vol. 58

2.0

0)

Q

o 1.0
o

"o.
O

400 500 600

Wavelength in Millimicrons

700

Fig. 4. Spectral-density curves for a series of concentrations
of the magenta image of Eastman Color Negative Film.

2.0

1.0

0.0

'green

;blue

Log E

Fig. 5. H &D curves for the magenta image of Eastman Color
Negative Film. (Densities to red are insignificant.)

W. T. Hanson, Jr.: Color Negative and Positive

227

2.0

400 500 600

Wavelength in Millimicrons

700

Fig. 6. Spectral-density curves for a series of concentrations
of the yellow-dye image of Eastman Color Negative Film.

2.0

0.0

Log E

Fig. 7. H &D curves for the yellow image of
Eastman Color Negative Film.

228

March 1952 Journal of the SMPTE Vol. 58

1.0 —

400

700

500 600

Wavelength in Millimicrons

Fig. 8. Spectral-density curves of a series of concentrations

of the dyes occurring in the reproduction of a scale

of neutrals on Eastman Color Negative Film.

amounts of the yellow-dye image are
shown in Fig. 6. In this case, no colored
coupler is used. The H & D curves of
this image are shown in Fig. 7. The
insignificant red densities of the yellow-
dye image have been neglected, but
there is significant green density, as
shown by the curve. It would be de-
sirable to correct this by means of a
colored coupler, but to date such correc-
tion has not been possible. Such a
correction would improve the reproduc-
tion of yellows and greens.

The sum of all three of these images
is shown in Fig. 8 by the spectral-density

curves of the reproduction of a scale of
neutrals on Eastman Color Negative
Film. Obviously these spectral-density
curves do not represent visual neutrals.
The reproduction of neutrals is quite
orange in color because of the presence
of the orange- and yellow-colored
couplers. This orange overcast which
must occur in all pictures on Eastman
Color Negative Film is eliminated in the
printing process by the proper sensitiza-
tion in the print film and the proper
selection of light intensity in the red,
green and blue regions of the spectrum.
However, after such correction has been

W. T. Hanson, Jr.: Color Negative and Positive

229

3.0

2.0

1.0

0.0

Log E

Fig. 9. H &D curves for Eastman Color Negative Film.
Exposure, intensity-scale sensitometer, 1/25 sec;
Illumination, daylight quality;
Density, printing density.

AO AOA AO AO AOAAOAOAOA
OAOAoAQA AO AO A°OOA AOAQ1

A O A O AOAO A O AO A A°-. A O A O A
OAOAQAoA AOAO AO^A AOAQ°

AOAOA AO&O AOA A AO A ° A
O AO AO A OA AO AO AO OA AOAQ°

Gelatin Overcoating

Green-Sensitive Emulsion and Uncolored

Magenta Dye Coupler
Red-Sensitive Emulsion and Uncolored

Cyan Dye Coupler
Blue Interloper
Blue-Sensitive Emulsion and Uncolored

Yellow Dye Coupler
Sub - Stratum

Safety Support

Removable Anti - Halation Backing

Fig. 10. Schematic cross section of Eastman Color Print Safety Film, Type 5381.
(All layers contain a magenta dye.)

230

March 1952 Journal of the SMPTE Vol. 58

made in the printing operation, the
final result is free of the defects intro-
duced by the overlapping absorptions of
the cyan and magenta dyes in the
negative.

The H & D curves for the reproduc-
tion of a scale of neutrals on Eastman
Color Negative Film is shown in Fig.
9. The three curves represent the
densities of the neutral scale as measured
with red, green and blue light. These
measurements were made through filters
with a physical densitometer using a
photomultiplier tube with an S-8 sensi-
tive surface. The filters were selected
according to the technique described by
Williams,5 with the intention that the
densitometer would measure the densi-
ties of the image in the same way that
the image would print onto the color
positive film. Thus, the curves are
expressed in terms of printing density.
Since these curves are in terms of integral
printing densities, they do not represent
the separate characteristics of the cyan,
magenta and yellow images, but the sum
of these three. However, to a first
approximation, the red density curve
represents the cyan-dye image; the
green density curve, the magenta-dye
image; and the blue density curve,
the yellow-dye image. Here again, the
orange-colored overcast of the Eastman
Color Negative image is indicated by the
high densities to green and the higher
densities to blue.

The Positive Color Film. The positive
film is called Eastman Color Print
Safety Film, Type 5381. It also con-
tains couplers which are dissolved in an
oily liquid and dispersed in the emul-
sions. In this case, however, the
couplers themselves are not colored.
The structure of this film is shown
diagrammatically in Fig. 10. The
first emulsion layer is a fairly fast
emulsion containing the yellow-forming
coupler. It is desirable that this layer
be on the bottom, since the yellow-dye
image contributes the least to overall

Table II. Processing Steps for Eastman
Color Print Film

Step Time

Temperature
70 F

Carbonate pre-

bath 1-2 min

Not critical

Positive color

development 12-15 min

Critical

Stop bath 4 min

Not critical

Water wash 4 min

Not critical

Bleach 8 min

Not critical

Water wash 2 min

Not critical

Sound-track

development

(strip appli-

cator) 10 sec

Not critical

Water wash 2 min

Not critical

Fixing bath 4 min

Not critical

Water wash 8 min

Not critical

Stabilizing bath 1-5 sec

Not critical

Drying 15-20 min

Edge waxing (both sides)

Note: Proper development time is a
function of agitation conditions and of
the particular machine being used. De-
veloper temperature should be controlled
to ±0.25 F; other solutions, ±2 F.

picture sharpness and the lower image
in a multilayer film is the least sharp.
The emulsion is fairly fast so that the
negative film with the high blue densities
described previously can be printed.
The next layer is a blue-dye interlayer.
The major purpose of this layer is to
absorb red light transmitted by the
upper layers. Within all three emulsion
layers, light is scattered by the silver
halide grains. Scattered red light, as
well as direct red light, exposes the red-
sensitive layer and thereby causes a
decrease in image sharpness. The
blue-dye layer underneath the red-
sensitive layer absorbs light transmitted
by the upper layers and prevents this
from being scattered and reflected back
to the red-sensitive layer from the
bottom layer. There is still some
residual scatter in the top two layers
which has an effect on sharpness. The
next layer is the red-sensitive emulsion

W. T. Hanson, Jr.: Color Negative and Positive

231

3.0

2.0

1.0

- Yellow
Magenta

- Cyan

Viewing light 4000 K

Log E

Fig. 11. H &D curves for Eastman Color Print Film.
Exposure, intensity-scale sensitometer, 1/25 sec;
Illumination, tungsten light plus color-correction filters;
Density, equivalent neutral density, calculated from
integral density measurements.

containing the cyan coupler. Next is
the green-sensitive emulsion containing
the magenta coupler. Over this is a
gelatin overcoat to protect the film
against abrasion. Throughout the entire
film is a magenta dye. This prevents
green light from being scattered through-
out the layers and decreasing the picture
sharpness of the magenta-dye image.
The processing steps are shown in

Table II.* The developer used for
processing the color positive is a deriva-

* The exact formulas for the processing
solutions must be adjusted for the various
processing machines of different design
and cannot be specified quantitatively.
Information based on the most recent
experience is available through the Motion
Picture Film Division of the Eastman
Kodak Company.

232

March 1952 Journal of the SMPTE Vol. 58

3.0

2.0 -

Neutral to 4000 K
Illupiination

400

500

600

700

Wavelength in Millimicrons

Fig. 12. Spectral-density curves for the cyan, magenta and yellow dyes

in Eastman Color Print Film and the neutral they form. Neutral

density of 1.72 in 4000 K blackbody illuminant.

tive of /?-phenylenediamine which is
known to produce "sensitization" in
human skin. Repeated contact with
the developer will lead to "dermatitis."
Great care must therefore be exercised
in handling this solution.*

After exposure to a step tablet on a
sensitometer (the light source being
adequately balanced for a particular
emulsion), the final processed film may
be expressed in terms of the three normal
H & D curves of a color film. In order

* Specific precautions which must be
followed are available from the Motion
Picture Film Division of the Eastman
Kodak Company.

to describe adequately the characteristics
of each of the dye images, the densities
should be expressed in terms of "ana-
lytical" density. For the curves shown
in Fig. 11, the densities were read on a
physical densitometer through red, green
and blue filters and the integral densities
converted to equivalent neutral density.6
Spectral-density curves of the three
image dyes are shown in Fig. 12. These
are shown in the amounts which make
a neutral density of 1.72 to a blackbody
illuminant with a color temperature of
4000 K. Spectral density of the neutral
is fairly selective which results in changes
in the appearance of the print when the
illuminant color is changed. If a print

W. T. Hanson, Jr.: Color Negative and Positive

233

M Mirror

H Heat-Absorbing Glass

F Filler

T Timing Shutter

A Printing Aperture

Fig. 13. Schematic illustra-
tion of position of niters
in light path in Bell &
Howell Model D printer.

is properly balanced for arc projection,
its appearance will change if a tungsten
light source is used.

Printing the Negative to the Positive. As
in the case of any integral tripack
printing operation, no registration prob-
lems are involved in printing Eastman
Color Negative Film onto Eastman
Color Print Film, and printing can be
done on a continuous contact printer.
The main requirements are that a light
source of sufficient intensity be available
and that some means for altering the
spectral quality of this light be supplied.
Satisfactory results have been obtained
using a Bell & Howell Model D printer,
with the light source adjusted according
to the recommendations of Kunz, Gold-
berg and Ives.7 A slot has been cut
in the lamphouse casting so that a filter
pack may be inserted into the light path
at a position shown schematically in
Fig. 13. Two pieces of Pittsburgh heat-
absorbing glass (No. 2043), 0.1 in. thick,
are in the position shown. With a
properly exposed negative and a typical
print-film raw stock, a 400-w, 115-v
lamp operating at 105 v gave satis-
factory print density at printer point
"12," with the printer operating at the
rate of 45 ft/min.

While printing with a continuous
tungsten light source, adjusted in color
quality by means of color-compensating

0)

D

GO

c

<D
0)

k.

O

•D
<D

a:

c

0>

0)

k.

e>

Q>

D

CO

Fig. 14. Filter arrangement for
"narrow-band" printing: red, Corning
#2408, 1 mm; green, Corning #3486 +
#4303, 1 mm; and blue, Corning #5113,
1 mm.

filters, usually gives a satisfactory print,
it does not give the maximum obtain-
able color quality. The dyes and the
colored couplers in the negative have
fairly narrow absorption bands so that
the color quality obtained in a print
is quite sensitive to changes in the
sensitivity distribution of the print film
or to changes in the spectral quality of
the printing light. Reference to Figs. 4
and 6 shows that the blue density charac-
teristics of the yellow and magenta
images change appreciably in the region

234

March 1952 Journal of the SMPTE Vol. 58

of 410 to 460 m/x. The results obtained
in the final print are affected noticeably
by changing the quality of the printing
light in this region of the spectrum.
A Kodak Wratten Filter No. 2B, or
similar violet-light absorber, should
always be used in the printing operation.
Similarly, reference to Figs. 2 and 4 shows
that the green density characteristics of
the cyan and magenta images vary con-
siderably in the region of 520 to 560
m/t, and the density of the cyan image
varies appreciably within the red
region of the spectrum. Variations of
the quality of the printer light in these
spectral regions will lead to varying
results.

Better control of the color quality
of the print can be obtained if the light
source used for printing is composed of
three spectral bands in the blue, green
and red regions of the spectrum. This
can be accomplished by mixing the light
transmitted by three filters, for example,
by introducing into the light beam on
the Bell & Howell Model D printer a
filter which is composecj, of a series of
strips of glass as shown in Fig. 14. This
filter is placed in the light path in the
printer at the same position occupied
by the color-compensating filters as
shown in Fig. 13. This strip construc-
tion of the filter provides uniform
illumination at the printing aperture.
The position of the filter in the beam
and the alignment of the mirror must
be critically adjusted in order to ensure
uniformity at the printing aperture.

In printing by the technique just
discussed, it is possible to use the regular
timing shutter on the Bell & Howell
printer for introducing density correc-
tions from one scene to the next. How-
ever, it is not possible to make changes
in the color quality of the illumination
between scenes in order to correct the
color balance of successive scenes. In
motion picture color printing, such
"color timing" is necessary. This can
be accomplished by the use of three
light sources in the optical system of a

printer such as the Bell & Howell
Model D. This is shown schematically
in Fig. 15. The light from each lamp
passes through a filter, and then the
three light beams are combined at the
printing aperture. With this system,
the narrow spectral bands of light as
described above will be obtained. In
addition, the intensity of each of the light
sources can be adjusted separately by
known means, such as varying the volt-
age, use of diaphragms, or neutral-
density filters, and thereby effect the
proper color timing. This type of
light source has been described by
Bornemann and McKusick.8

Eastman Color Print Film can also
be printed from color-separation nega-
tives. In this case, the printing must
be done in a step printer with adequate
registration pins. Negatives obtained
from Eastman Multilayer Stripping
Negative Safety Film, Type 5249, de-
scribed by Capstaff,9 are a typical
example. Similarly, separation positives
and duplicate separation negatives may
be made from Eastman Color Negative
and these, in turn, printed onto Eastman
Color Print Film or some other color
film.

Sound Track. The sound track on
Eastman Color Print Film is developed
by edge application of a reducing agent*
after the rehalogenizing bleach bath.
In the color developer, the sound-track
and picture images are developed simul-
taneously to dye and metallic silver.
The next step in the process removes
all of the unexposed silver halide in
both picture and sound-track areas.
Following the fixing bath, the bleach
bath converts all of the developed silver
image to silver bromide. In the sound-

* The exact formulas for the processing
solutions must be adjusted for the various
processing machines of different design
and cannot be specified quantitatively.
Information based on the most recent
experience is available through the Motion
Picture Film Division of the Eastman
Kodak Company.

W. T. Hanson, Jr.: Color Negative and Positive

235

R Red Filter

G Green Filter

B Blue Filter

M Dichroic Mirror

A Printing Aperture

Fig. 15. Schematic illustra-
tion of three colored light
sources for "color timing"
with Bell & Howell Model
D printer.

236

Fig. 16. Sound-track processing: A, air squeegee; B, applicator roller;
C, dial indicator for adjusting applicator; and D, film enters wash tank.

March 1952 Journal of the SMPTE Vol. 58

4.0

3.0

* 2.0

c

0)

Q

1.0

Fine Groin Cine Positive
Eastman Color Print Film

Log E

Fig. 17. H & D curves of sound-track images.
Density, ERPI Densitometer with infrared-sensitive cell;
Exposure, Eastman Color Print Film, tungsten light plus color-
correction filters for neutral dye image;
Eastman Fine-Grain Cine Positive Film, tungsten light.

track area, this silver bromide is again
reduced to metallic silver by an edge-
application treatment. The sound track
is thus composed of a combined dye and
silver image. The edge-application
equipment is shown in Fig. 16.

The H & D characteristics of the
sound-track image are a function of
the color of the light used in exposing

the sound track. The highest contrast
is obtained if this light is of the color
quality which gives a neutral dye image.
With such exposure, the density scales
of the silver images in the three emulsion
layers coincide. Under these conditions,
the H & D curve as measured with the
infrared-sensitive photocell is as shown
in Fig. 17.

W. T. Hanson, Jr.: Color Negative and Positive

237

An analysis of the characteristics of
the sound track on this film has been
reported by Evans and Finkle.10

References

1. G. E. K. Mees, "Direct processes for
making photographic prints in color,"
J. Franklin Inst., 233: 41-50, Jan. 1942.

2. T. H. Miller, "Masking: a technique
for improving the quality of color
reproductions," Jour. SMPE, 52: 133-
155, Feb. 1949.

3. W. T. Hanson, Jr., and P. W. Vittum,
"Colored dye-forming couplers in sub-
tractive color photography," PSA
Journal, 13: 94-96, Feb. 1947.

4. W. T. Hanson, Jr., "Color correction
with colored couplers," /. Opt. Soc.
Am., 40: 166-171, Mar. 1950.

5. F. C. Williams, "Objectives and
methods of density measurement in

sensitometry of color films," J. Opt
Soc. Am., 40: 104-112, Feb. 1950.

6. A Report of the Color Sensitometry
Subcommittee, "Principles of color
sensitometry," Jour. SMPTE, 54: 653-
724, June 1950.

7. C. J. Kunz, H. E. Goldberg and G. E.
Ives, "Improvement in illumination
efficiency of motion picture printers,"
Jour. SMPE, 42: 294-314, May 1944.

8. W. Bornemann and W. McKusick,
"Illuminating system and light control
for 16mm continuous optical printer,"
Jour. SMPTE, 54: 480-482, Apr. 1950.

9. J. G. Gapstaff, "An experimental
35mm multilayer stripping negative
film," Jour. SMPTE, 54: 445-453,
Apr. 1950.

10. G. H. Evans and J. F. Finkle, "Sound
track on Eastman Color Print Film,"
Jour. SMPTE, 57: 131-139, Aug. 1951.

238

March 1952 Journal of the SMPTE Vol. 58

Printer Control in Color Printing

By C. A. HORTON

The use of an electronic photometer is described for maintaining color and
intensity balance in 35mm color printers. Some precautions are given on
the use of color-correcting niters and data are provided on the hue shift with
temperature and consequent reduction of transmission of certain glass filters.

JL RINTING OF motion picture color film
brings with it problems which do not
arise in black-and-white printing. There,
it is usually enough to specify an ex-
posure adequate to produce a definite,
easily measured, minimum density in the
print. The choice of print stock and
processing conditions determine the con-
trast and tone reproduction from a given
negative. In color printing, the quality
of the printing light must be controlled
even more closely than its intensity.
The problem in printing a color film is to
adjust and maintain an illumination in
the printer gate which will produce a
pleasing picture on the theater screen.

These strict requirements are espe-
cially true in negative-positive systems
where the process gamma of the print
stock may approach 3.0. The tolerances
which have been found necessary are
0.05 log 7 on illuminance and 0.02 log I
on color changes. Such variations as
these, or larger ones, may easily arise
from aging of the printer lamp, breaking

Communication No. 1351 from the Kodak
Research Laboratories, a paper presented
on April 28, 1950, at the Society's Con-
vention at Chicago, 111., by C. A. Horton,
Eastman Kodak Co., Kodak Park Works,
Rochester 4, N.Y.

and, hence, replacement of the heat-
absorbing glass, or instability of the
absorbers in the color filters being used.
An accurate and reproducible method of
measuring the quality and intensity of
the light in the printer gate is almost in-
dispensable. A convenient photometer
for printer control should have a linear
scale, a stable zero, a wide sensitivity
range and freedom from fatigue, and
should be easily fitted into the printer
gate.

The use of a photronic cell and a
galvanometer for control of 1 6mm Koda-
chrome printers has been described to
this Society by Aex in 1 947. * However,
the cell fatigue and the short scale of this
instrument make it unsatisfactory for
controlling printers using high-gamma
materials.

The Densichronf shown in Fig. 1 has
been found to satisfy most of the above
requirements. It consists of a photocell
in an a-c magnetic housing, connected to
an a-c amplifier with a logarithmic re-

*Paul S. Aex, "A photoelectric method

for determining color balance of 16-mm

Kodachrome duplicating printers," Jour.

SMPE, 49: 425-430, Nov. 1947.

t Manufactured by the Welch Scientific

Company.

March 1952 Journal of the SMPTE Vol.58

239

Fig. 1. Densichron and photocell unit.

sponse meter. It is available with either
red- or blue-sensitive photocells. The
blue cell has been found sufficiently
sensitive to measure printer lights
through red, green or blue filters. The
logarithmic scale is convenient since it
may be interpreted directly as log /. Its
scale has been found to be linear to better
than 0.10 over a log I range of 2.8, as
shown in Fig. 2. In all parts of the
scale, fatigue is less than 0.01 log /over a
three-hour period.

In order to use the instrument as a
photometer, some constant source of
light is needed as a reference zero. For
printer control, the absolute value of the
light need not be known. A small bat-
tery-operated lamp connected through a
milliammeter is adequate or, if a stand-
ard sensitometer is available, its direct
beam may be used. It is advisable to
take readings of the zero of the instru-
ment through tricolor filters in order to
detect any relative change in photocell
sensitivity. These tricolor filters will be
referred to as "analyzing filters." By

having such readings recorded it is al-
ways possible to adjust the gain to correct
for slow drifts in overall sensitivity or an
accidental movement of the gain control.
The present instrument has been in use
six months without showing any change
in relative sensitivity.

The choice of tricolor filters for
analyzing the light should be determined
from the print-film sensitivity and the
photocell sensitivity. Ideally, the prod-
uct of the values of the photocell sensi-
tivity and the transmission of the filters
should match the film sensitivity. Since
neither of these sensitivities is usually
known accurately, it is fortunate that
this requirement does not have to be ful-
filled strictly.

When printing is being done through
filters which produce narrow spectral
bands, the analyzing filters may be any
set, provided they isolate the same red,
green and blue regions of the spectrum as
are used in the printer. When printing
with white light, or white light modified
by color-compensating filters, the choice

240

March 1952 Journal of the SMPTE Vol. 58

is more limited and care must be taken to
choose filters which give, with the photo-
cell, a response close to the peak of the
film sensitivities. No specific rules for
the selection of the filters can be given
except that a set is satisfactory if it pre-
dicts filter changes that agree with
photographic tests. Once a set of filters
has been decided upon, it may be used as
long as the print material has the same
sensitivity distribution.

The most satisfactory filter set found
for control of printers using Eastman
Color Print Safety Film, Type 5381,
was: red — Kodak Wratten Filter No.
70; green — Kodak Wratten Filter No.
16 + No. 61; blue — Kodak Wratten
Filter No. 35 + No. 38A. With this
set, predictions or transfer of color and
intensity balance from one printer to
another are correct to 0.02 log E when
printing through color-compensating fil-
ters. When transferring a balance from
this condition to a printer equipped for
printing with narrow spectral bands,
prediction is correct to about 0.05 log E.
The set of narrow-band printing filters
used was: red — Kodak Wratten Filter
No. 29; green — Kodak Wratten Filter
No. 16 + No. 61 ; blue — Kodak Wrat-
ten Filter No. 35 + No. 38A + No.2 A.

In addition to the tricolor filters, the
present instrument has a small disk of
flashed opal and one of Corning 9780
infrared-absorbing glass 1^ mm thick
over the photocell. These are necessary
when using the instrument as a color
densitometer since the dyes have little
density to the infrared. These precau-
tions are probably not necessary when
using the photocell as a photometer, but
no tests have been made to verify this
conclusion.

The Densichron probe containing the
photocell is small enough to fit easily in
the gate of a Bell & Ho well printer. By
rotating the probe slowly through a
small angle, it is quite easy to get a re-
producible maximum intensity reading.
In some printers where the probe does
not fit in the gate, a curved rod of trans-

parent plastic may be used to conduct
the light to the cell. In this case a
mechanical guide should be used to lo-
cate the probe and rod in the gate, since
it is difficult to get reproducible readings
when the assembly is held by hand.

When a printing balance is known on
one printer, or exposing device, it is fre-
quently necessary to set up the same
balance on other printers. By making
the appropriate changes in the tricolor
readings from the first printer to com-
pensate for differences in speed or time of
exposure, these new readings may be set
up on the second printer by adjusting
the filter pack and timing shutter until
the Densichron shows the desired values.
This procedure usually brings the printer
in balance or so close to it that one photo-
graphic test is adequate before starting
the printing of full-length pictures.

When a printer test is off balance or
shows improper exposure, and density
readings on the print, or experience sug-
gests that a change should be made in
the printing light, the photometer is
more reliable than the catalog densities
of the compensating filters. Figure 3
shows spectrophotometric curves of a
well-known set of compensating filters.
Inspection of these curves shows that the
addition of a Kodak Color Compensating
Filter CC-50C to the filter pack would
seriously disturb the blue and green light
balance as well as make the desired cor-
rection to the red intensity. By taking
red, green and blue photometer readings
before and after each filter change, the
balance may be corrected exactly to the
prescribed set of values. The densities
of the complete set of color-compensating
filters as read by the Densichron through
the filters given are listed in Table I. If
measurements are not made by a pho-
tometer in the printer gate, these values
may be used to estimate the amount of
neutral density being introduced in the
light beam by the addition of one or
more of these filters.

For printer control, daily readings are
made through red, green and blue filters

C. A. Horton: Color Printing Control

241

2.8
2.4
2.0
1.6

"

0.8

0.4

0.4 0.8 1.2 1.6 2.0 2.4 2.8

Log I white light

Fig. 2. Linearity of Densi-
chron response.

1.2

1.0

I 0.8

I

0.6

S

<| °4
0.2

0.0

400

CC 50C
CC 50M
CC SOY

500 600

Wove length (mp)

700

Fig. 3. Spectrophotometric
curves of color-compensat-
ing filters.

2.0

1.9

1.8

1.7

468

Minutes

10

Fig. 4. Density change of red
glass filter, Corning No. 2408,
Melt 1151, with rise in tem-
perature.

242

March 1952 Journal of the SMPTE Vol. 58

2.8

-

-

• 39 F

x 208 F

2.4

"

2.0

-

1

1 1.6

-

-

o

u

§ 1.2

-

0.8

_

-

11

0.4

-

\

Fig. 5. Shift in the absorption

U

of a red glass, Corning No.

VS. ,

2408, Melt 1151, with change

00

i

i i i 1 i

, ,

i , i . y *

in temperature.

400

500

600 700

Wave length

(mp)

Table I. Densities of Color-Compensating of the illuminance in the gate, with and

Filters as Measured by

the Densichron without the printer-balancing filters in

Through the Tricolor

Filters Shown. position. A comparison of these six

Density to Kodak
Filter

will show whether slow drifts are due to

Designation

No. 29

No. 61

No. 49 lamp or to filter changes.

Special care must be taken if glass fil-

GG-05G
GG-10G
GG-20G

0.06
0.10
0.16

0.04
0.04
0.05

ters are used in the printer, since many of
Q'QS them, particularly the reds and yellows,

GG-30G

0.28

0.07

0^06 show a large change in density with

CG-40G
CG-50G

0.33
0.43

0.08
0.09

0.06 temperature. This difficulty has been
0 • °6 met with in the use of both Corning No.

GG-05M

0.03

0.06

0.05 2408, Melt 1151, and Corning No. 3384,

GG-10M
GG-20M
GG-30M

0.03
0.04
0.05

0.10
0.16
0 28

0.06 Melt 600, glasses. The decrease in
Q ' JQ transmission of the filter as measured by

GG-40M

0.05

0.31

0 ! 1 2 tne Densichron may amount to 0.25 log /

GG-50M

0.06

0.47

0.13 between room temperature and printer-

GC-05Y

0.03

0.03

0.09 operating temperature. Figure 4 shows

GG-10Y

0.03

0.03

0.12 a typical curve of density to red light

GG-20Y
CC-30Y
GG-40Y

0.03
0.03
0.03

0.03
0.03
0 04

0 • 23 against time for a Bell & Howell printer
0 ' 37 with Corning No. 2408, Melt 1 1 51 , in the

GG-50Y

0.03

0.04

0*44 beam. The equilibrium temperature

C. A. Horton: Color Printing Control

243

This change of density of glass filters is,
in general, due to a movement of the
absorption edge toward longer wave-
lengths with increasing temperature.
Spectral-density curves for the above-
mentioned red glass at two different
temperatures are shown in Fig. 5.
Though inconvenient, this density change
with temperature need not reduce the
accuracy of the printer control if suffi-
cient time is given for the filters to reach
temperature equilibrium before readings
of the intensity are made.

When printing through narrow-band
filters, such as are used in making dupli-
cate positives from a color negative or

printing duplicate positives or negatives
on a color print film, the analyzing filters
should still be used in addition to the
printing filters.

When it is necessary to change to a
print emulsion whose relative red, green
and blue speeds are different from those
of the emulsion previously used, the
speed differences determined from sensi-
tometry may be applied directly to the
tricolor readings and the filter pack ad-
justed as indicated by the photometer.
Similarly, if densitometry of the negative
predicts timing changes to red, green and
blue, these may be set up on the printer
by means of the photometer.

244

March 1952 Journal of the SMPTE Vol. 58

Desirable Characteristics of 16mm
Entertainment Film for Naval Use

By LOWELL O. ORR and PHILIP M. GOWETT

Current 16mm release prints are evaluated for sound quality, chiefly by
measuring dynamic range. Projection equipment and conditions are
described.

w,

E HAVE GATHERED DATA on the

quality of 16mm entertainment film
release prints as we have found it at
the Navy Motion Picture Exchange,
Brooklyn. By way of comment, and in
order to narrow the issues to be pre-
sented, we should mention that the
16mm films discussed here are those
resulting from reduction of 35mm
entertainment films which are circulated
to all commercial motion picture
theaters, and further, that such 16mm
prints are not used by the Navy alone,
for equal numbers are used by the
Army and Air Force Motion Picture
Service, and approximately half as many
by the Veterans Administration. In
addition, they are being used on com-
mercial ocean liners, and in various
foreign countries. We have every
reason to believe that in such various
uses, exhibition conditions are such that,

Presented on October 17, 1951, at the
Society's Convention at Hollywood, Calif.,
by Lt. Lowell O. Orr, USN, New York
Naval Shipyard, c/o Motion Picture
Exchange, Brooklyn 1, N.Y., and Philip
M. Cowett, Dept. of the Navy, Bureau of
Ships, Washington 25, D.C.

while not exactly the same as those
prevalent in the Navy, they are never-
theless similar to, and in many cases
closely approximate, those of the Navy.
We would further like to point out that
in referring to the producer herein, we
mean the actual producer, or releasing
distributor, who is the prime contractor
under Navy Motion Picture contracts.
As such, the prime contractor is solely
responsible for the quality of the release
prints supplied.

In undertaking our study we felt
that good results could be obtained from
presenting a systematic analysis of
prints as presently released for Navy use.
It is not our intention to be hyper-
critical, but rather, through the relation
of our observations, to tend to indicate
what the current practice is with respect
to the sound quality of 16mm entertain-
ment motion picture release prints.

It is a truism that, before any suitable
improvement in quality can ensue, a
true picture must be had of the situation
existing at this time. That, then, is our
motivation: to establish a plateau, you
might say, representing the current
practice in 16mm film production.

March 1952 Journal of the SMPTE Vol. 58

245

Fig. 1. Standard Navy 16mm IC/QEB projector set.

+ 5
0
-5
-10
-15

-20
2

^]

>ctive

^Uppe,

Limit

^

^

*

. —

• —

I

_ow

»r I

irr

lit

X

ir

1

'^s

^N

SN

\'

^S

.\

\

s

\

^

N

^

3 50 100 200 500 1,000 2,000 5,000 10,000 20,0

246

Frequency in Cycles per Second
Fig. 2. 16mm sound reproducer electrical characteristic.

March 1952 Journal of the SMPTE Vol. 58

Short Form Specification

16mm Review Rooms and Reproducing Equipment

Amplifier Power Output:

10 w, 1% distortion
15 w, 2% distortion
Over passband, 50 to 7000 cycles

Signal-to-Noise Level :

50 db below 10-w level signal from 400-cycle
SMPT& standard level film

Frequency Response:

As given in SMPTE recommendations
Uniformity of Scanning-Beam Illumination:

H db using SMPTE test film
Flutter:

0.25% peak using SMPTE test film and RCA field-
type flutter meter

Fig. 3. SMPTE specifications for reproducing equipment.

We, in the Navy, have problems that
are somewhat different from those of
the normal user of 35mm entertainment
film in that the situation in which the
film is reproduced is generally far from
desirable. We, for example, exhibit
film topside, where high ambient noises
caused by exhaust blowers, noises of the
ship underway and cross winds, all tend
to force a limited dynamic range for
optimum intelligibility. A further situa-
tion of reproduction which is quite
common is that encountered aboard
an aircraft carrier, where high ambient
noise (80-90 db) results in the net end
of masking low-level sequences com-
pletely and, unless the print is one of
high intelligibility, preventing under-
standing of much significant dialogue.

As a basis for understanding the data
on quality, we should first describe the
type of equipment used in the screening
room and the exact installation with
regard to how the various measurements
were made and how our data were ob-

tained. This equipment is the standard
Navy 16mm IC/QEB projector set
(Fig. 1), manufactured by the De Vry
Corporation and designed to conform to
the requirements, regarding its fre-
quency-response characteristic (Fig. 2),
as set forth by the 16mm Subcommittee
on Sound Reproduction (Fig. 3). It
has a power-output capacity of 20 w at
less than 2% distortion over the pass-
band of frequencies (Fig. 4). The
frequency-response and tone-control
characteristics of the amplifier are
shown in Figs. 2, 5 and 6. Similarly,
the portable loudspeaker frequency-
response characteristics are shown in
Fig. 7.

In order to evaluate prints as to
dynamic range, we utilized a second
amplifier which was bridged directly
across the sound output from the two
projector sound heads. The function
of this second amplifier was to operate
the volume-level indicator which is
incorporated in a still projector and is

Orr and Cowett: 16mm Release Quality

247

0.002 0.005 0.01 0.02

0.05 0.1 0.2 0.5 1

Power Output in Watts

10 20 30

Fig. 4. Amplifier total harmonic distortion versus power output
(frequency range 100 cycle /sec to 5000 cycle /sec).

+ 10

248

50

100

200 500 1,000 2,000

Frequency in Cycles per Second

5,000 10,000 20,000

Fig. 5. Amplifier tone control characteristics.

-t-IU

-1-5
0

r5

-10
-15

-20
2

ectiv

e N

-Uppe

1

r Limit

\

ji

_

/

^

•-

—

^

v,

Blower L

mit

D 50 100 200 500 1,000 2,000 5,000 10,000 20,0

Frequency in Cycles per Second
Fig. 6. Amplifier frequency response design limits.

March 1952 Journal of the SMPTE Vol. 58

•nu

+5
0
-5
-10
-15

-20
2

/up

per

.im

t

Objective

^

X"

/

/

Lo

wer

_im

t-

>

S.

\

/

\

0 50

100 200 500 1,000 2,000 5,000 10,000 20,0

Frequency in Cycles per Second
Fig. 7. Loudspeaker acoustical response frequency design limits.

identical with what is used in the review
rooms in Hollywood. The VU (volume
unit) meter is calibrated as to the zero-
level reading by reproducing the SMPTE
400-cycle standard-level film. The dy-
namic-range amplifier is adjusted to
indicate full scale on the volume-level
indicator when using this level film.
This was necessary to enable us to
measure low-level dialogue sequences on
the film, which would give no meter
indication whatsoever if a zero-level
meter setting were used.

The Motion Picture Exchange review
room as shown in Fig. 8 is 60 ft long,
30 ft wide and 20 ft in height. It is
made of cinder block which supplies
some acoustic deadening. Therefore,
our screening room is not unlike most
of those in Hollywood, with the single
exception that it is somewhat more
reverberant. The 16mm projection
equipment used in the screening room is
identical in performance with that of
standard review-room 35mm projection
equipment.

Use of Sulfide Photoresistive Cell

One more factor of interest, to us at
least, and a very important factor, was
that we had pioneered, in the IC/QEB
equipment, with the use of the sulfide
photoresistive cell. Our reasons for
using the cell were to secure a wider
frequency range, higher signal-to-noise
level, elimination of photocell hiss,

elimination of photocell microphonics
and the elimination, in the sound head
itself, of high impedances which are
always a source of trouble in high
humidities.

In order to corroborate our thinking
with regard to the use of this cell, and
to check its overall performance, two
projectors were used in the evaluation of
film. One projector contained the
conventional cesium photoelectric cell,
while the second projector was equipped
with a sulfide photoresistive cell. Both
of these projectors were then adjusted
in output level, using the Society's
400-cycle standard level film, so that
they gave the same reading on the VU
meter. It then became standard prac-
tice to screen all film on the two pro-
jectors in order to determine any differ-
ence in the reproduction of film from
either one of the cells. We have
amassed considerable data on the various
film producers' products in this manner
and can state that, with the exception of
dye tracks, there is no difference in the
performance between the sulfide cell
and the cesium cell. The blue dye track
develops a signal of 5 db to 10 db less
than the silver track. No other signi-
ficant changes in print sound quality
exist. The cell comparison data could
have been presented, but have not been
since there was no difference in per-
formance other than as noted.

Orr and Cowett: 16mm Release'Quality

249

Fig. 8. Motion Picture Exchange review room.

Now, getting back to our problems
with regard to high ambient noise, and
the masking effect of such noise on low-
level sequences, we desired to set up a
test condition at the Motion Picture
Exchange that would simulate quite
closely conditions encountered in the
field. Therefore, since we have record-
ings at the Material Laboratory in the
New York Naval Shipyard of all types
of ships' noises, used for determining the
best frequency characteristics of battle-
announce equipment, etc., it was con-
venient for us to procure records of these
ships' noises and, through a reproducer
system set up in the screening room,
duplicate conditions aboard ship.

We also have very accurate data on
the intensity of the noise at various parts
of a ship, so it was not necessary for us to
leave the screening room in order to

determine the best dynamic range of a
print. We emphasize the above because
one of the most significant faults with
prints, as released to us, has been the
tendency to use an extremely wide
dynamic range on the assumption that
the film is going to be listened to in a
theater, under optimum listening condi-
tions, where such a range is practical.
Now that we have established the test
condition, we would like to point out
that this study was made possible
through cooperation and collaboration
with the Society of Motion Picture and
Television Engineers and more especially
with its Subcommittee on 16mm Sound
Reproduction.

The factor of print quality, insofar
as distortion is concerned, or frequency
range, having to do with the naturalness
of the sound, and the overall quality

250

March 1952 Journal of the SMPTE Vol. 58

of the sound have all been carefully
considered in our analysis of prints. The
quality of sound, of course, is not subject
to any method of measurement and, in
a sense, is a matter only of the listeners'
acceptance of what he considers good
quality. Our discussion here, therefore,
deals not with this phase of print
quality, soundwise, but purely with the
factor, more important to us for the
moment, of the dynamic range in release
prints. In securing these data we had
the opportunity, as you may appreciate,
of working with specimens from every
major producer. The producers are
not identified by name, but are marked
in such a way that we can identify their
respective products. In Table I are
the data for both black-and-white and
color, including both variable-area and
variable-density tracks. The maximum
peak, derived from the volume indicator,
is given in terms of decibels below the
selected zero level as indicated below.
The average peak is also in terms of
decibels. The minimum peaks, or the
low-level parts, are not indicated here,
as they are in many cases too low in
level to show on the instrument.

From these data the reader will appre-
ciate that low-level dialogue sequences
would be completely masked by any
distracting noise or poor acoustical
conditions. It is our hope to have film
in the future in which this will not be the
case. We have, then, an analysis of
thirteen producers' products with regard
specifically to dynamic range. As pre-
sented in Table I, these products
indicate what the current practice is
with respect to the dynamic range of
16mm entertainment film release prints.

To establish a basis for comparison
with the data in Table I, the measure-
ments on the current SMPTE 16mm
Sound Service Test Film Type SPSA
are given in Table II. Table III is a
summary average of the data contained
in Table I. By way of further analysis
there are shown in Table IV the most
and least favorable readings taken in

this survey of both types of sound tracks
in black-and-white and color. It should
be noted that the producer of a "best"
sound track is also capable of producing
a "worst" sound track.

From the preceding it can be seen
that to reproduce satisfactorily all prints
offered to the Navy, the equipment
would have to be designed with a reserve
gain of at least 25 db. That, then,
means that particular color prints are
20 db below standard black-and-white
prints as released for system check by
the Society of Motion Picture and Tele-
vision Engineers.

The preceding data on the dynamic
range of current release prints are the
basis for experiments conducted at the
Motion Picture Exchange in determining
an acceptable degree of compression of
the dynamic range.

The synthetic ship's noise generator
was energized in the review room to
establish an 80-db acoustic noise level
approximately 30 ft from the program
speakers. The acoustic noise level of
80 db is not an uncommon noise level
aboard Navy vessels.

Feature films were then reproduced,
with the distracting noises previously
described, and their intelligibility de-
termined.

We found, by actual experience, that
the degree of compression which did
not completely destroy the sense of
realism was limited. However, we have
come to the conclusion that a real
improvement in overall sound intelligi-
bility, without destroying the artistic
value of the film, or increasing the print
sound distortion, could be realized by
raising the low-level dialogue sequences
6 db. Peak levels as well could be
raised 3 db without difficulty.

This, then, is an attainable improve-
ment in the dynamic range of release
prints.

We are presently using the industry
averages shown in Table III in evaluat-
ing the acceptability of prints for Navy
use. (It is our intent to eliminate from

Orr and Cowett: 16mm Release Quality

251

=§

»<

1

"° S

si

7I7IMIIIIIII

I i CM •

CN 00 ^-i 00 00 I t^

Mill I

Tf ON 00 10

CO"T-.\OCM[ [oo|cMoolo^-i

7777 7 77 77

OOVO •

O . . 10 f-.

I T-H I \0 00 I 1-H

I I II I II II

^^ VO t""*- VO C^ SO ^" LO I I ^" \O \O

I I I I I I M I I I

ir)r^-ooooc<i-^-\oc\ i

i— ILOCN^— ir^-ooo
ts co eg ^-H (N ,-. co

csorjocNTt-

1-1 CM CM CS T-l

Table I

Key: Producers identified as A - M; num-
ber of prints given in parentheses; B &W
V-A, black-and-white variable area; B &W
V-D, black-and-white variable-density; G
V-A, color variable-area; G V-D, color
variable density.

Navy distribution the worst examples
given on the "extreme quality" figures
[Table IV] since there is obviously no
possibility of manufacturing a projector
to reproduce these low-level prints
satisfactorily, and if there was, the
signal-to-noise level of such a print
would render it valueless.)

It is significant, as pointed out at the
beginning, that there are many users of
16mm copies of 35mm entertainment
films. It is notable that in Glass "A"
releases in 35mm some 350 prints will
be made for commercial exhibition.
Every care and consideration is given to
the making of such 35mm prints since
the audience to which they are to be
shown represents the major source of
revenue to the industry. In order to
insure and protect this source of revenue,
the prints must in all respects be heralds
of the art of the motion picture. Nothing
is permitted which would detract from
that art.

There is no disputing the soundness
of the precautions taken to prevent
release of inferior 35mm prints. In
fact, the engineer has been successful
not only in improving and enhancing
the input into the sound and motion
picture camera, but he has also translated
the sound and image, through proc-
essing techniques, into 35mm release
prints which are truly symbols of the
art of the motion picture.

Why has this same effort not been
made in the direction of 16mm release
prints? Certainly the same factor of
inviting audience appreciation is pres-
ent. The Navy and the other Armed
Forces do represent a large segment of
the motion picture viewing population.
From that standpoint alone, we are

252

March 1952 Journal of the SMPTE VoL 58

Table II. SMPTE Test Film, Black-and-White Print.

Max. level

Avg.

Dialogue test (sound excellent)
Piano test (sound excellent) .
Orchestra (sound excellent)
Opening music

-6 db
-6

-2
-2

- 8

- 8

- 8

Table III. Overall Average of Data in Table I, for Total of 240 Prints and Releases
Over Period From 3-15-51 to 9-14-51.

Type of track

No. of
prints

Max. peak

Avg. peak

Black-and-white, variable-area ....
Black-and-white, variable-density . . .
Color, variable-area .....'...
Color, variable-density

... 119

... 55
... 59

... 7

-6db
-8
-9

-7

-12 db
-14
-16
-14

Table IV. Volume-Level Extremes.

Best Film

\

Worst Film

Pro-

Peak, db

Pro-

Peak, db

Type of track

ducer

Max. Avg.

ducer

Max. Avg.

Black-and-white, variable-area . .
Black-and-white, variable-density .
Color, variable-area
Color variable-density

. E
. A
. G, I
C

-0 - 4
-3 - 6
-2 -10
-4 -10

D
A
G
A

-14 -20
-14 -26
-16 -26
-20 -30

sure that you will agree that the de-
precation of the "Art" in the form of
poor sound and picture 16mm release
prints is undesirable.

This, then, is not only our problem,
but more largely it is the problem of
the motion picture producer and engi-
neer in turning out a 16mm release
print that can, as far as practicable,
herald the art of the motion picture to
the extent that the current 35mm print
is symbolic of that art.

Discussion

John Hilliard: I'd like to ask Lloyd
Goldsmith, Joe Aiken, Samuel Cohen,
Norwood Simmons, Fred Albin, Art
Blaney and Eddie Reichard, if he's here,
to come up and be available for questioning
in this period.

We will proceed with a direct question-

and-answer period so that the Navy can
have available information which they
specifically came out here for.

The first question that they are interested
in, perhaps, will be a brief review of optical
reduction from 35 to 16, and if Sam is not
here, Eddie would you come up and
briefly review for us the process that you
use in connection with optical reduction
of 35mm prints to 16? Tell Mr. Cowett
and Mr. Orr the process involved in both
black-and-white and color. I see Mr.
Cohen is here. Sam, we've asked that
you review for us briefly the optical re-
duction technique that you use in your
laboratory to produce the 16mm prints
similar to those used by the Navy. In
other words, explain what the technique
is in connection with both variable-area
and variable-density.

Sam Cohen: Is this with regard to sound
track and picture?

Orr and Cowett: 16mm Release Quality

253

Mr. Milliard: Both, but principally in
connection with sound track because
they are having difficulties in evaluating
problems in connection with variable-area
and variable-density — something that Mr.
Cowett can elaborate on as we go along.

Mr. Cohen: I may be taking a bad step
here: but in every case we recommend
re-recording the sound negative and not
making dupe negatives from the 35 ma-
terial. Dupe negatives made from the 35
material do not give nearly the quality
obtained when a 32-35 track is recorded
with 16mm characteristics.

Mr. Milliard: They are unfamiliar with
the technique of optical reduction from
35 to 16 and could you inform them?

Mr. Cohen: In optical reduction there
are various machines that reduce the
35mm to a dupe negative 0.8 in width
and 0.4 in length. To take area as one
specific case, we take the 35mm fine-grain
which is made to a density of 1.90 to get a
slightly filled 35mm fine-grain track. This
is reduced to a sound-recording stock and
is exposed to reach proper density by
developing in a high-contrast developer.
We obtain a negative density slightly lower
than an original recorded negative. That
is, an original recorded RCA negative
would be a density of 2.75 and a dupe
negative would be 2.5 on sound-recording
stock developed in high-contrast developer,
optically reduced 0.4 in length, 0.8 in
width, and this gives a variable-area 1 6mm
negative from which your reprints can
be made. On variable-density, this can
be done either on a positive or negative
stock. We have achieved better results
by duping to a panchromatic stock and
developing it in a negative developer of
low contrast. Panchromatic stocks de-
veloped to a gamma of 0.55 will closely
reproduce the original bias and unbias
densities

Mr. Milliard: In your judgment, is
there any difference between the develop-
ment for variable-area and variable-
density that would reflect a change in
quality for the 16mm work aboard ship.

Mr. Cohen: If they are not developed
in the proper developers there will be a
definite loss. In variable-area, for in-
stance, many times laboratories try to
make composite dupe negatives from

variable-area material and naturally
they're miles apart because the picture
dupe should be developed to a gamma of
0.55 on a low-gradation stock, where the
sound should be developed to 3.00 plus
gamma on a high-contrast stock. So if
you try to make a composite or a single
variable-area dupe and develop it in
negative-action developer, the results would
be atrocious and the same goes with
variable-density. If you try to put that
on a high-contrast stock and develop it
in a high-contrast developer, you would
have bias - unbias going from 0.30 to
1.30 or thereabouts. You couldn't com-
pensate in printing.

Mr. Milliard: There has been indicated
some difficulty with variable-density as
compared to variable-area and I would
like to have Mr. Cowett make a few re-
marks along that line to see if we can help
him in that connection.

Philip Cowett: Every report that we
have received from the field, from many
ships out in various parts of the world,
indicates that variable-density is the one
source of headache and all ask, "Can
all of our prints have variable-area sound
tracks?"

Mr. Cohen: Are those prints you're talking
about or dupe negatives? I don't under-
stand. Do you want the release prints
to be on variable-area track?

Mr. Cowett: Yes.

Mr. Cohen: I see. Well, there is a
definite reason for that. We run into that
a good deal with television stations. A
picture recorded for 35mm, either in
density or area, does not have the charac-
teristics to give volume necessary in the
16 projector; the density track seems to
suffer more; and improper laboratory
control can hurt density track much faster
than it can hurt area. If it is printed too
heavy, volume is lost immediately; and
on area, quality is lost more than volume —
volume to some extent, but quality is lost
much faster than volume.

Mr. Cowett: Is it then possible for the
motion picture industry to produce all
16mm prints with variable-area tracks?

Mr. Cohen: No, there are some other
factors that enter into it. It is really not
necessary if the original material is made
properly. We are doing 50% area and

254

March 1952 Journal of the SMPTE Vol. 58

50% density sound at the present time,
and from a volume and quality standpoint
each is equally good. The Navy should
specify a sound track made for 16mm.
In many cases I know that just 35mm fre-
quency-characteristic tracks are being
used for production.

Mr. Hilliard: That is the reason Mr.
Cowett is here. He would like to obtain
sufficient information so that he can
stipulate what should be the optimum for
both types of recording in connection with
the 16mm prints that the Navy uses.

Mr. Cohen: Every Navy contract I've
come in contact with stipulates certain
fine grains and a certain number of
prints. None of them that I've heard of
has stipulated a sound track recorded for
16mm, which comprises the majority of
the prints.

Mr. Cowett: I imagine that, since I am
not a film man, Mr. Marks may later
have something to say on that, but it would
seem to me that since we were getting a
16mm print, the sound track should also
be for a 16mm.

Mr. Cohen: But many of the people
producing the pictures are in studios
where there is no 16mm equipment, where
16mm is done outside and the contract
doesn't often call for it, and they show the
35mm print; when they get their approval
they ship the 35 to get the approval, and
it's run in the 35 projection room, and it
stands to reason that if they want that to
be the best, they can obtain it, in order to
sell more pictures, which is reasonable.

Mr. Cowett: We used to view the 35mm
prints for acceptance and then wouldn't
bother too much looking at the 16. But
now we run off each print and each print
is rejected if it's not good.

Mr. Hilliard: I'd like to ask Joe Aiken
what experience he's had in connection
with their work at the Navy laboratory,
in connection with variable-area and
variable-density. What do you think
are the difficulties involved here with Mr.
Gowett?

Joe Aiken: We have had very little
comparative experience between 16mm
variable-area and variable-density prints
at the Naval Photographic Center, as our
prints have all been variable-area for the
past several years. However, since this

question of dynamic range and overall
level has been presented, I will describe
our recording practices to see if there is a
parallel with the problem which Mr.
Cowett has brought up.

The Naval Photographic Center pro-
duces Navy training films primarily.
Dr. Carpenter has discussed certain phases
of them at this Convention. We produce
a part of the training-film program; the
majority are produced under contract by
commercial studios and under Navy
technical supervision. In many of them,
sound effects and music are employed in
much the same manner as in entertain-
ment films. Usually training films are
produced first in 35mm. Following their
acceptance, those produced at the Photo
Center are re-recorded to 16mm for
release printing, with a frequency charac-
teristic slightly restricted at both ends of
the spectrum, and keeping the average
recorded level as high as practical. When
the ratios of levels of voice, music and
effects are established in mixing for the
35mm sound track, we keep 16mm re-
production in mind. We therefore hold
the dynamic range within closer limits,
with less spread between highest and
lowest levels, than is customary in 35mm
entertainment films. Although extremely
low voice passages are not permitted, the
films are quite acceptable in their 35mm
versions.

Mr. Cowett has stated elsewhere that
the difficulties described in his paper
have not been experienced in projecting
1 6mm prints of Navy training films. It is
my personal opinion that those studios
that wish to produce both entertainment
films for 35mm theater release, and 16-
mm prints for projection under less de-
sirable conditions, could afford to con-
sider the use of two techniques when voice,
music and sound effects are mixed. One
mixing technique is suitable for the 35mm
theaters, and the other, for 16mm projec-
tion, which confines the dynamic range
within closer limits, and where important
voice passages are maintained at adequate
levels.

Mr. Cohen: On that, here, we do all the
re-recording right from the first trial print
of the 35mm in order to avoid the expense
of a complete re-recording job which
would be quite an expensive operation

Orr and Cowett: 16mm Release Quality

255

and the results are, the sound men think,
quite satisfactory, without going to the
original of a new dub job. It's consider-
ably less.

Mr. Milliard: Sam, would you briefly
indicate your experience in connection
with this color problem, that is the Navy's
problem of large variations in volume with
the difference in type of dye tracks.

Mr. Cohen: Every color process that's
going through today requires a different
track treatment. Certain dye tracks are
being sulfided or subjected to various
other treatments to enable them to be used
with their present exciter lamps. Present
systems haven't run into any difficulty
except where the track hasn't had enough
volume in the original recording. It does
require a greater amount of volume than
on a black-and-white track; and if the
track has just been recorded for black-and-
white, then is being used in a color process,
it will not be completely satisfactory; and
if different processes are used — such as
for Technicolor — and the same product
used on another process, the results would
not be of equal quality.

Mr. Hilliard: Mr. Gowett indicated
informally in a sound session yesterday
that the green-dyed sound track on the
lead sulfide cell gave little or no output.

Mr. Cohen: I don't think that the color,
whether it's green or brown or black, has
as much to do with it as the opaqueness
or the transmission of the track itself.

Mr. Hilliard: Well, that is really the
factor that we're involved in — the lead
sulfide cell and the green color. You
know the lead sulfide cell is most sensitive
in the infrared region.

Mr. Cohen: In going through track
problems, we use various types of track
application and various processes in our
laboratory. When viewing the tracks
and listening to them you can have a
different color, as some tracks go from the
yellow to a dark brown in color and still
retain the same amount of volume, if all
of the image is there and if there is proper
transmission density. But as for the green,
we have used a track that does have a
green color and in theatrical production
the volume has been sufficient, but that
volume was recorded there originally.
By the way, the majority of the time we

use a variable-area track, in which you
can naturally get quite a volume if you
go to 100% modulation, although I
believe (in present theater practice) it runs
around 40% modulation. You can
increase that volume tremendously.

Mr. Hilliard: I'd like to direct a question
to Dr. Frayne in connection with photo-
cells that might be available to help reduce
the variation with color. Is there any
comment that you can make at this time?

John Frayne: Before I answer your
question, I think I should clear up further
some of what Sammy Cohen has been
covering: for example, you lose sound
level on a track if the photocell sees it as
practically a transparency. It's not
necessarily a matter of opacity; the track
may be too transparent in some cases.
For example, some of the dye tracks which
may have a visual density of around 2.0
in the case of variable-area, to a photocell
with a cesium surface they may appear
to be as low as 0.2. In that case, of
course, you lose all the contrast in the
track and therefore lose volume. It is
well known that the Bell Telephone
Laboratories are actively engaged in the
phototransistor development which may
or may not have application to sound
systems. There are none yet available
commercially, but they would seem to be
a very natural device for the sound picture
business.

Mr. Hilliard: You had a question in
connection with contracts?

Mr. Marks: Mr. Cohen, you mention
that you are willing to supply something
so long as the Navy would tell you what to
supply, that since the Navy contracts
didn't mention anything about 16mm
print specifications, you therefore had
free choice in supplying the type of prints
that you did supply.

Mr. Cohen: No, there's one error. I
work for a laboratory, not for a producing
company, and it's the producing company
that has the contract; it's the producing
company that delivers to the laboratory.
We are not in the sound recording business
so we process whichever films the producer
delivers to us. We don't tell the producer
what he is to produce.

Mr. Marks: I'm not taking issue with
you, but you implied that you had Navy

256

March 1952 Journal of the SMPTE Vol. 58

contracts. You had producers' contracts.

Mr. Cohen: No, I don't have Navy
contracts. I said that we do work for
people who have Navy contracts and they
bring in certain materials and I'm telling
you from practical experience just what
material comes in. I am in a position to
see what comes into the laboratory,
probably in a very good position to tell
what type of material is going out, but
it is the Navy's responsibility directly to
write the contracts.

Mr. Marks: You deal as a subcontractor
with the Navy, you don't have anything
to do with the Navy contract itself.

Mr. Cohen: No, none whatever.

Mr. Marks: So, therefore, you couldn't
say that we do not specify any particular
treatment in the contract.

Mr. Cohen: Well, we have to deliver
all the material to the Navy. If the
contract calls for a fine grain and two
35mm prints and five reduction prints,
as a number of them are coming through,
that material goes. We would not be
making the 16mm prints from a 35mm
original negative if there was a 16mm
negative recorded.

Mr. Marks: Well, I agree in that respect.
But the matter of fact is that the Navy
does not specify the type of 16mm release
prints or sound track for a very tangible
reason, for the same reason that this study
was made — because there was no ex-
perience with it until very recently. Now
we intend to apply the results of this
study.

Mr. Cohen: That is what I'm trying to
bring out here. I'm not finding fault
with the Navy. I'm trying to help the
Navy. I'm telling you what should be
put into the contract in order to get a
result, so that on board ship they would
have prints with volume.

Mr. Marks: We are not prepared to
insert anything into the contracts based on
this study alone. That is why we are
inviting the study of the Society of Motion
Picture and Television Engineers in order
to help us arrive at a specification which
is practicable, both from our standpoint
and from the capacity of the industry to
produce. We don't want anything un-
reasonable from the companies. However,

from a contractual standpoint, every time
an order is placed, regardless of whether
the type of film is specified, there is an
implied liability upon the part of the
producers to supply a product which is
fit for the use intended. We maintain
that at the present time nothing is being
supplied, except in rare cases, which is
ultimately fit for the use intended, sound-
wise, when compared with the Society's
400-cycle test film. Along with that,
every producer has been charged with
the knowledge that we are using this
JAN 16mm projector.

Mr. Cohen: If all this were true, there
would be no problem.

Mr. Marks: Well, I agree and that's
why we're bringing it to the attention of
you people, in order to help us to help
ourselves. And I think, personally, it
should be a matter of pride with the entire
industry to attempt to produce a print
which will have the same artistic impact
that it has on the general public in the form
of a 35mm print.

Loren L. Ryder: In the discussion that
is taking place a quality comparison has
been made between reproduction of area
and density 16mm sound films. During
these discussions no mention has been
made of supersonic direct-positive prints
of the density type. Direct-positive super-
sonic density prints on 16mm film when
reproduced on proper equipment are
comparable to 35mm optical prints. If
a decision is made, it should be made
after a study of what may be done and not
based solely on past practices.

Lloyd Goldsmith: I think Mr. Cowett's
paper as read today is a most timely one.
As I see it, speaking from the sound-
quality standpoint, the basic problem, as
also stated by Mr. Cohen, is to get a re-
recorded release 16mm negative. That
is the first step. In the process of making
that negative from previously recorded
35mm materials, the volume range must
be restricted over that normally employed
in 35mm recording. That is the second
step, and it was very well brought out in
Mr. Cowett's paper. The third step,
if necessary, is to make a correction in
the frequency response over that originally
employed in the 35mm material. All of
this is not new. It has been known

Orr and Cowett: 16mm Release Quality

257

certainly since the experience we had in
the last war in securing from the motion
picture industry improved 16mm prints
for use by the armed forces. Speaking for
Warner Brothers, at least, in 1 943 we were
convinced that a separate release negative
for 16mm was desirable and we have been
making such a negative from that time.

I might mention that in making the
16mm release negative from original
35mm material, in re-recording, we do
not, at Warners, change the frequency
response very much. However, that is
largely due to the fact that the 35mm
material released for showing in theaters
has a frequency range, speaking now for
music, with a low end around 60 cycles,
no intentional cutoff, and a high end
around 6000 cycles, with an intentional
sharp cutoff filter of 6700 cycles. In the
case of dialogue the original material is
re-recorded with the same upper-end
limitation, flat to 6000, but the low end is
cut off as sharply as possible at 100 cycles.
In re-recording this material from 35 to
16, to make a new release negative for
16, we make no change in the frequency
response. This we feel is predicated on
good release printers which give good
contact and will faithfully reproduce the
high frequencies, whatever there are, to
6000 cycles.

The one thing we do, however, in
preparing this release negative, is to re-
strict the volume range. We intentionally
pull up the low-level dialogue passages
from 4 to 6 db. As indicated in Mr.

Gowett's paper, it would appear that the
desirable added amount of compression
would be something in that range. I
might add that our original material is
fairly well compressed in the 35, so we feel
that an additional 4- to 6-db pull-up for
16 release is satisfactory. To help the
re-recording mixers accomplish this, we
force them to re-record with the gain in
the mixer monitor's horn circuit reduced
from 4 to 6 db. Therefore, he has to pull
up the low-level dialogue from 4 to 6 db
for a corresponding intelligibility which
he has been used to. The high-level
passages, particularly the main and end
titles, in music, that are normally 100%
modulated, he still holds to 100% modula-
tion by means of peak-reading volume
indicators, as it is recognized that there
must be no intentional overloads in the
16mm release material, unless it be of
certain sound effects which can stand
such overloads without apparent distortion.

I would like to stress again those three
steps: the special re-recorded release
negative, compression of volume range, and
restriction of frequency characteristic if
necessary. That is the secret of good
16mm release sound quality from 35mm
original material.

Mr. Cowett: I should like to thank the
members of the motion picture industry
who appeared here on this panel and hope
that some day the entire industry may
follow the process outlined by Mr. Gold-
smith which appears to be a workable and
a desirable plan.

258

March 1952 Journal of the SMPTE Vol. 58

High-Speed Motion PictureCameras
From France

By PAUL M. GUNZBOURG

Two high-speed motion picture cameras developed by the French firm
of Merlin-Gerin-Debuit are briefly described. The first operates at a
speed of 3000 frame/sec, using 100-ft rolls of 16mm film; the second,
using standard 35mm film, can be operated up to a speed of 100,000
frame/second. Both cameras employ a rotating lens drum. In the
slower camera ordinary oscillographic film unwinds continuously by con-
ventional means. The second camera uses the device of a film strip
which is attached to the interior of the lens drum rotating with it.

3000 Frame/Sec Camera

The optical principle of the camera,
shown in Fig. 1, is based on the juxta-
position of a stationary objective and a
series of mobile objectives passing rapidly
before it. For the mobile objectives-
compensators, commercial lenses with
=*= 1% tolerances are used.

A primary objective with a focal length
of 140 mm, housed in a focusing mount,
is capable of focusing at distances
ranging from 4 ft to infinity. For dis-
tances from 2.5 in. to 4 ft, other objec-
tives of 50-mm (2-in.), 80-mm (3.5-in.),
100-mm (4-in.) and 120-mm (5-in.)
focal length can be substituted in similar
mounts.

Optical compensation is provided by a
rotating drum with 80 small objectives

Presented on April 6, 1 949, at the Society's
Convention at New York, N.Y., by Paul M.
Gunzbourg, Mac Donald International,
Inc., 115 Broadway, New York 6, N.Y.

set at equal distances around its periph-
ery. These are three-element anas-
tigmats with a focal length of 20 mm at
an aperture of//3.5. Figure 2 shows the
optical drum.

The film is driven by a double row of
sprocket teeth in the lens drum and
passes in the conventional way from the
supply to the take-up spools. The
drum is made in one piece and, in rota-
tion, controls the movement of both
film and lenses. This principle provides
for synchronization of both film and
images without mechanical means. Un-
due tension of the film is avoided by
having 6 to 8 teeth of the lens drum in
mesh, resulting in movement of the film
without tear or breakage. Power is sup-
plied by a universal a-c and d-c 110-v or
220-v motor of high-starting torque, de-
veloping about 20 hp at starting.

A synchronizing device, operated by
an adjustable electrical contact, per-

March 1952 Journal of the SMPTE Vol.58

259

Fig. 1. Interior view of 3000 frame/sec camera.

Fig. 2. Optical drum of 3000
frame/sec camera.

mits the camera circuit to be closed or
opened in accordance with the length of
film desired to be exposed. Since the
current draw is about 0.5 amp, a con-
tactor relay is necessary for the starting
and stopping of the motor. This device
may be adjusted according to the syn-
chronization requirements of the phenom-
enon being photographed or to the
length of film in the camera.

A set of changeable slits of various
widths is provided with the camera.
These slits, fitted between the fixed ob-
jective and the mobile objectives on the
lens drum, permit the decrease of the
exposure time per frame at a given frame
frequency by a ratio of 1:4. A cen-
trifugal brake with which the motor
spindle is equipped permits the adjust-
ment of the speed within the range of
This adjustment can be ap-

260

March 1952 Journal of the SMPTE Vol. 58

Fig. 3. Interior view of 100,000 frame/sec camera.

plied from the moment when 10-12 m
(33-40 ft) of film have been unwound.

Accessories of the camera consist of a
reinforced braced tripod and a focusing
microscope to be attached to the view-
finder. The camera itself weighs about
16^ Ib without motor and accessories,
and can be easily handled.

100,000 Frame/Sec Camera

The same principle of optical compen-
sation is used in the 100,000 frame/sec
as in the 3000 frame /sec camera. The
basic difference in the design of the
two cameras lies in the device adopted
in the faster camera to allow for move-
ment of the film at the far higher rate of
speed required.

Since no film could withstand being

driven at such high speed by any of the
conventional film drives, the film in this
case is applied directly against the inner
surface of a rotating drum, 0.605 m in di-
ameter, and can thus be rotated with it
at the required speed of 250 m/sec (820
ft/sec) without damage. The length of
film that can be used is, of course, deter-
mined by the circumference of the inner
surface of the drum wall, which measures
1.90 m (74.8 in.). One revolution of
the drum results in 750 pictures, repre-
senting a 50-sec projection at normal
speed when the frames are placed in se-
quence. For example, with the drum
rotating at 6000 rpm (film velocity 190
m/sec) the frame frequency resulting is
75,000/sec for a total duration of 0.01
sec.

P. M. Gunzbourg: High-Speed Cameras

261

Fig. 4. Billiard ball falling into glass
of water. Speed: 2500 frame/sec.

The optical drum carries on its periph-
ery 750 fixed compensator objectives,
arranged in three rows, of 250 lenses each,
parallel to the drum axis. Each row is
staggered one-third of the level of the
adjacent row below it. One lens from
each row is uncovered in turn, a slit
being used to limit the angle and the
exposure duration per frame. The ob-

jectives are three-element anastigmats
with 20-mm focal length and //3.5
aperture. The film is held taut on the
inner surface of the drum against the ob-
jectives, by a flexible bronze strip ap-
plied on the reverse side of the film act-
ing as a rear pressure plate. An interior
view of this camera is shown in Fig. 3.

As the drum rotates, the compensator
objectives pass before a fixed objective
having a diameter of 70 mm and a focal
length of 350 mm. The fixed objective
is fitted with an automatic magnetic cap-
ping shutter, synchronized with the mo-
tion of the drum, and operating when the
drum reaches the desired speed of ro-
tation. The shutter closes after one com-
plete revolution of the drum. The
length of film in the camera is thus fully
utilized.

The sequential images produced in
this camera are presented as three circu-
lar images, each 6.5 mm in diameter,
lying obliquely across the film, with
frame 1 produced by a lens in row 1,
frame 2, by the next lens to be uncovered
in row 2, and the third, by the corre-
sponding lens in row 3. For projection
purposes, the images are optically
printed, approximately three times en-
larged, onto standard 35mm film, as
shown in Figs. 4 to 7. Lighting sources
used in the examples shown here were
either flashes of 3^ oz of magnesium set
off behind ground glass, or two projec-
tion-light sources with parabolic reflec-
tors fitted with 2-kw bulbs, also behind
ground glass.

Power is provided by a compound
150-v, d-c motor requiring about 0.5 hp.
Viewing is through a shutter-equipped
window. Other features inside the
housing include contactor and push-
button devices for starting and stopping
the driving motor, and for unwinding and
developing the film automatically. The
camera weighs approximately one ton.

262

March 1952 Journal of the SMPTE Vol. 58

P. M. Gunzbourg: High-Speed Cameras

263

53

T>

1

..

8

S

i

II
ll

1

o 0

V o

-8

264

March 1952 Journal of the SMPTE Vol. 58

P. M. Gunzbourg: High-Speed Cameras

265

Biographical Note

A. C. Dowries

"Arthur G. Dowries, Chairman, Board of
Editors" has been appearing month after
month for 11 years in this Journal: no
more propitious words could appear on the
Journal's masthead.

We do not apologize for thus introducing
a man who has retired from a long and
successful career as chemist, engineer and
research leader. Those who have known
him over some years will agree that in
speaking of him it is in order to treat the
immediate immediately — while not slight-
ing what is more distant.

The quality of the recent Volumes of
this Journal has been the direct result of
a great deal of constant attention by Mr.
Downes, who reviews every paper, assesses
the evaluations made by those he chooses
from the Board of Editors to study each
paper, and prepares the verdicts and often
extensive advice for authors. This service
is not a casual part-time hobby. It is
a job of carefully winnowing each year
at least a couple of solid bushels of papers —
to find some completely acceptable for
publication, some satisfactory in part and
others which must be rejected. The Chair-

man of the Society's Board of Editors is
responsible for assessing them all.

(We should note here that these extensive
activities of the Chairman of the Board of
Editors for today's Journal are made
possible by efficient secretarial service
given by the staff of National Carbon.)

Arthur Caldwell Downes was born in
Ipswich, Mass., in 1882. (This and a few
other milestones are here recorded, chiefly
from American Men o/ Science, 8th ed., 1949) :
B.S. from Massachusetts Institute of
Technology in 1904; chemist at Hartford
Laboratory Co., 1904-05; National Car-
bon Co., Cleveland and Fostoria, Ohio,
1905-17; Assistant Superintendent of
National Carbon's Fostoria Works in
1918-21, Cleveland Works in 1921-22 and
Niagara Works in 1922; directing work
on illuminating carbons, electrical brushes
and new carbon products, he was head of
the Works Laboratories in Cleveland in
1922-25 and the Development and Re-
search Laboratory from 1925 until his
retirement in 1947.

He has been a member of this Society
since 1927 and a Fellow since 1934. He
served as Editorial Vice-President in years
very critical for the Journal and for Con-
ventions and Papers Programs — 1941-46.
During these six years he also served as
Chairman of the Board of Editors.

In 1947 Mr. Downes was made a Fellow
of the Illuminating Engineering Society
which then cited his 40 years as an illumi-
nating engineer, his IES committee work
and his many contributions to IES Trans-
actions, principally with reference to
spectral characteristics of sunshine and its
substitutes. In 1928 he received the
award of the Academy of Motion Picture
Arts and Sciences for his illumination
research. He is also a member of the
American Chemical Society and the
American Institute of Chemical Engineers.

A. C. Downes did a good deal of work
in Hollywood beginning in 1935 when
high-level illumination was being de-
veloped by Mole-Richardson Co. for the
studios to use with the new Technicolor
process. It is, some think, farther than

266

the map shows from Arthur Dowries'
birthplace in Ipswich, Mass., to Holly-
wood. In between, it is true, is Cleveland,
where one of his associates reports, "He
has always been known simply as 'A. G.'
and his long-time and expert knowledge
of carbon formation procedures and* of
the light-emitting properties of illuminat-
ing carbons has been continually respected
and put to good use by his co-workers."

Though long known as A. G., he is
also known as one who uses direct current
methods to get things done — forthrightly,
efficiently and with understanding. And
this is how he got on in Hollywood,
according to a note from Elmer Richard-
son:

"Here in Hollywood we probably go
overboard with our informality. No one
works in the Hollywood technical group

long before he is better known by his
first name than his surname. Right from
the start among ourselves, we always
referred to A. G. Downes as Arthur, but
at first we did not want to offend him so
we all stuck with 'Mr.' One day when we
were all together, some one said: 'Let's
have some fun,' so Pete (Peter Mole) was
nominated to start addressing our good
friend, Mr. Downes, as 'Arthur'; and so
Pete, in his quiet way, with a twinkle in
his eye 'broke the ice.' I think Arthur
was a bit flabbergasted at first but after
he got accustomed to it, I think he was as
happy as we were to drop the 'Mr.'

"Arthur Downes to me is a friend, a
man with that combination of intelli-
gence, experience and knowledge that is
best defined as wisdom, and with human
qualities that make cooperation with him
a delight to all concerned."

71st Semiannual Convention

The Advance Notice listing the sessions
and abbreviated titles of papers went to
members in the Western Hemisphere on
March 10. That was the mailer which
includes the tear-off postal card for con-
veniently arranging hotel accommodations
at The Drake in Chicago during the
Convention. If you have mislaid yours,
please refer to p. 173 of the February
Journal, which has the information you
need.

Here is the schedule of 11 sessions in
which Program Chairman George Colburn
had some 52 papers arranged at press
time:

April 21-25

Monday afternoon and evening

Television
Tuesday morning

Screens and control of brightness
Tuesday afternoon

Armed Forces production
Tuesday evening

Magnetic projection; Film inspecting;

Future use of educational film
Wednesday morning and afternoon

High-speed photography

Thursday morning
Open

Thursday afternoon
Color ; Laboratory

Thursday evening
General Session

Friday morning

Sound and editing

Friday afternoon
New equipments

There will be demonstrations of equip-
ment, guaranteeing lively and concrete
interest, and we can be sure that Bill
Kunzmann's arrangements are auspicious
for the Get-Together Luncheon, the
Cocktail Hour and the 71st Semiannual
Banquet and Dance.

Bill and all the Chicago folks responsible
for making the many Convention wheels
turn are meeting in Chicago on March 13
for a planning session for which all the
signs are good. The roster of chairmen,
which was completed at an early date,
was published in the February Journal.

267

Book Reviews

The Television Program

Its Writing, Direction and Production

By Edward Stasheff and Rudy Bretz.
Published (1951) by A. A. Wyn, 23 W.
47 St., New York 19. 355 pp. incl. glossary
and index. Numerous illus. and examples.
6 X 9 in. Price $4.95.

The Television Program is probably the
most complete study of television pro-
duction practices and techniques to date.
It should be recommended for a thorough
reading by everyone in television and the
connected industries, either as a means
of reviewing and comparing techniques or
as a method of exploring the nature of the
new medium.

The authors bring a solid background of
actual experience to their work. Edward
Stasheff has been in the industry since
1945, serving as educational consultant
to CBS, assistant program manager to
Station WPIX and as a teacher at Colum-
bia and Michigan Universities. The tele-
vision record of Rudy Bretz is even longer,
including work as a cameraman, as a
teacher and as program manager of WPIX.

For easy assimilation, the book is divided
into four parts: (1) the nature of the
television program, (2) the writing of the
program, (3) the writing of the fully
scripted show, and (4) the producing and
directing of television.

Part One analyzes the television program
and points up the differences between it
and other media. Television production
units are described, with the authors
realistically bringing out the limitations
imposed by budgets, time and space.
Types of television shows are listed and
described by formats and finally the reader
is given a cursory run-down of the basic
shots and visual transitions currently used
in television.

Two sections are devoted to writing the
television show. Of particular interest
to engineers would be sections 12 and 14,
respectively, "Technicalities of Writing
for the TV Camera" and "Transitional
Devices."

Part Four, "Producing and Directing
the Television Program," has the ring of
authenticity that comes from personal
experience. Not only are the functions of

producer-director set forth, but the
spirit motivating the production art is
nicely translated into words. Television
aspirants looking for a "bible" of the
television production art, complete to the
current moment of publication, can find
it here in Part Four.

It should be especially noted that each
section of the book is illustrated with charts,
diagrams and reproduced "air" scripts,
the latter embellished with photographs
of the action as it would be seen by the
television camera.

Next to actual experience and observa-
tion, research is an effective way to learn
television production. For those interested
in using this method, The Television
Program will be found invaluable. — Dik
Darlejy, Director, American Broadcasting
Company, ABC Television Center, Holly-
wood 27, Calif.

Motion Pictures, 1912-1939

A catalog compiled by the Library of
Congress. Published (1952) by the Copy-
right Office, Library of Congress, Wash-
ington 25, D.C. 1256 pp. Bound in
buckram. Price $18.00.

The press release of the Library of Con-
gress describes this as a monumental catalog
that lists more than 50,000 motion pictures
registered in the Copyright Office from
1912 through 1939 and notes that the
catalog contains much information that
has hitherto been available only after
prolonged research in the files of the
Copyright Office. The release also con-
tains the information which follows.

As time passes and old producing com-
panies and their films are forgotten, this
volume will become increasingly valuable
as a reference book on films and film his-
tory. The information given about each
film includes, insofar as possible, the title,
date, producing company, sponsor, in-
formation about the published work on
which the film was based, physical de-
scription, credits, claimant and date of
copyright, and the author of the film story.
The material for the entries, which are
listed alphabetically, was obtained mainly
from the record books of the Copyright
Office, the original applications for the

268

registration of the copyright claims, and
descriptive material that was supplied at
the time the films were registered.

The cumulative catalog has a 268-page
index, which lists the individuals and
organizations associated with each motion
picture, and a "Series List," which provides
the name of the copyright claimant and
the title and date for each motion picture
of a series. Any particular film may be
located in a variety of ways — by title,
producing company, copyright claimant,
alternate title, name of the work on which
the film was based, series title, author of

the film story, sponsor, and releasing or
distributing agents.

Motion Pictures, 1912-1939 is the first
publication in the cumulative series of the
Catalog of Copyright Entries. Work has started
on a supplementary volume that will cover
motion pictures copyrighted in the years
1940 to 1949. These two cumulative
volumes and the subsequent semiannual
issues of Motion Pictures and Filmstrips in
the regular series of the Catalog of Copy-
right Entries will constitute a comprehensive
bibliography of United States motion
pictures from 1912 to date.

Current Literature

The Editors present for convenient reference a list of articles dealing with subjects cognate
to motion picture engineering published in a number of selected journals. Photostatic
or microfilm copies of articles in magazines that are available may be obtained from The
Library of Congress, Washington, D.C., or from the New York Public Library, New
York, N.Y., at prevailing rates.

American Cinematographer

vol. 32, Nov. 1951
Set Lighting by Remote Control (p. 444)

A. Rowan
Reflected Light for Color Photography

(p. 446) L. Allen

Dual-Purpose Projector (p. 450) R. Lawton
Planning and Estimating TV Spot An-
nouncement Films (p. 454) /. H.
Battison

vol. 32, Dec. 1951

Motion Pictures on Tape (p. 500) F. Foster
Trick Effects in TV Commercial Films
(p. 502) /. H. Battison

Audio Engineering

vol. 36, Jan. 1952

The Two Types of Theatre Video (p. 16)
/. W. Sims

Bild und Ton

vol. 4, Oct. 1951

Zur Messung Fotografischer Zentralver-
schlusse (p. 300) H. Peck

Die Bewegungskamera und ihre An wen-
dung (p. 307) W. Rieger

Abmessungen fur 1 6-mm-Transportrollen
und die 1 6-mm-Schaltrolle (p. 318) A.
Heine and L. Busch

British Kinematography

vol. 19, Oct. 1951
The Gevacolor Processes (p. 100) H.

Verkinderen
The Economics of Film Production (p.

110) C. Vinten

Standardization of Projection Lamps (p.

117) M. Furness

vol. 19, Nov. 1951
A Photographic Technique for Producing

High Quality 16mm Prints (p. 132) A.

Tutchings
A Method of Making Travelling Mattes

Using a Single-Film Camera (p. 139)

G. 1. P. Levenson and N. Wells
A Non-Reflecting Room and Its Uses for

Acoustical Measurement (p. 148) F. H.

Brittain

Electronic Engineering

vol. 23, Dec. 1951

Picture Storage Tubes (p. 472) R. E. B.
Hickman

Electronics

vol. 24, Dec. 1951

Improving a Film-Camera Chain (p. 103)
C. /. Auditore

vol. 25, Jan. 1952

Specifications for Color TV Field Tests
(p. 126)

Ideal Kinema

vol. 17, Dec. 6, 1951

Third Dimension Demonstration by Means
of Sextuple Screen (p. 15 and p. 19)

International Projectionist

vol. 26, Dec. 1951
Movie Studio Carbon Arc Lighting (p.

11) H. B. Sellwood
The GPL Simplex Direct-Projection

Theatre TV System (p. 22) F. N.

Gillette

269

vol. 27, Jan. 1952
DuPont's New "Thin" Film Related to

Dacron Fiber (p. 10)
Kollmorgen's New Optics Plant (p. 15)
To Mask or Unmask (p. 16)

Kino-Technik

no. 11, Nov. 1951
Der plastische Film vermag dem Kino

neue Impulse zu geben (p. 224)
Storungen bei der Vorfuhrung von Ton-

filmen (p. 228)

no. 1, Jan. 1952

Siemens-Projektor 2000 — Ein neues Gerat
fur die Schmalfilm-Projektion (p. 10)
L. Busch

Neue Aufnahmetechnik durch das "Travel-
ling Matte"— Verfahren (p. 12)
Storungen bei der Vorfuhrung von Ton-
filmen, Pt. 2, (p. 16) K. Braune and H.
Tummel

Motion Picture Herald

vol. 186, Jan. 5, 1952

(Better Theaters Section)

How Theatres Can Be Revised for "Full

Vision" (p. 8) B. Schlanger and W. A.

Ho/berg

vol. 186, Feb. 9, 1952

Operation and Maintenance of Theatre
TV Equipment, Pt. 6, 35mm Inter-
mediate System (p. 40) A. Nadell

Photo-Technik und-Wirtschaft

vol. 2, Nov. 1951

Uber die Farbentwicklung im Agfacolor-
Negativ/Politiv-Verfahren Reckziegel (p.
446)

RCA Review

vol. 12, Dec. 1951

Fundamental Processes in Charge-Con-
trolled Storage Tubes (p. 702) B. Kazan
and M. Knoll

Radio and Television News

vol. 47, Jan. 1952

Practical Sound Engineering, Pt. II (p.
66) H. Tremaine

(The concluding article of this series
detailing how a complete distribution
system achieves flexibility by means of
patch bays)

Tele-Tech

vol. 10, Nov. 1951
Analysis of Latest Lawrence Color-Tv

Tube (p. 38) /. H. Battison
Image Iconoscope for Improved TV Film

Scanning (p. 44) R. Theile
Combined Special Effects Amplifier for

Television (p. 50) W. L. Hurford

Tele-Vision Engineering

vol. 3, Jan. 1952

Video Studio Techniques (p. 8) C. D.
Parmelee

New Members

The following members have been added to the Society's rolls since those last published.
The designations of grades are the same as those used in the 1950 MEMBERSHIP DIRECTORY,
Honorary (H) Fellow (F) Active (M) Associate (A) Student (S)

Armistead, Mark, President, Mark Armi-
stead, Inc. Mail: 1041 N. Formosa
Ave., Hollywood 46, Calif. (M)

Butler, John W., Executive, Signal Corps
Photographic Center. Mail: 11 W.
Eighth St., New York 11, N.Y. (M)

Con way, David L., Director, Photo-
graphic and Special Events, WHEN,
Meredith-Syracuse Television Co.
Mail: Maple Hill Farm, R.D. #2,
West Monroe, N.Y. (M)

De Titta, Arthur A., Pacific Coast Super-
visor, Movietonews, 1417 N. Western
Ave., Hollywood 27, Calif. (M)

Fox, George S., Producer, Designer,
Cameraman, George Fox Corp. Mail:
6626 Romaine St., Hollywood 38, Calif.
(M)

Fritzen, John, Technical Services, Cine-
color Corp. Mail: 11583 Huston St.,
North Hollywood, Calif. (M)

Gamon, George A., Motion Picture Engi-
neer, Sound Service Co., Pty., Ltd.
Mail: 6 Alameda St., Parkdale, Mel-
bourne, Australia. (A)

Goldberg, Ernest W., President, Golde
Manufacturing Co. Mail: 1140 Michi-
gan St., Wilmette, 111. (M)

Gonzalez, Jesus G., Recording Engineer,
Estudios Tepeyac. Mail: Coquimbo
868, V.G.A. Madero, Mexico City,
D.F. (M)

Goodman, Louis S., Executive Director,
Film Research Associates. Mail: 150
E. 52 St., New York 22, N.Y. (M)

Gordon, Barry O., Instructor, Motion
Picture Photography, Ryerson Institute.
Mail: 42 Roseland Dr., Alderwood,
Toronto 14, Ont., Canada. (M)

Gromak, Theodore B., Engineer, Motio-
graph Corp. Mail: 409 S. Villa Park
Ave., Villa Park, 111. (M)

270

Hall, Robert E., Motion Picture Film
Technician, U.S. Air Force, Wright
Field. Mail: 359 Hilside Rd., Skyway
Park, Fairborn, Ohio. (A)

Heininger, Francis, Writer, Director,
De Frenes Co. Mail: 40 W. Ashmead
PL, N., Philadelphia 44, Pa. (M)

Herbst, R. G., Metallurgist, Bell & Howell
Co. Mail: 9519 Leamington Ave.,
Skokie, 111. (M)

Hurley, Albert B., Manufacturing Execu-
tive, Hurley Screen Co. Mail: Hunt-
ington Bay Rd., Huntington, N.Y. (M)

Karasch, Joseph N., Motion Picture
Photographer, Director, United Auto
Workers, AFL. Mail: 541 Powers
St., Port Washington, Wis. (M)

Kislingbury, William, Cameraman,
Optical Effects, Universal-International.
Mail: 10423 Cheviot Dr., Los Angeles
64, Calif. (M)

Lemmon, Lt. Gene C., U.S. Air Force,
Box 0-20, Edwards Air Force Base,
Edwards, Calif. (A)

MacAllister, Richard, Producer, 16mm.
Mail: 717 Erie Ave., San Antonio 2,
Tex. (A)

Maxfield, Harold H., Design, Structural
Steel, Canadian Brazilian Services.
Mail: 241 Torrens Ave., Toronto,
Ont., Canada. (A)

Miller, James T., Manager, Film Process-
ing, Bry Color Laboratories. Mail:
2020 W. Arthur St., Chicago 45. (A)

Morrissey, Thomas G., Chief Engineer,
Station KFEL. Mail: 5700 W. 28
Ave., Denver 14, Colo. (A)

Nottorf, Robert W., Chemist, E. I. du
Pont de Nemours & Co. Mail: Box
175, Parlin, N.J. (M)

Ozga, Franciszek, Research, Newman
Rudolph Lithographing Co. Mail:
1424 N. Damen Ave., Chicago 22. (M)

Pew, L. Glen, TV Engineer, KPIX
Television, Inc. Mail: 13 Ricardo
La., Mill Valley, Calif. (M)

Poch, Waldemar J., Engineer, Radio
Corporation of America. Mail: 3
Haines Dr., Moorestown, N.J. (M)

Rajagopalan, R., Sound Recordist, c/o
Udaya Studios, Alleppey, South India.
(A)

Ricciardelli, Gino, Assistant Chief Engi-
neer, WNBF-TV. Mail: 151 Robinson
St., Binghamton, N.Y. (M)

Samuelson, Carl, Chemist, Cinecolor
Corp. Mail: 1136 Green La., La
Canada, Calif. (A)

Shaw, Robert B., Mechanical Engineer,
Off. Sect, of Defense. Mail: 8302
Flower Ave., Takoma Park, Md. (M)

Sheldon, Eric J., Vice-President, O. W.
Ray Corp. Mail: 83 Bretton Rd.,
Yonkers, N.Y. (M)

Sobolov, Harold, Instructor, TV Hour

Director, American Broadcasting Co.

Mail: 281 E. 205 St., Bronx 67, N.Y.

(A)
Stantz, Lou veer H., Chief Engineer,

WNBF-TV. Mail: 168 Moeller St.,

Binghamton, N.Y. (M)
Stimson, Allen, Photometric Engineer,

General Electric Co. Mail: 40 Federal

St., Lynn, Mass. (M)
Stinerock, John V., Film Processing

Quality Control Engineer, Eastman

Kodak Co. Mail: 2123 East Ave.,

Rochester 10, N.Y. (A)
Trad, Victor, President, TV Engineer,

Trad Television Corp. Mail: 82 Al-

myr Ave., Deal, N.J. (M)
Trainer, Merrill A., Manager, Broadcast

Equipment Products, Radio Corporation

of America, RCA Victor Div., Bldg.

15-5, Camden, N.J. (M)
Utlek, Sigmund, Laboratory Technician,

Reeves Sound Studios, Inc. Mail:

646 Rosedale Ave., New York, N.Y.

(A)
Woolf, Robert S., Manager, Teletran-

scription Dept., Du Mont Television

Network. Mail: 10 Du Pont Ave.,

White Plains, N.Y. (M)
Ziegler, Allison V., Recording Engineer,

Westrex Corp., Ill Eighth Ave., New

York 11, N.Y. (M)

CHANGES IN GRADE

Bashner, Melvin C., (S) to (A)
Current, Ira B., (A) to (M)
Daniel, George (A) to (M)
Dieter, Henry, (A) to (M)
Hirschfeld, Gerald J., (A) to (M)
Jones, Ronald W., (A) to (M)
LaRue, Mervin W., Sr., (A) to (M)
Macbeth, Norman, (A) to (M)
Ochse, Brand D., (A) to (M)
Pessis, Georges, (S) to (A)
Rocklin, Ralph J., (A) to (M)
Schroeder, Walter A., (A) to (M)
Sherry, Frank E., Jr., (A) to (M)
Spottiswoode, Raymond J., (A) to (M)
Spring, Donald N., (A) to (M)
Streech, Wilbur J., (A) to (M)
Trentino, Victor, (A) to (M)
Youngs, William E., (A) to (M)
Von Vollenhoven, Leopold, (A) to (M)
Ward, Alvis A., (A) to (M)
Whitman, Vernon E., (A) to (M)
Wight, Ralph, (A) to (M)
William, Eric, (A) to (M)
Winn, Curtis B., (A) to (M)
Wolff, Joe M., (A) to (M)
Wutke, Louis, M., (A) to (M)

271

Chemical Corner

Edited by Irving M. Ewig for the Society's Laboratory Practice Committee. Suggestions
should be sent to Society headquarters marked for the attention of Mr. Ewig. Neither
the Society nor the Editor assumes any responsibility for the validity of the statements
contained in this column. They are intended as suggestions for further investigation by
interested persons.

Chemical Treatment Pro-
duces an Oil-Retaining
Rust-Resistant Surface

The Octagon
Process, Inc.,
15 Bank St.,
Staten Island,

N.Y., markets a chemical preparation
"Rustshield," which is a phosphatizing
compound that imparts to a steel or iron
surface a rust-resistant, highly absorbent
quality thereby greatly increasing its oil-
retention properties. The surface needs
less lubrication and remains rust-free
longer.

Increasing Emulsion
Speed

In an article by
L. Jacobs, Jr., in
U.S. Camera, 14:

41-3, March 1951, a new chemical called
"Hydram" is described as increasing
negative emulsion speed as much as ten
times. Hydram is intended for use with
conventional developers. The effect of
this chemical is not to increase the threshold
speed values but to increase the contrast
in the toe part of the H <2?D curve. Hydram
is not recommended when the developer
contains sodium bisulfite, potassium meta-
bisulfite, tartaric or citric acid.

Add More Life The British Journal of
to Your Hypo Photography, 98: 191-92,
April 1951, contains an
article which has various suggestions for
increasing fixer longevity: (1) use of an
acid short stop; (2) use of a two-bath
fixer arrangement; (3) removal of silver
by some suitable means and replenishing
the various chemicals which have been
depleted ; and (4) the addition of ammon-
ium sulfate to the fixer. Ammonium
sulfate imparts to a fixer increased fixing
speed and longer life. Various formulas
are also discussed.

New Plastic Sheeting A transparent

plastic sheeting

manufactured by The Alsynite Co. of

America, 4670 DeSoto St., San Diego,
Calif., may have applications in the
laboratory or on the lot. This plastic is
shatterproof, eliminates glare by light
diffusion, is fire resistant, reduces heat
transmission, has a high impact and load
strength, is light in weight and is easy to
install. It comes in flat panels and various
colors. The sheet may be handled just
like wood with a saw, nails or drill. Al-
synite says that it is an improved substitute
for glass.

Dispensing of Liquid
From Carboys and
Demijohns Greatly
Simplified

L. B. Russell
Chemicals, Inc.,
60 Orange St.,
Bloomfield, N.J.,
sell a small in-
expensive ($15.50) hand-operated dis-
penser which fits into the mouth of carboys
and demijohns of any type. Heavy
carboys do not have to be rocked or tilted.
The mere pressing of the bulb of the
gadget dispenses the liquid from the
carboy and the hazard and odor of splash-
ing liquids are avoided. This device is
made of acid-resistant plastic and fits
all the way down to the bottom of the
container so that all the liquid may be
drawn off.

Liquid Stainless Steel The Lockrey-
Frater Corp.,

38-13 Tenth St., Long Island City, N.Y.,
may have the inexpensive answer to the
laboratory's problem of protecting and
decorating equipment with their "Liquid
Stainless Steel." This is a paint-like
material which is a suspension of finely
divided actual stainless steel combined
with a vinyl plastic. The liquid when
applied will dry fairly rapidly and leave
a coat of stainless steel on wood, metals,
composition board, concrete, brick, etc.

272

It gives the appearance and much of the How to Get Better BFI #20, a pro-
permanence of the metal itself. The Film Washing prietary formula of
coating offers the impermeability to The Brown Forman
moisture that 302 stainless steel does by Industries, 1908 Howard St., Louisville,
an overlapping and interlocking of the Ky., is claimed to increase washing
flakes as they dry. Liquid Stainless Steel efficiency and to reduce to one-twentieth
may be applied by spray, brush, or dip the amount of hypo remaining in the film
and gives a permanent coating in the which would be there if washed with water,
bluish-gray, non-shining cast of stainless One gallon of BFI #20 will treat 36,000
steel. ft of 35mm film.

Meetings

The Central Section of the SMPTE has scheduled two papers for its meeting at the
Bell & Howell Co., 7100 McGormick Blvd., Chicago, on March 27. Bruno G. Staffen,
development engineer of the Jensen Manufacturing Co., will describe a new low-cost
theater speaker system, and there will be a description of the new Bell & Howell magnetic
and optical 16mm sound projector by J. B. Weber, H. H. Brauer, F. J. Schussler and
M. G. Townsley. C. E. Heppberger is Central Section Chairman, and John S. Powers
is Program Chairman.

The Atlantic Coast Section of the SMPTE will meet on April 16, 7:30 P.M., at the
Henry Hudson Hotel, New York City, when Robert Dressier of Paramount Pictures
Corp.'s Chromatic Television Laboratories will present a paper and a demonstration on
electrooptic sound recording on film.

71st Semiannual Convention of the SMPTE, April 21-25, The Drake, Chicago

Other Societies

American Physical Society, Mar. 20-22, Columbus, Ohio
Optical Society of America, Mar. 20-22, Hotel Statler, New York
American Physical Society, May 1-3, Washington, D.C.
Acoustical Society of America, May 8-10, New York

American Institute of Electrical Engineers, Summer General Meeting, June 23-27,

Hotel Nicollet, Minneapolis, Minn.

American Physical Society, June 30-July 3, Denver, Colo.

National Audio- Visual Association, Convention and Trade Show, Aug. 2-5, Hotel Sher-
man, Chicago

Photographic Society of America, Annual Convention, Aug. 12-16, Hotel New Yorker,

New York

American Institute of Electrical Engineers, Pacific General Meeting, Aug. 19-22, Hotel

Westward Ho, Phoenix, Ariz.

Illuminating Engineering Society, National Technical Conference, Aug. 27-30, Wash-
ington, D.C.

International Society of Photogrammetry, Conference, Sept. 4-13, Hotel Shoreham,

Washington, D.C.

Test films are the customary tool for checking picture and sound performance in theaters,
service shops, in factories and in television stations. Twenty-seven different test films
in 16mm and 35mm sizes are produced by the Society and the Motion Picture Research
Council. Write to Society Headquarters for a free catalog.

275

New Products

Further information about these items can be obtained direct from the addresses given.
As in the case of technical papers, the Society is not responsible for manufacturers* state-
ments, and publication of these items does not constitute endorsement of the products.

A developmental transistor was unveiled
to the daily press during the last week of
February, and the Gamden, N.J., press
department of RCA-Victor has released
the accompanying photograph, which
shows, in approximate life size, a transistor
in an advanced stage of construction (right)
and the finished transistor, with its com-
ponents embedded in a protective casing
of plastic (left). This transistor, based
on the Bell Telephone Laboratories' de-
velopment, is designed to perform many
of the functions of a vacuum tube and can
substitute for it in many applications.
Because of its minute size and low power
requirements, it is expected that it will
make possible an important reduction in
the size of many electronic devices used
for television, radar and hearing aids.

The Berkshire Lab strobe, Model 18,
has been designed as a small, light, eco-
nomical stroboscope unit, using a standard
neon bulb. It is available from the
Berkshire Laboratories, 546 Beaver Pond
Rd., Lincoln, Mass., with specifications:
Power source .... 11 5-v, 60-cycle, a-c
Power consumption .... less than 1 w
Flashing rate 60 cycles/sec

(determined by line frequency)
Flash duration approx. 100 ^sec

(measured at 50% of peak intensity)
Net price, f.o.b. Lincoln, Mass . . $9.95

The manufacturers suggest that it will
prove useful for determining the speeds

A light-weight sound-proof blimp for
the Arriflex camera has been announced
by Kadisch Camera & Sound Equipment
Co., 128 West 48th St., New York 19,
N.Y., manufacturer of the Arriflex. The
blimp has an external control for follow-
focus, and a built-in synchronous motor,
and accepts 200- and 400-ft magazines.
A new extension eyepiece through the
blimp permits viewing through the lens
during shooting. The new blimp is easily
accessible for rapid changing of magazines.

of rotation of motors, machines, phono-
graph turntables and other objects. The
instrument is housed in a chromium-
plated case and is the same size as a
standard two-cell flashlight.

274

The Bell & Howell Filmosound 202

16mm optical-magnetic recording pro-
jector is now being marketed to non-
professionals to make possible movies with
sound at a cost of $200 for a 10-min film.
Bell & Howell particularly emphasizes
the usefulness of such an assembly to
small manufacturing or marketing com-
panies which make use of training and
sales films, to educational and church
users, to those who wish to exhibit films
in a number of languages or dialects (as
in India), etc.

The Filmosound magnetically records,
plays back and erases, while the film is
being projected, on single perforated color
or black-and-white film which has been
magnetically striped. Bell & Howell
says that their Soundstripe will outlast
the film, may be re-recorded indefinitely,
and may be economically applied to the
film at a cost of 3^ a foot. The Filmo-

sound, which costs $699 with a 6-in.
speaker contained in the case, operates
for recording at either 16 or 24 frames/sec.
A 12-in. auxiliary speaker is also available.
Although the sound quality is better at
24 frames/sec, Bell & Howell reports that
either gives acceptable results, equal to
or better than commercially available disc
recordings.

Soundstripe, which is a magnetic iron
oxide stripe, may be striped over the full
sound track area of single-perforated film ;
thus silent movies taken or duplicated on
single-perforated film or optical sound
film with obsolete sound tracks can be
treated with full Soundstripe. Or half
the sound track on optical film may be
processed with the magnetic stripe, thus
making it possible to record and play
back the magnetic track or play back the
optical sound track.

275

The Utility Television Monitor Model
CA16 is now being produced by Gonrac,
Inc., 19217 East Foothill Blvd., Glendora,
Calif. This monitor has been designed
for general purpose use by television
studios, both in control rooms and on
stage, with these specifications reported:
a fully rectangular picture presentation of
9 in. X 12 in. on the 16GP4 kinescope;
wide-band video amplifier with a smooth

roll-off above 7 me; and a total of 14
tubes in addition to the kinescope.

Design features planned for the con-
venience of operating personnel are:
coaxial input connectors and a switch to
select either composite video, or separate
video and composite sync; both inputs
equipped with parallel receptacles for
multiple connection; heavy-gauge steel
cabinet housing with carrying handles;
and a removable front to facilitate cleaning
the kinescope face and the protecting
safety glass.

The 1952 Catalog of Films From Britain

is now available upon request to British
Information Services, 30 Rockefeller Plaza,
New York 20, N.Y. Nearly 300 16mm
sound films are described. These are
available as rentals or purchases from
regional British Information Services offices
and some dealers. In addition to a de-
scriptive listing of all general and special-
ized films, the British Information Services
reports that the special section "The
Motion Picture — The Art and Its Artists
(Experimental and Classic Documentaries,
including Academy Award Winners)" was
added in response to the requests of the
many film societies, colleges and uni-
versities.

Six American Standards have been added to the Motion Picture Set of 60 which the
Society has had available for sale. To holders of the present set the Society has made
available the six new standards: PH22.11-1952, PH22.24-1952, PH22.73-1951, PH22.74-
1951, PH22.76-1951 and PH22.82-1951. The price is $1 plus 3% sales tax on deliveries
in New York City.

The new set of 66 standards in a heavy three-post binder with an index is available at
$14.50 plus 3% sales tax on deliveries in New York City; foreign postage is $.50 extra.

All standards in sets only are available from Society Headquarters. Single copies of
any particular standard must be ordered from the American Standards Association,
70 East 45th St., New York 17, N.Y.

Back issues of the Journal available: Don Canady, 5125 Myerdale Drive, R.R. 15,
Cincinnati 36, Ohio, desires to dispose of a complete set, in excellent condition, from
JarTuary 1930 to date, plus one issue of September 1928. Anyone interested in acquiring
the complete set should communicate directly with Mr. Canady.

SMPTE Officers and Committees: A new roster of Society Officers and the
Committee Chairmen and Members will be published in the April Journal.

276

The Nature and Evaluation of the
Sharpness of Photographic Images

By G. C. HIGGINS and L. A. JONES

The ability of a photographic material to produce pictures having good defini-
tion is commonly referred to as its sharpness, which is a subjective concept.
The objective quantity ^GX2)/^'DS is shown to be a physical measurement
which correlates with sharpness judgments. (G^8)*, is the mean of the
square of the density gradients, AD/ Ax, across an abrupt boundary between
alight and a dark area in the developed image and AS* is the density difference
between these areas. {Gx2}/^ is evaluated only for those values greater than
0.005 in density per micron which represents the threshold gradient. It is
shown that, contrary to the generally accepted belief, resolving power does not
correlate well with sharpness judgments and in some cases is even misleading.

1\.N IMPORTANT property of a photo-
graphic material is its ability to produce
pictures having good screen definition.
This property of a material is commonly
referred to as its sharpness. Sharpness
defined in this manner is a subjective
concept.

The obvious usefulness of an objective
measurement which will predict the
sharpness of pictures made on a photo-
graphic material led to an investigation
of the nature of sharpness and the physi-
cal properties of the picture which are
important in producing sharp images.

Communication No. 1459 from the Kodak
Research Laboratories, a paper presented
on October 15, 1951, at the Society's Con-
vention at Hollywood, Calif., by G. C.
Higgins and L. A. Jones, Eastman Kodak
Company, Kodak Park Works, Rochester
4, N.Y.

During the course of an investigation by
Jones and Higgins1 on photographic
graininess and granularity, the mode of
functioning of the human visual mecha-
nism was examined in some detail.

It is generally accepted that the mag-
nitude of the neural response, which initi-
ates the sensory or perceptual response
which occurs when a cone in the eye is
stimulated, is determined by the sudden-
ness with which the stimulation changes.
The cones, which are the receptors in the
eye for photopic or daylight vision, there-
fore respond primarily to temporal illu-
minance gradients, A5/A/. When ex-
amining any object in a visual field, the
eye is constantly moving, with the result
that the cones repeatedly scan the image
formed on the retina. The distribution
of luminance in the object produces a

April 1952 Journal of the SMPTE Vol. 58

277

Db

2.0 -

1-5-

1.0

0.5

Do

\\\\\\\\\\\\\\

11111

10

20 30

Distance on sample

40

50

Fig. 1. Schematic diagram representing a knife-edge exposure and the micro-
densitometer trace, D, across the developed image. The straight line, £, between
points A and B and the hypothetical dotted curve, F, represent traces having the
same average gradient as curve D.

distribution of illuminance in the image
formed by the lens of the eye. The mo-
tion of the eye then permits each of the
cones to scan this illuminance distribu-
tion, the response of the cones being pro-
duced by the temporal illuminance gra-
dients, AB/At. This temporal luminance
gradient consists of two components, the
temporal component, Ax/ At, produced by
eye motion, and the spatial component,
A/?/ A*, produced by the object being
viewed, the product of the two compo-
nents being A#/A/. The spatial lumi-
nance gradients, A/?/ A*, in the visual field
therefore represent the physical aspects
of the object which control the perception
of detail.

The concept of gradient sensitivity has
proved useful in finding an objective
measure of a granularity which corre-

lates with graininess. It appeared logi-
cal, therefore, to apply the gradient sensi-
tivity principle to the problem of obtain-
ing a physical measurement which will
correlate with sharpness judgments of
pictures.

When a photographic material is ex-
posed while partially shielded by a knife-
edge in contact with the emulsion, as
shown schematically at the top of Fig. 1 ,
the developed image does not end
abruptly at the knife-edge but encroaches
on the shielded area and has a diffuse
boundary. A microdensitometer trace
across a knife-edge image, made as shown
schematically at the top of Fig. 1, is repre-
sented by curve D in the lower part of
this figure. The ordinates represent
density and the abscissas, distance on the
sample in microns. When judging the

278

April 1952 Journal of the SMPTE Vol. 58

sharpness of this image, the cones of the
eye move back and forth across this bound-
ary in much the same manner as the
fingers move back and forth across a
piece of cloth when judging its roughness.
The density gradients, AD/ Ax, across this
boundary become log illuminance gra-
dients, A log B/Ax, in the image formed
on the retina. The motion of the eye
converts these spatial log illuminance gra-
dients into temporal log illuminance
gradients which are the stimuli for the
cones. The gradients are evaluated in
terms of AD/ Ax rather than AT/ Ax, since
the response of the eye to luminance
differences is known to be essentially
logarithmic.

It has been suggested that the maxi-
mum value of the gradient, AD/ Ax, as
shown at C, should be an indication of
the sharpness of the image. However,
experiments by Wolfe and Eisen2 in these
Laboratories have shown that the maxi-
mum gradient does not correlate with
sharpness judgments. These same in-
vestigators have shown that the average
gradient between any two points on this
curve, such as A and B, also fails to corre-
late with sharpness judgments.

The average gradient, {G^/w, between
A and B is independent of the density
distribution between the points. Curve
D, which represents a microdensitom-
eter trace, the straight line. E, between
A and B, and the hypothetical dotted
curve, F, all give the same value of aver-
age gradient, (GI}Av. If the physical as-
pect of the sample which determines the
response of the cones is AD /Ax, then the
sharpness of the image should depend
upon the rate at which the gradient
changes across the edge. That is, the
distribution of density across the edge
represented by the three lines joining the
points A and B should lead to three dif-
ferent sensations of sharpness. From the
study of gradient sensitivity in connec-
tion with the investigation of graininess
and granularity, it is known that the
threshold gradient sensitivity in the
photopic range is approximately 0.005 in

density per micron. However, this
threshold gradient, as indicated by the
points A and B, is only an approxima-
tion and may have to be modified as
more data are accumulated.

While there are numerous methods of
evaluating the gradients in such a man-
ner that the results will depend upon
their distribution across the boundary,
we have chosen to use the mean of their
squares between the limits of 0.005 per
micron. This average, (Cz2)Av is equal
to fj*(dD/dx)*dx/(Xb - Xa). We chose
to use (Gx 2)Av since it is equal to the prod-
uct of the average gradient measured at
equal increments of D and the average
gradient measured at equal increments of
x; (Gx2)Av = (G)Av(D)-{G)Av(;c). It seems
probable that in obtaining the average
gradient, its evaluation should depend
upon equal increments of D, since the
problem involved is that of perceiving
luminance differences and, for any given
viewing condition, AD corresponds to a
difference in log illuminance on the ret-
ina. This method of averaging has
been found to yield fruitful results in
obtaining a numerical specification of
the contrast of printing papers, which
is a somewhat similar problem.

On the basis of the knowledge of the
mode of functioning of the eye, it seems
probable that the subjective impression
of sharpness should depend not only upon
(G^)^ but also upon the density differ-
ence, DS, between the light and the dark
areas. On the trace shown in Fig. 1.
DS is equal to Db — Da. The objec-
tive quantity, (Gx2)Av • DS, was therefore
investigated as a physical measurement
which it seemed reasonable to expect to
correlate with picture sharpness. We
suggest that physical measurements based
upon the density variation across a
knife-edge image be termed "acutance."
The formula (Gx2)Av • DS, therefore, gives
values of acutance.

Wolfe and Eisen2 prepared matched
transparencies of the same scene printed
on fine-grain positive film from ten differ-
ent negative materials. The sharpest

Higgins and Jones: Evaluation of Image Sharpness

279

Figure 2A

Fig. 2. The sharpest (B) and the least sharp (A) pictures made from
negatives on ten different photographic materials.

280

April 1952 Journal of the SMPTE Vol. 58

i

Figure 2B

Higgins and Jones: Evaluation of Image Sharpness

281

Distance on sample

Fig. 3. Microdensitometer traces across the images in the positive as printed
from knife-edge images on ten different negative materials.

picture, B, and the least sharp, A, made
from these negatives are shown in Fig. 2.
It is evident that the maximum differ-
ence in sharpness between the pictures
made on these ten materials is relatively
small. By subjective judgments of the
relative sharpness of the positives from
the ten negative materials, Wolfe and
Eisen assigned numerical sharpness
values to all positive transparencies.
This method, which is an introspective
psychological one, yields numbers of a
purely ordinal nature which are not re-
lated to the objective character of the
stimulus. The differences in sharpness
in many instances were so small that
different observers ranked a given pair of
pictures in different orders, even though,
based on the judgment of all observers,
there was a real difference in sharpness
between the two reproductions.

Knife-edge images were printed onto
all negatives, and these images in the
negative were, in turn, printed by contact

onto fine-grain positive film. The
microdensitometer traces across the
knife-edge images in the positive are
shown in Fig. 3. While the differences
in the traces appear quite small, curve
10, which represents the trace on the
sharper picture in Fig. 2, shows a higher
slope and a more abrupt toe and shoulder
than curve 1, which represents the trace
on the least sharp picture shown in Fig.
2. As noted previously, the maximum
slope of the curves does not correlate with
sharpness. For example, curves 3 and 4
have essentially the same maximum
slope, while all observers find pictures on
material 4 sharper than on material 3.
The significant difference between these
two curves is the rounding-off of the
shoulder and the slightly lower density
scale on curve 3 as compared with
curve 4.

The acutance values, (Gx2)Av • DS, were
calculated for all traces and are plotted
as a function of sharpness in Fig. 4.

282

April 1952 Journal of the SMPTE Vol. 58

The coefficient of correlation between the
objective and the subjective measure-
ments is 0.994. The relation shown in
Fig. 4 is psychophysical, since it shows
the correlation between a subjective
(psychological value) and an objective
(physical) factor. All materials are
ranked in the same order and are spaced
approximately the same on the sharpness
scale. The very small difference in
sharpness between prints from negative
materials 1 and 1 0, as shown in Fig. 2, is
represented by an acutance difference of
1620 or more than 100% of 1350, the
value for the least sharp material.

For many years it has been the prac-
tice in the photographic field to report
values of maximum resolving power for
the different materials. Resolving power
is usually measured by photograph-
ing a series of line gratings and determin-
ing the number of equal-width lines and
spaces that are just resolvable when the
developed image is examined visually
under adequate magnification. While
these measurements were intended spe-
cifically as a measure of the ability of the
film to record fine detail, such as images
of double stars or fine parallel lines, it has
been generally assumed that these resolv-
ing-power values were a measure of the
ability of the material to produce sharp
pictures. However, experience has shown
that resolving power as usually measured
does not correlate well with sharpness
judgments and in some cases may even
be misleading.

The lack of correlation between resolv-
ing power and sharpness is strikingly
shown by the prints in Fig. 5. The
same negative was printed onto two ex-
perimental positive materials to give the
best-matched tone reproduction possible.
The positive material used in printing
picture A has a maximum resolving
power in excess of 230 lines per millimeter,
while the positive material used in print-
ing picture B has a maximum resolving
power of 1 30 lines per millimeter. Even
though the material used in making print
B has a very much lower resolving power,

^2000

60 80 100 120

Relotive shorpness

Fig. 4. Acutance, {<V}Av • Gtf, plotted
as a function of sharpness.

the picture is clearly much sharper than
print A, obtained by printing on the
high-resolving-power material. When
making the pictures shown in Fig. 5,
knife-edge images were printed onto the
two positive materials. The micro-
densitometer traces across these knife-
edge images are shown in Fig. 6. The
difference between these two traces is
readily apparent. Trace A, represent-
ing the less sharp material, has a very low
slope and a long toe and shoulder, while
trace B, representing the sharp material,
has a relatively high slope and an abrupt
toe and shoulder. The density scales
are essentially the same for both mate-
rials. The value of <Gx2)Av-Z>S for the
very sharp material is 12,210, while the
value for the unsharp material is only
2,800.

The basic principle underlying the
method of obtaining a physical measure-
ment correlating with sharpness judg-
ments of the photographic image should
also apply to the evaluation of lenses
where the luminance gradients of impor-
tance are those in the areal image formed

Higgins and Jones: Evaluation of Image Sharpness

283

Figure 5A

Fig. 5. Prints from the same negative printed onto two experimental posi-
tive materials; material A having a maximum resolving power of 230 lines per
millimeter and material B, a maximum resolving power of 130 lines per
millimeter.

284

April 1952 Journal of the SMPTE Vol. 58

Figure 5B

Higgins and Jones: Evaluation of Image Sharpness

285

If

if

0 10

Relative focal position (mm)

Fig. 6. Microdensitometer traces
across knife-edge images printed onto
the two positive materials used in making
the pictures shown in Fig. 5.

Fig. 7. Maximum resolving power
and relative sharpness of pictures
plotted as a function of the relative
distance from lens to film when making
the negatives.

Fig. 8. Photographic reproductions of the image of a point source as formed with

a lens. B was made at the image distance giving maximum sharpness, and A was

made at the image distance giving maximum resolving power.

286

April 1952 Journal of the SMPTE Vol. 58

by the lens. From the standpoint of
photographic reproductions it is, of
course, also necessary to examine the
manner in which these luminance gra-
dients in the areal image are reproduced
as density gradients in the negative and
in the positive. Wolfe and Eisen3
examined a 12-in. lens designed for aerial
photography by photographing the same
picture repeatedly, the photographic
material being placed at different dis-
tances from the lens. These negatives
were then printed onto photographic
paper and the resulting prints were
judged for sharpness. A standard re-
sol ving-power test chart was also photo-
graphed under the same conditions em-
ployed in making the picture negatives.
Maximum resolving power, as measured
with a high-contrast test object, and rela-
tive picture sharpness are plotted as func-
tions of image distance in Fig. 7. The
abscissa values represent distance in
millimeters from an arbitrary origin.
As shown, the position of maximum re-
solving power is approximately 1 mm.
from the position of maximum sharpness.
Wolfe and Eisen3 also photographed a
point source and a knife-edge under the
same conditions that were employed in
making the picture negatives. Photo-
graphic reproductions of the images of
the point source are shown in Fig. 8. At
the position of maximum sharpness,
shown at B, the image of the point is
fairly large but has very sharp edges,
with practically no variation in density
outside the central image, while at the
position of maximum resolving power,
shown at A, the image of the point is
represented by a small dot surrounded by
relatively large variations in density in
the form of several light rings. Micro-
densitometer traces across the knife-edge
images are shown in Figure 9. The
trace made at the position of maximum
sharpness, B, has a very high slope and a
little toe or shoulder, while the trace
made at the position of maximum resolv-
ing power, A, has a relatively low slope
and a very pronounced toe and shoulder.

Fig. 9. Microdensitometer traces
across photographic reproductions of
the knife-edge image as formed with a
lens. B was made at the image distance
giving maximum sharpness, and A was
made at the image distance giving
maximum resolving power.

The values of (G^Av * DS as obtained from
the traces representing maximum sharp-
ness and maximum resolving power are
620 and 165, respectively. The acu-
tance criterion indicates the focal posi-
tions giving maximum sharpness, while
the criterion of maximum resolving
power represents a focal distance 1 mm re-
moved. The pictures shown in Fig. 10
were made with the film at the position
of maximum sharpness, B, and at the
position of maximum resolving power, A.

All data taken to date indicate that
acutance measured as (G^Av • DS can be
used to predict the sharpness of pictures
made with different photographic ma-
terials. The data also indicate that this
concept is useful in evaluating the sharp-
ness characteristic of an image produced
by a lens. However, the density differ-
ence, DS, across the knife-edge image is
essentially the same for all samples in-
vestigated. The data, therefore, are
not conclusive as to whether the DS term
should be introduced, or if so, whether it
should be introduced as a weighted func-
tion.

While it is shown that resolving power
as usually measured cannot be used to

Higgins and Jones: Evaluation of Image Sharpness

287

Figure 10A

Fig. 10. Photographic reproductions of the same scene made at the image

distance giving maximum sharpness, B, and at the image distance

giving maximum resolution, A.

288

April 1952 Journal of the SMPTE Vol. 58

Figure 10B

Higgins and Jones: Evaluation of Image Sharpness

289

predict with certainty the ability of a
photographic material or a lens to pro-
duce sharp pictures, it is nevertheless an
important property of the materials.
When viewed at 14 in. under optimum
conditions, the eye can resolve a maxi-
mum of about ten black and white lines
per millimeter. The resolving power of
the film or lens must be sufficient to
satisfy the limit set by the eye for a given
viewing condition. We believe that,
from the standpoint of sharpness, the im-
portant property of the image is the acu-
tance of the edges of lines which are just
resolved by the eye. Acutance measure-
ments on lines of different widths and

different frequencies, as well as different
contrasts, should give this information.
Resolving power is therefore a limiting
condition which does not furnish infor-
mation as to the sharpness of detail which
is well resolved by the eye.

References

1 . Loyd A. Jones and George C. Higgins,
"Photographic granularity and graini-
ness: III. Some characteristics of the
visual system of importance in the evalu-
ation of graininess and granularity,"
/. Opt. Soc. Am., 37: 217-263, Apr. 1947.

2. R. N. Wolfe and F. C. Eisen, unpub-
lished work.

3. R. N. Wolfe and F. C. Eisen, unpub-
lished work.

290

April 1952 Journal of the SMPTE Vol. 58

Progress inThree-Dimensional Films
at the Festival of Britain

By RAYMOND SPOTTISWOODE

The planning for the Telecinema is described, then the building and the
projection equipment. Also discussed are the developing of stereoscopic
cameras and new formulas, producing the films, and introducing stereo-
phonic sound and large-screen live television shows. The success of various
parts of the program is evaluated and possibilities for the future assessed.

-L HE FESTIVAL OF BRITAIN, 1951, was
planned as a mid-century stock-taking
of Britain's achievements in the arts
and sciences, combined with an attempt
to pierce into the future and foreshadow
the developments of the next 50 years.
The Great Exhibition of 1851 had stuck
obstinately to the present; in fact it
had dismissed electricity as a mere toy,
and had treated the finding of oil as no
more than a convenient replacement for
candles. The planners of 1951 were
determined not to be caught napping.
Their centerpiece was an exhibition
site on the South Bank of the Thames
in London; and here, in a series of
daringly executed buildings, they pre-
sented thematically the story of the

Presented on October 19, 1951, at the
Society's Convention at Hollywood, Calif.,
by David R. Brower, Assistant to the
Manager, University of California Press,
Berkeley, Calif., for the author, Raymond
Spottiswoode, Kingsgate, Sudbury Hill,
Harrow-on-the-Hill, England, who was
Technical Director, Stereofilm Program,
Festival of Britain, 1951.

people of Britain, their origin, their
environment, their way of life, their
discoveries.

From the very beginning, the motion
picture had its place in this thematic
treatment. Despite periodic ups and
downs, British studios have made notable
contributions to the art of the film, and
these were commemorated in 1951 by
a cooperative production, The Magic
Box, which told the story of William
Friese-Greene, one of the pioneers who
aided in the invention of the movie
camera.

The Festival authorities provided on
the South Bank a new building, the
Telecinema, and a new program in
which, for the first time in the world,
live big-screen television and three-
dimensional films were to be combined
on an equal footing as an entertainment
foreshadowing the movies of the future.

Glancing ahead for a moment, it may
be recorded that the Telecinema and
its program was one of the outstanding
successes of the Festival. With only

April 1952 Journal of the SMPTE Vol. 58

291

Fig. 1. The Telecinema building.

400 seats, it grossed in five months about
$225,000, converted at the old rate of
exchange normally used for economic
comparisons. The total audience was
very nearly half a million; but this could
have been greatly increased, if the
Festival had not had commitments to
include in the program a number of
documentary films which had been
specially produced for it. As it was,
with seven to nine shows a day, the public
had to queue for between one and three
hours to get in — a period often very
much longer than that of the program
itself. Yet throughout the 22 weeks,
there was not a single complaint, and
many people returned to the Tele-
cinema again and again.

In the short space of this paper, I
shall try to describe the Telecinema
building and the events which led up to
the completion of the 33-min series
of stereoscopic and stereophonic films.
If I say little about television — for
which 1 ,220 live shows were produced —
it is only for lack of space; it played a

vital part in the construction of our
programs.

The Building

Work on the Telecinema was started
late in 1949. The Festival was
extremely fortunate in its choice of
architect. Wells Coates, though hamp-
ered by a narrow site pressed close
against a railroad bridge, succeeded in
producing a building of elegant and
simple lines, with a seating capacity of
400 and adequate space for the many
supplementary services required (Fig. 1).
The inside of the theater (Figs. 2 and 3)
is austerely simple, but it is saved from
any feeling of severity by its attractive
color scheme of varying shades of blue.
The Festival motif was introduced in a
Venetian-blind curtain of original de-
sign. The building was laid out exclu-
sively for use with modern safety-base
film, thus allowing certain precautionary
measures to be dispensed with, and
permitting a type of construction (some-
times from its shape called "lobster-

292

April 1952 Journal of the SMPTE Vol. 58

Fig. 2. Interior of the Telecinema — view from the stalls.

Fig. 3. Interior of the Telecinema — view from the circle.
Raymond Spottiswoode: Three-Dimensional Films

293

claw") in which the projection room
is enclosed in the space between the
upper and the lower tiers of seats (see
Fig. 4). This gives a horizontal pro-
jection beam, with a picture free from
keystone distortion, and also provides a
platform within 45 ft of the screen for
mounting out of sight the Schmidt-type
television projection equipment. In
the Telecinema, this projector (built by
Cinema-Television, Ltd.) was placed
centrally, and was swung out of the way
for film projection by means of a turn-
table and rails. This structural arrange-
ment necessitates a rather high position
for the screen, and the front seats in
the theater are accordingly given a
reversed slope. The Stableford screen
was of the high-gain, non-depolarizing
type, equally suitable for television,
three-dimensional and flat films, and,
in spite of the metallic surface, a re-
markably wide light distribution is
secured by special design. Uniform
screen brightness from the side front
seats is aided by giving the screen a
slight cylindrical curvature of a radius
equal to the projector throw. Though
the screen itself has a width of 20 ft,
the image width is only 15 ft, the re-
maining area forming a band around
the picture which receives a diffused
light picked up from the film itself and
projected onto the screen by a device
produced by the British Thomson-
Houston Co.

Figure 5 shows the disposition of some
of the equipment in the projection-room,
as viewed through the large glass window
which enables the audience when enter-
ing the theater to see "what makes the
wheels go round." The television
equipment consists of a camera and
control console (not shown) which feed
a video signal to the console on the
extreme left, from which the signal
passes to the projector placed imme-
diately in front of the front wall of the
projection room. The film projectors
are BTH S/U/P/A machines syn-
chronized by selsyn interlock, and

behind them stand two BTH-HMV 4-
track magnetic recorder-reproducers for
handling the stereophonic sound tracks.
Non-sync magnetic machines and com-
plex switchgear complete the projection
room installation.

This is the equipment which rendered
such satisfactory service throughout the
Festival in 1951. But in the early part
of 1950 no equipment of any kind was
available in England for producing or
projecting stereoscopic and stereophonic
films. All of it had to be designed and
built, and the films produced, in only
14 months. First to be put in hand
was the magnetic recording and re-
recording equipment. In order to
reduce inter-track magnetic interference,
it was decided to employ no more than
four sound tracks, the wide dynamic
range making a control track unneces-
sary. Three of these tracks were to
feed three banks of loudspeakers placed
symmetrically across the screen (Fig. 6),
the outer ones being set as far apart as
possible to widen the sound base. Thus
only a single track remained for feeding
the groups of loudspeakers mounted and
wired in parallel behind the balcony
and stalls, and (again in parallel) in
the main ceiling and in the ceiling of the
rear stalls.

In the writer's opinion, the use of
three channels behind the screen has not
been adequately justified, the Philips
company in Holland having given
extremely convincing demonstrations of
back-of-screen stereophonic sound em-
ploying only two channels, the center
loudspeaker being fed with low-frequency
nondirectional sound from a bridge
circuit.

Shortly afterwards, construction work
was started on a stereoscopic camera
based on two Newman-Sinclair units
facing inwards in conventional fashion
toward a pair of mirrors, and so mounted
that the inter-lens separation (stereo-
base) could be varied from 1 to 8 in.,
and the half-angle of convergence
(stereoangle) from 0° to 5°.

294

AprU 1952 Journal of the SMPTE Vol. 58

Fig. 4 Section of the Telecinema through center line.

Fig. 5. Projection room as seen by those entering the Telecinema.
Raymond Spottiswoode: Three-Dimensional Films

295

Fig. 6. Profile of the Telecinema.

Key to Television Operations

1. Scene being enacted in foyer of the
theater.

2. Floodlights.

3. Television camera.

4. Sound monitor control.

5. Amplifiers.

6. TV control console.

7. TV projector.

8. Runway and turntable for TV pro-
jector when films are being shown.

9. Actual scene being taken in the foyer
projected on to the high-grain screen
simultaneously.

10. Main loudspeakers.

11. Auxiliary speakers which can control
sound from any portion of the screen.

12. Sound frame funnels to audience.

13. Insulated main walls.

14. Insulated studs.

15. Air-control vents.

Key to Projection Room

16. Stereophonic four-sound-track mag-
netic reproducers, one of which is
coupled to the two projectors.

1 7. Two projectors giving synchronous left-
and right-eye pictures.

18. Interval music sound tracks.

19. Film re winder.

20. Projector control panel.

21. Main switch gear.

22. Glass screen to foyer.

23. Vent from projectors.

24. Incoming B.B.G. Television to control
console.

Other Parts

25. Balcony, 150 seats.

26. Suspended roof.

27. Roof lights.

28. Loudspeakers.

29. Entrance to balcony.

30. Mezzanine floor and manager's office.

31. Main entrance.

32. Entrance to stalls.

33. Loudspeakers.

34. Stalls, 252 seats.

35. Ground level and exit from stalls.

296

April 1952 Journal of the SMPTE Vol. 58

During the months that elapsed when
this equipment was taking shape, two
other projects were put in hand. In
order to augment the program, it was
decided to invite the National Film
Board of Canada, known throughout
the world for its experimental films,
to undertake a three-dimensional
abstract film with stereophonic music,
the first to be made- any where. Their
response was most generous, and the
film Around is Around (which was pre-
sented with the paper by McLaren1 at
this Convention) was put into produc-
tion.

A Theory of Stereoscopic Transmission

Secondly, as a result of careful study
of the literature of the three-dimensional
film, it became apparent that knowledge
of the transformations and distortions
of the stereo image was still exceedingly
scanty, and most of the recommendations
were empirical, in spite of the excellent
preliminary work carried out by Rule,2

Norling3 and others. The present
writer, with his brother, N. L. Spottis-
woode, therefore set about evolving a
comprehensive theory of stereoscopic
transmission, which is shortly to be
published as a book of that name by the
University of California Press. A single
master equation determines the shape
of the image under all possible variations
in the camera and projection systems,
while a series of about 80 subsidiary
equations makes possible the design of
convenient calculators, and elucidates
many peculiarities of the three-dimen-
sional image not hitherto studied.

The four films in our Telecinema
program (widely different in their style
and subject matter) were all produced in
conformity with this theory. Today,
the director of a three-dimensional film
has only to state what position in the
ultimate movie theater he wishes a
landscape or a studio scene to occupy,
in order to fit the mood or the editing
of a sequence; and in a few moments

Raymond Spottiswoode: Three-Dim ensional Films

297

the stereotechnician beside the camera
will have established the precise shooting
conditions for realizing this intention.
If, moreover, there are psychological
factors which will tend to alter this
geometrical placement of the image in
cinema space, he will be able to make
proper allowance for them. By the
same token, the producer of an animated
cartoon film (using standard one-lensed
equipment) can now work with as much
accuracy in three dimensions as he
formerly did in two.

We owe it to the organizers of the
Festival and to the British Film Institute
that we were thus able to devote many
months to a subject of no immediate
utility, but which none the less greatly
simplified the productions which were
to follow, and which will, it is hoped,
be of service to the industry in general
if three-dimensional films come into
widespread use.

Production

The special stereoscopic camera was
not completed in time to shoot with it
the Monopack Technicolor film which
had been planned. Accordingly, the
Newman-Sinclair cameras on their
special base were allocated to the pro-
duction of a black-and-white film which
was shot in a week at the London zoo.
The film is built round the character of
an eminent professor who believes that
an audience cannot appreciate a three-
dimensional film unless it has first
grasped the principles of stereoscopic
transmission. (Any resemblance to the
present writer is wholly coincidental!)
While he becomes more and more
mixed up in tangled phrases and demon-
strations which don't come off, the
camera cuts away to sequences which
clearly show the heightened reality of
the three-dimensional film.

To make possible the production of an
actual film in color, Technicolor Ltd.
of England came forward with the
generous offer of two three-strip cameras
mounted alongside one another on a

base which permitted a variable angle
of convergence. Figures 7 and 8 show
this assembly from different angles.
The simple device of a very slightly
tapered wedge (Fig. 9) enabled the
stereoangle to be adjusted with speed
and accuracy. Parallax measurements
under the traveling microscope showed
that the actual parallaxes between
infinity points on the two camera images
differed by only 3 to 5 ten-thousandths
of an inch from those arrived at by cal-
culation. A universally jointed drive
(Fig. 10) took care of the convergence
angle and enabled one of the cameras to
be swung aside for film inspection. The
normal Technicolor selsyn system was
employed to follow focus on the two
cameras.

The only fundamental disadvantage
of this excellent arrangement was the
necessarily wide separation of the lens
axes; with virtually no gap between
the cameras, this distance was 9.5 in.
This gave, in the theater, a stereoscopic
width magnification (mw) of about
0.25, and a depth magnification (md)
at a mid-position in the theater of about
an equal amount. This suggested that the
film should be composed mainly of long
shots, in which what we call extra-
stereoscopic factors — perspective, mask-
ing, light and shade, and so on — should
as far as possible counteract the miniatur-
izing effect produced by the exaggerated
stereobase.

Our choice of subject fell on the'head-
waters of the River Thames, little known
to Londoners, especially as they appear
in the winter months, when the twin
lenses of the stereo film camera, mounted
on a moving platform, would reveal
the receding planes of the bare tree
branches in all their architectural
beauty. Despite the worst March
weather in 80 years, a short version of
this film was produced in time for the
Festival, and was entitled The Distant
Thames; later a complete film, Royal
River, took its place. In a questionnaire
issued to audiences, this film received a

298

April 1952 Journal of the SMPTE Vol. 58

Fig. 7. Two three-strip cameras mounted together,
as supplied by Technicolor Ltd.

Fig. 8. Two three-strip cameras supplied
by Technicolor Ltd. showing dolly.

Raymond Spottiswoode: Three-Dimensional Films

299

majority of first choices over alfthe other
films in our program.

Stereophonic Sound

Royal River, Around Is Around and the
short introductory film, Now is the Time
to Put on Tour Glasses, were all designed
for stereophonic sound accompaniment.
The equipment previously outlined was
completed by a special re-recording
console, the chief feature of which was a
set of "pan-pots" similar to those de-
signed by the Walt Disney studios for
Fantasia. These are variable distribu-
tion networks which enable a single
input to be "moved around" to any
required output sound track and thus to
any required group of loudspeakers.
Re-recording was carried out in the
Telecinema itself before the Festival
opened, and afterwards during the
night, when the building was closed to
the public. In this way, the precise
effect of the multiple sound tracks could
be judged during mixing.

Reactions to Telecinema

The television and stereo program
was first presented at a world press
show in the Telecinema on April 30,
1951, and it may be of interest to
analyze some of the widespread reactions.
It was to be expected that certain of the
more tradition-bound critics should
regard the stereofilm as just one step
nearer to complete naturalism, and they
viewed with alarm the prospect of highly
three-dimensional film stars should
Hollywood take up this new kind of
movie. For other reasons, the trade
and technical press were not altogether
sympathetic. For them the three-
dimensional film meant a challenge to
long-established entertainment values;
without the blessing of the industry, it
must be regarded as an attack from
outside, like television. The first re-
sponse was therefore to say that it had
all been done before, and wasn't worth
doing again.

The public, however, caring little for

these aesthetic and commercial argu-
ments, showed great enthusiasm for the
new films, and there was in fact never an
empty seat during all the 1,220 per-
formances, despite the normal com-
mercial admission charges. Certain of
the critics, moreover, showed a welcome
perception of new possibilities in film.
The dignified Times declared,

"[In The Distant Thames] the sight and
the imagination were being drawn into
depths and perspectives the screen has
never before possessed the secret of re-
vealing .... The impact of third-dimen-
sional image and sound is far greater
and more fascinating than expectation had
imagined; the spectator who has once
been lent a pair of those magic glasses
and, by taking them, becomes a partici-
pant, will feel like a tiger who has tasted
human blood and will be content with no
other."

And, towards the end of its run, the
Telecinema was described by a promi-
nent trade paper as a gold mine, and
the paper urged the industry to press
ahead with the commercialization of
large-screen television and three-di-
mensional films.

This response was the more gratifying
since our program was extremely modest
in scope and capable of great improve-
ment in its entertainment value. If
these little films, made on a budget of
a few thousand pounds, could attract
such enormous audiences, and cause an
audible thrill to run through the house
at each performance, what would not
be the stimulating effect on the box
office of three-dimensional films made
with all the resources of Hollywood?

It is this thought which prompts the
following tentative comments on the
future of the stereofilm. On the most
restricted scale, we are hopeful that the
Telecinema will remain in existence
under the progressive management of
the British Film Institute as a place
where three-dimensional films and live
television can continue to foreshadow
the entertainment of the future. Those

300

April 1952 Journal of the SMPTE Vol. 58

Fig. 9. Wedge used for adjustment of stereoangle.

Fig. 10. Universally jointed drive for adjustment of convergence angle
and for swinging one camera aside for film inspection.

Raymond Spottiswoode: Three-Dimensional Films

301

responsible for this development in
England intend to push forward with the
production of other films. But what of
the entertainment world, based on
Hollywood? Will three-dimensional
films, which have so long remained just
around the corner — like television not
many years ago — finally step out and
make the flat film as obsolete as the
silent film?

This is obviously not a step to be taken
lightly by the industry's leaders. It is
their responsibility to protect and exploit
the present investment in flat films,
and above all the present roster of film
stars, who might not weather the transi-
tion to a far less flattering form of presen-
tation any more easily than the silent
stars who had to find voices. Hollywood
will have to decide, now or in the future,
whether its box-office revenues are
sufficiently menaced by the attractions of
other kinds of entertainment to justify so
radical and therefore risky a change.
The studios will also have to bear in
mind that television can add a third
dimension more easily than can film, and
that this step forward is likely to be taken
as soon as the novelty of color begins to
wear off.

Use of Glasses

If these arguments are beginning to
recommend a change, the industry is
undoubtedly deterred by a technical
consideration on which I should like to
say a few words, though it demands a
paper in itself. Exhibitors are almost
unanimously against all three-dimen-
sional systems which demand the use of
special viewing glasses, whether of the
permanent or the throwaway kind, and
their objections are entitled to the ut-
most respect. Under the rather special
conditions of the Telecinema, we feel
that this problem was virtually solved.
The glasses, made by the Polaroid Corp.,
had extremely attractive frames resem-
bling beach glasses, and a large filter
area, so that the audience was perfectly
at ease when wearing them. Distribu-

tion and collection, with the aid of
specially partitioned boxes, was accom-
plished by the normal staff of usherettes
in periods of less than two minutes.
Losses were small and there were no
complaints of discomfort.

There is, however, among the public
and the press a tendency to regard any
stereo system requiring glasses as in some
sense "old-fashioned" — this in spite of
the fact that Polaroid was invented only
15 years ago. The following points
about this system are therefore worth
noting. It is the only practical system
in which there is a continuous trans-
formation of the image with movements
of the spectator — in other words, there
are no nonstereoscopic or pseudoscopic
viewing areas. (For modern movies,
the anaglyph system can be disregarded.)
Secondly, the image separation is ex-
tremely efficient; under good commer-
cial conditions, there is a leakage of only
about 0.15% of each image into the
"wrong" eye. No lenticular system yet
constructed approaches this efficiency.
There is no deleterious effect of any kind
on the definition of the image, so that in
adding the third dimension other neces-
sary image qualities are not sacrificed.
Finally, the conversion of theaters is
cheaply and simply carried out, and the
special screen is just as effective with
flat pictures. These points should, I
think, be given greater weight in discus-
sion, especially when it is considered
what a large part of the population
wears glasses, and does not object to
putting on an extra pair on the beach or
when driving a car.

None the less, if two equally perfect
three-dimensional systems were devised,
one requiring glasses and the other not,
there would not be a moment's hesita-
tion in picking the one to use. It is
therefore worth considering some funda-
mental points about these "glass-less"
systems. Firstly, the problem of image
selection at the screen is very much more
difficult than most inventors think;
many able men are working today on

302

April 1952 Journal of the SMPTE Vol. 58

systems which have long ago been
abandoned as profitless, or can be de-
monstrated as having no future. A few
inventors — notably Ives, Kanolt, Noail-
lon and Gabor — have made fundamen-
tal contributions in the motion picture
field during the last 20 years. The best
treatment of this subject is the little-
known group of patent specifications by
Dennis Gabor,4 whose research was
carried out for the British Thomson-
Houston Co. in England.

Two basic and very serious problems
are made clear by this work. First, that
the image-separating screen is of formid-
able complexity, and requires separate
calculation and construction for each
theater, according to the placement of its
seats. Second, the resolving power of
the lenticular structure gives an image
definition much lower than would be
acceptable for feature films, unless a
manufacturing technique is assumed
which is far ahead of what 'can be
accomplished today. Thirdly, if the
audience's heads are not to remain
rigidly fixed, as in the Soviet system
associated with Ivanow, a plurality of
images must be provided for each eye,
so that the eyes pass smoothly from one
viewing zone to another and not into a
position of blurred or pseudoscopic vis-
ion. The stills displayed in shop win-
dows benefit from this plurality of im-
ages, because they give the passerby the
illusion that he is walking past an object
which he can see "in the round." But
the moviegoer is essentially a stationary
person, who is fully satisfied with the
single view of the world which flat films
have long given him. Hence the multi-
ple views required by lenticular systems
(to permit random head movements)
are in a very real sense wasted. When
it is considered that the storage capacity
of 35mm film is already strained to the
limit by the demands of high picture
resolution and almost perfect color re-
production, it will be seen that the re-
quirement of multiplying this capacity

by a factor of 5 or 10 puts an impossible
burden on the manufacturers of film.

Thus we have the contrast between a
virtually perfect system, simple and in-
expensive, which requires glasses; and
systems dispensing with glasses which are
today far from practical attainment, and
which almost certainly would not repay
the huge sums needed to develop them
further. I believe that there is a way out
of this dilemma, and that it is to be found
by harnessing the science of electronics
to solve some of the problems which are
too refractory to be dealt with by optics.

For the Future

In all that has gone before, it has been
assumed that the three-dimensional film
meant the true binocular film, and not
the flat film as projected on a giant
screen, or spread out to the limits of
vision, as in the Cinerama process. It is
certainly true that a wide field of view
enhances the feeling of being "in the
scene," and is thus necessary in any
attempt to give audiences a stronger
sense of participation in the dramas of
the screen. But I do not feel that there
is any adequate substitute for true three-
dimensional presentation; nor do I
think that anyone who has worked ex-
tensively in this field and watched the
reactions of audiences to these "films in
space" would willingly revert to the flat
films of today.

References

1. Norman McLaren (Appendix by
Chester Beachell), "Stereographic ani-
mation— the synthesis of stereoscopic
depth from flat drawings and art
work," Jour. SMPTE, 57: 513-520,
Dec. 1951.

2. J. T. Rule, "The geometry of stereo-
scopic projection," /. Opt. Soc. Am.,
31: 325-334, Apr. 1941.

3. J. A. Norling, "Three-dimensional
motion pictures," Jour. SMPE, 33:
612-634, Dec. 1939; J. A. Norling,
"Progress in three-dimensional pic-
tures," ibid., 37: 516-524, Nov. 1941.

4. British Patents, 541,751 and 541,753.

Raymond Spottiswoode: Three-Dimensional Films

303

The Cash Customers

at the Festival of Britain Telecinema

By NORMAN JENKINS

This informal report describes the reaction of the audience to a most unusual
programme of large-screen television, plus stereoscopic films accompanied
by stereophonic sound.

JL HE SPONSORS, The British Film
Institute, who have also commissioned
the special equipment necessary, claim
that the Telecinema is the first place in
the world in which big-screen television,
three-dimensional pictures and stereo-
phonic sound can be seen on an equal
footing with the established sound
film.

The South Bank Exhibition of the
Festival of Britain houses the Tele-
cinema (officially the Telekinema), the
title of which forces me to apologize
for those of my countrymen responsible.
For everything else all concerned deserve
the highest praise mixed only with the
modicum of criticism I feel it is necessary
to record if only for the sake of
objectivity.

The auditorium of the Telecinema
is long and narrow, quite unlike the
rather broad type of cinema to which
we have been accustomed by building

A contribution which Norman Jenkins,
16 Rozel Rd., Ixindon S. W. 4, England,
has made in response to a request by
Society headquarters.

which took place in the thirties. There
has, of course, been no building of
cinemas after World War II.

The theatre walls cut square into the
proscenium. This looks like nothing
more than a modern picture frame which
appears to be of material, five or six
feet wide, splayed inward and neatly
mitred at the corners. The flat faces
are perforated and the substance looks
rather like Celotex or other proprietary
sound insulation sheeting. The holes
are larger, however, and the material
is continuous rather than in tile form.

The space between the proscenium
and the screen proper is all screen.
That is to say the whole proscenium
opening is projection material of a
specially curved metallic surface. The
central portion is used for the picture
and the peripheral space used for the
projected surround.

The surround consists of a variable
intensity of either white or coloured
light. For television the surround is
fixed value coming from a standard
slide projector. For films the light

304

April 1952 Journal of the SMPTE Vol. 58

comes from a reflex arrangement which
uses the picture itself, a few frames in
advance, to modulate the light from the
lamphouse of one projector.

Please note that the theatre walls
cut into the proscenium and next comes
the comparatively narrow surround, thjen
the picture. Compared with the cross
section of the theatre at that end the
picture size is large.

In the space of a very few months
equipment has been planned and manu-
factured and a cinema designed and
built. In the same space of time the
technique as well as the technics of
large-screen television, stereo-cinema and
stereophonies have been developed.
All of these entirely variable and
theoretical concepts have become actu-
alities now earning money and indeed
playing to completely full houses at
every performance seven days a week.

Of the equipment itself, the design
of the cinema and the technics generally,
I propose to say very little. Raymond
Spottiswoode, who is technical director
of the project, has contributed a paper
describing these in far greater detail
than I obviously could do. [That paper,
with illustrations, also appears in this
issue of the Journal.] My interests are
largely in detailing some of the reactions
of the cash customer.

To know the reactions of the cash
customer is, of course, vital to the
establishment of any kind of commercial
service of stereoscopy and stereophony
in film entertainment and large-scale
television. Whether or not I am best
fitted to make the necessary observations
I cannot say and for that reason the
Society's Board of Editors has permitted
me to make this purely personal con-
tribution, deliberately personal, with
but individual responsibility for the
opinions expressed. I should further
explain that I have no commercial
interests whatever in cinema entertain-
ment, but I have what I believe is a
very wide technical knowledge and
experience developed simultaneously

with a most critical experience of cinema-
going: in short, a film amateur, a pro-
fessional cash customer.

First Visits

The first time I visited the Telecinema
was before the exhibition opened offi-
cially. The attendant gave me a pair
of stereo spectacles and showed me to
a seat while part of a stereoscopic short
subject, The Distant Thames, was being
projected to an extremely small audience.
I must say I regretted this experience,
not only because it thrust me into a
purely private showing but because I was
not a part of a normal audience seeing
a properly staged show.

Nevertheless, I was tremendously
impressed by the cinema itself, the decor,
the proscenium and even the attendants.
I have seen less distinguished appearing
and far less soigne'e programme sellers
at charity shows. The chic clothes
they wore and their air of friendliness
were so exactly right as to baffle de-
scription.

For the part of The Distant Thames
I saw, I had considerable difficulty in
resolving sharp definition. I have since
come to the conclusion that the fault was
probably then in the equipment or its
adjustment. Subsequent viewings have
found the film sharp enough.

At the end and after the lights had
gone up I was literally startled to hear
a number of birds cawing and chirruping
loudly in the course of flight around the
auditorium. I knew, of course, it was
reproduction and that was what I had
come there for, but it was the first time
I had become conscious of the stereo-
phonic sounds. I had not noticed
anything like it during the running of
the film.

For some reason or another I was not
invited to the press show. The national
daily press greeted the programme with
enthusiasm and was followed later by
the technical press in similar terms.
What few reports I saw in the technical
papers were all favourable. I have

Norman Jenkins: Telecinema Audience Reactions

305

avoided reading all the reports hereto-
fore because I did not wish to form any
kind of bias before writing this report.
The first regular showing I did attend
was during the first two weeks. Un-
fortunately, the sound broke down, for
a period entirely, and for the rest only
a single track was used so there was no
stereophony. In my later visits there
was no interruption due to technical
difficulties.

Large-Screen Television

This was the first time I had seen the
large-screen television and although I
had been discussing this with the
cinema manager, Mr. Hazell, who had
come to the South Bank from the Odeon
at Penge where he had been accustomed
to large -screen television, it was some
time before I realized what I had been
looking at.

The large-screen television programme
has, from the opening day, commenced
by showing the entry of the first cash
customers. The performances are sepa-
rate and whilst the house is filling the
projector is running and showing a
picture. This is picked up from the
main entrance foyer, where not unduly
bright lights suffice for the Marconi
camera. Those in the auditorium see
others entering and proceeding to the
staircases.

When I entered the nearly full circle
and saw a picture on the screen the
thought did not register that it was a
televised one. I was in just the mood
of observation, rather than criticism.
The picture was good, large and some-
what soft in tones of grey rather than
black, but apart from that it looked
rather like average to good 16mm. It
was not until the commentator began
speaking that I realized what it was
and not then until he moved his head
and body. When he did this the lines
showed momentarily, sinking back into
the picture on cessation of movement.

The number of lines used by Cinema-
Television Ltd. is the same as that used

for BBC transmissions. I am not a
television user (speaking personally
again, I do not see my money's worth
in the possession and use of a receiver)
and see programmes only occasionally.
My memory of them, both prewar and
recently, had led me to expect large-
screen television to be something far
more crude than this and much less
acceptable. The best pictures I had pre-
viously seen were on a nine-inch tube
and even the lines showed more than I,
a film man, could accept. But this
large-screen television is good by any
standard.

Both on this and on subsequent visits
I found that the audience would laugh
at the least funny incidents. At the
first large-screen television show I saw,
I can well remember the commentator
doing a live and impromptu interview
with a gentleman from Mauritius. The
latter was nervous and had a little trick
of licking his lips. Every time his
tongue came out the audience laughed
and with repetition became hilarious.
I wondered what the man himself
must have thought if he had heard the
laughter. After all, he was only in the
foyer and his audience not more than a
dozen feet from him.

The incident reminds me of the extraor-
dinary feat of sound proofing which
the architect and his technical advisers
have done. Charing Cross railway
bridge (Hungerford Bridge) is at no
great distance from the outside of the
cinema and the noise from a dozen or
so rail tracks is practically continuous.
Inside the cinema there is no detectable
sound.

But to revert to the audience for
television. At other shows I have
noticed that the people coming in and
behaving anything less than completely
phlegmatically will raise a laugh, whilst
the commentator can also raise a laugh
for very little. I take this as part
evidence of a rather specially conditioned
audience. These folk have been waiting
in line for an hour or more to get in.

306

April 1952 Journal of the SMPTE Vol. 58

Before that they have been tramping
round a concrete floored exhibition
getting more and more footsore. Almost
any kind of well padded seat would have
special attractions. But in addition the
character of the exhibition must be added
to the evaluation. This show is like
nothing else they have seen and is an
inspiring and uplifting experience. They
have seen some remarkable things,
presented in a most unusual way.

This audience is in a definable frame
of mind as I see it: everyone is expecting
something unusual. They have, after
all, come to see stereo films and hear
stereo sound, and to see large-screen
television. They expect these things
to be of a quality similar to those other
strange and remarkable things they have
seen. This audience is not likely to be
critical, it is ready and eager to be
amused, a "natural" for a comic. I
think many showmen will know exactly
what I am driving at.

The Picture Programme

Dodging backwards for a moment it
is worth taking a look at the programme.
It has been the same one for all these
months and it is very likely that beyond
minor adjustments it will continue with-
out alteration until the close of the show.

First of all there is Now Is the Time.
There is nothing quite like this in the
average cinemagoer's experience. It
is animated cartoon, but with the
possible exception of short shorts adver-
tising the usual soap or cigarettes,
there is no point of contact. This film
is, of course, stereoscopic and in colour
and has the added unfamiliarity of
synthetic sound — photographed pat-
terns.

The next film is Around Is Around.
This is another film with no point of
contact with usual cinematic experience.
Briefly, the stereo pairs of this film are
produced by the traces of cathode-ray
tubes, synthetically displaced and photo-
graphed in color. The sound tracks of

this film are recorded in multiple and in
depth and width — stereosound.

The next film is really typical of the
average magazine reel. (I mean abso-
lutely no disparagement; I know too
well the compelling circumstances
normally applying to newsreel magazine
production.) A Solid Explanation was
made in black-and-white by the Pathe
Documentary Unit of Associated British-
Pathe Ltd., and is a one-set background
for a commentator somewhat heavy-
handedly explaining that stereoscopy is
a matter of to and from, as he does this
and that, and as to the zoo animals he
describes. The aquaria and outdoor zoo
sequence that follows go to form the
only familiar scenes of comparison.

The Distant Thames is a perfectly
straightforward piece of photography of
the river and as such should form a
point of contact. Unfortunately for this
contention almost all of the film is in
motion. There is a very short sequence
at Windsor Castle where the zoo film
technique is merely duplicated in Techni-
color.

It is, I think, well enough known to
both technicians and film producers
that a camera moving sideways across
a subject, or better still around it in an
arc, will produce an illusion of stereos-
copy, even in a two-dimensional film.
Paramount, I believe, tried (or succeeded,
I just don't know) to patent this for a
series of animated model-cum-cartoon
films they produced in the thirties.
Well, The Distant Thames confuses the
issue by spending an estimated 98% of
its footage in sideways or forward (very
little) movement. I do submit that
this is an unusual experience for cinema-
goers. The sound accompaniment for
The Distant Thames was post-recorded
stereophonically in the Telecinema itself.

The concluding film in the programme
is a cartoon in a most unusual technique.
It consists of a static series of illustrations
to the recitation of "John Gilpin."
None of the illustrations moves. The
pictures are in black-and-white only.

Norman Jenkins: Telecinema Audience Reactions

307

The camera moving across in certain
sequences and quick cutting in others
livens up the fast action demanded by
the story. You know about John
Gilpin, perhaps? He was a citizen of
credit and renown. He went on horse-
back for a pic-nic to the Bell at Ed-
monton.... The sound accompani-
ment is nonstereophonic.

Details of the films and the credits
are familiar to many, for the technical
press has been generous in giving space
to publicize this venture into the un-
known. From those reports I have
seen I must remark how little criticism,
either informed or otherwise, has been
offered: it has mostly been purely
descriptive and noncommittal.

Audience Reaction

The reception given by the audiences
present at those times when I have
visited the cinema has been near enough
the same as far as I can judge. The
first film opens to an appreciative hush,
following polite applause for the tele-
vision commentator. The donning of
the stereo spectacles causes a hum of
excitement and anticipation, although
there must have been many who re-
member the MGM prewar stereo films,
when the more familiar red and green
(and nonreturnable) cardboard viewers
were used. The man behind me last
week made a loud reference to the fact
that this was nothing new.

The effect of depth in Now Is the Time
is instant and clear cut. I do not
suppose there is anyone in the whole
audience, unless one-eyed, who could
not appreciate this. The picture is
brilliant and as well illuminated as
any normal cinema screen. It is only
15 feet wide anyway and there are two
50-ampere arcs kept at a constant level,
one for each of the overlapping pictures.

As the animated drawings moved
forward, apparently out into the audi-
torium, there were always some gasps
of surprise and laughter which, by the
end of the programme when the swans of

the Thames film did the same with their
long necks, had sobered down con-
siderably. Numerous children stretched
out their hands to see if they could touch
the images. Applause after each film,
by the way, was generous to start with
but faded away.

The musical accompaniment of Now
Is the Time is so appropriate, synthetic
as it is, that the novelty and to some
extent eerie effect of the film is enhanced.
This film and Around Is Around are so
much in tune with the spirit of the exhibi-
tion of which the Telecinema is a part
that I for one, when I first saw this
programme, felt a thrill of new ex-
perience.

I wish that Solid Explanation formed
no part of the programme and that The
Distant Thames had perhaps been re-
placed with another, or had been placed
at the commencement of the programme.
I have been left with very mixed feelings:
either this programme should have been,
as it is represented as being, a true means
of comparing stereoscopic films with the
normal cinema, or it should have been
so completely experimental that there
was no point of comparison.

As it is, with the comparison that is
made by Solid Explanation and The
Distant Thames, completely realistic as
is the one and quite beautiful as is the
other, there is but a poor impression to
be gained from the first and no com-
parison of value that can be made with
the second. It is not as though The
Distant Thames can be compared with
any recognizable technique in trave-
logues or documentaries. This film
relies on the natural beauty of the
subject and two technical tricks, one
forward motion and the other stereos-
copy. Of the effect of stereo sound,
please note later comment.

Of the effect on the audiences there
is not much more that can be said
factually. Of the impressions I have
gained from listening to several per-
formances — listening, that is, to com-
ments in the locality of my own seat

308

April 1952 Journal of the SMPTE Vol. 58

and chatter from folk on the way out —
and from discussions with others who
have seen the programme either with me
or at other times, I have gathered the
following. Most folk are sufficiently
impressed to speak of their experience as
"wonderful" and regard the whole thing
rather as a technical miracle. What
they have to say to their friends intrigues
sufficient numbers to keep a queue
outside waiting up to a couple of hours
for a performance — but I have yet
to hear people saying "you must see the
Telecinema" in the tone adopted to
recommending a feature film of the
quality of, say, The Lavender Hill Mob.

Incidentally, if you saw and liked Kind
Hearts and Coronets you will know that
Alec Guinness and the Ealing Studios
comedies can be good. They excel
themselves in The Lavender Hill Mob.
It is not yet on general release and will
certainly be passed on from one to
another as a "wow." But I fear that
the effect of stereoscopic films and
stereosound does not even equal that
of an unusually good feature film.

Audience reaction to the programme
as far as films are concerned has been
dealt with but this report would not be
complete without some reference to the
effect on the audience of the remainder
of the technical effort. Although the
architecture and equipment are not con-
sidered in detail in this paper, I do feel
it necessary to explain the effect of the
proscenium design in relation to the
screen size and the system of projected
surround, and also to mention the effect
of the loud speaker placement and the
effect of stereosound.

Modulated picture surround was tried
out by British Thompson-Houston engi-
neers when the Odeon Cinema in
Leicester Square was first commissioned
and this, the Telecinema, is the second
attempt. I believe that neither experi-
ment is conclusive. As far as the
Telecinema is concerned, it is my
personal opinion that the proportions of
the screen end of the theatre, the screen

size, the proscenium and the surround
are by no means right.

But that depends upon the initial
intention. If all concerned were of the
opinion that stereoptics and stereophony
would make the cinemagoer think he
was in the picture and of it instead of
merely being a privileged spectator, a
dreamer of clear-cut dreams, then in
my opinion such an idea is proved to
have failed. The effect of the present
design is very effectively to present a
window, through which unusually
beautiful effects of depth in recession
may be observed and occasional effects
of depth in protruding procession.

I am sure Mr. Spottiswoode, who is
much more qualified than I am, can
explain why this is so because the
change in effect is so marked. In one
case everything is in the theatre and in
the other everything is so much smaller
and seen beyond the window frame.

I have done much experimental work
myself in projection with the object of
creating the perfect illusion and have
found to my own satisfaction (the
personal aspect of these comments
must not be lost sight of) that the best
effect is obtained by aiming at a picture
suspended in space, a picture ma-
terializing, as it were, in one's own home
or in the theatre where there is no sur-
round noticeable at all. It is a remark-
able fact that the continent of Europe has
not taken so much notice of the necessity
for proscenium design as we have here
in the United Kingdom. In France,
Belgium and Holland I have noticed
that the picture is usually far too big
for both proscenium and theatre in just
the same way as at the Telecinema.

I do not wish to make this an oppor-
tunity for airing my own theories, but
I am not yet convinced one way or the
other of the efficacy or necessity for a
picture surround. In seeing super-
imposed subtitles on foreign films I
have noticed, as others may also, that
white lettering gains contrast where it
appears on areas of picture that are not

Norman Jenkins: Telecinema Audience Reactions

309

necessarily black, or if they are on black
then where the letters are near to areas
of light tone — not necessarily, again, of
completely white areas. That is not
very well expressed, perhaps, but it is
descriptive of a transitory effect and
may strike a chord in those who have
had similar experience.

Assessment of the Stereosound

Of the stereosound in the Telecinema
I must say that from personal experience
it is by no means as successful in illusion
as the stereo picture. The latter is
noticeable from any seat and from any
angle. The depth in sound is effective
from central seats only and best from
the central seats in the circle. In side
seats there is an occasionally noticeable
roving sound.

On the occasion of one visit I had a
downstairs seat on the left-hand gang-
way, about one quarter or less from the
back wall. By dint of knowledge and
conscious effort I could hear sounds
corning from the rear and side, but only
when I decided that I ought to be hear-
ing them in that manner.

Subsequent visits and some thought
given to the troubles I knew the record-
ing people were experiencing have
produced the opinion that it is the
methods used as much as the natural
circumstances which are responsible.
For instance, in The Distant Thames bird
noises are supposed to travel round the
auditorium. They do, undoubtedly, but
background music appears to have no
direction, or else it comes from the
screen end only. To me, there was
auditory confusion. If at any one
moment only one sound direction were
used and a directional sequence were
employed to make that sound travel,
the illusion would succeed whatever the
position of the hearer.

The only check employed was to
question a neighbour on the downstairs
gangway in deliberate non-clue language.
I asked him "What did you think of the

direction of the sound." He had no idea
how, or reason for answering, to please
and said "Why! from the front, I
suppose." Both of us were within ten
feet of one of the nearer rear speakers
but he certainly hadn't noticed anything
coming from it. And I had only by
conscious effort. The point was con-
firmed by a friend a little further to the
rear on the same occasion.

From the same seat I did notice that
sounds following movement in depth
certainly did so with considerable realism
but I question whether it was better
done or results were better than first-
class recording and a normal single-
channel system would produce.

Having made these remarks by way
of criticism and for the record it would
be wrong not to say that in summary
there is here at the Telecinema a con-
crete example of very considerable and
noteworthy achievement. Theoretical
concepts have been brought to reality
in a remarkable space of time and if some
of them point to ways which should not
be followed, then they may equally point
to ways that must. Someone at some
time had to make a start and it is with
national pride that we here see that the
British Film Institute has taken the lead.

The Telecinema is to stay. It has
a site that no one is likely to covet and
it is to be hoped that sufficient money
has been taken at the door during the
exhibition to finance more experiment.
I am certain that when dialogue, for
instance, is recorded and characters
to left and right are heard to speak, as
is usual, one at a time, then the effect
of stereophony will be much more easily
heard and understood than at present
where unplaceable noises have to fight
for their presence with overall back-
ground music.

I am looking forward with the keenest
pleasure to seeing more and more
programmes not only in the Telecinema
(whose title I hope they will change
but I fear they won't) but also in the

310

AprU 1952 Journal of the SMPTE Vol. 58

general run of cinemas. There is noth-
ing in the equipment, either in the large-
screen television or in the stereo systems
to prevent this — there is only expense.

To the average cash customer I
think the Telecinema was a passing
novelty, and a glimpse into the future.
To me, and I am sure to many other
technicians, it has been a tremendously
impressive experience and certainly
a privileged occasion.

The professional reputations of those
concerned have been very considerably
enhanced by the universally favourable
and successful reception given to the
realization of their efforts. I would like
permission to name them all.

Those Responsible

Architect, Dr. Wells Goates

Programme Producer and British Film Institute
Representative for the Festival, John D.
Ralph

Technical Director, Stereofilm Production, Ray-
mond Spottiswoode

Supervisor of Television Production, Malcolm
Baker-Smith

Cinema Manager, A. F. Hazell

Stereophonic Recording, Ken Cameron (By
courtesy of the Grown Film Unit)

Stereo-Projection System, The British Thomp-
son-Houston Co., Ltd.

Stereosound Recorders and Reproducers, The
British Thompson-Houston Co., Ltd., and
His Master's Voice

Large Screen Production Television System,
Cinema-Television Ltd.

Norman Jenkins: Telecinema Audience Reactions

311

Optical-Magnetic Sound
16mm Projector

G. A. del VALLE and F. L. PUTZRATH

Heretofore, the task of recording sound on 16mm film has been a job for the
engineer and the most aggressive amateur. With the advent of successful
striping of 16mm film with magnetic coating, synchronized sound with
picture is now a reality for a greater number of people. The instrument
described in this paper has been designed to record, reproduce and erase
magnetic sound track, as well as to reproduce photographic sound track of
16mm film.

SINCE iron oxide coated tapes
became an accepted medium for sound
recording, the possibility of applying
the same material to 16mm motion
picture work has been very evident.
Work in our laboratories at Camden
for developing and designing equipment
to handle this film has been going on
for several years.

The problems of applying the narrow
strips of magnetic material to 16mm
acetate stock have not been simple,
and this doubtless explains the relatively
late appearance commercially of mag-
netic sound on 16mm film. Reeves
Soundcraft Corp. made satisfactory
film samples in the latter part of 1950.
Magnetic stripe on 16mm and 8mm
was earlier described and demonstrated
by Marvin Camras.1-2

Presented on October 18, 1951, at the
Society's Convention at Hollywood, Calif.,
by O. B. Gun by, for the authors, G. A.
del Valle and F. L. Putzrath, Radio Cor-
poration of America, Engineering Products
Dept., RCA Victor Div., Camden 2, NJ.

The projector that we are about to
describe is basically an RCA 400 Senior
projector (Fig. 1) which has been
modified to accept the component parts
required for recording and reproducing
magnetic sound track without altering,
in any way, the characteristic simplicity
of its threading.

This projector actually performs four
functions: (1) It reproduces photo-
graphic sound track; (2) it erases and
records magnetic sound track; (3) it
reproduces magnetic sound track; (4)
it can be used as a public address
system. Any one of these four functions
can be chosen by simply turning two
knobs, one (Fig. 2) to select the amplifier
operation desired, and one (Fig. 1) to
select the type of sound track to be
played. Recording level is checked
by a glow-lamp indicator which is
located on the upper portion of the
amplifier panel.

For recording and reproducing mag-
netic track, a very small record-play

312

April 1952 Journal of the SMPTE Vol. 58

Fig. 1. RCA-400 magnetic projector showing location of erase
head, and track-control switch.

combination head has been mounted
inside of the sound drum, as shown in
the partially disassembled view, Fig. 3.
The erase head has been mounted
just ahead of the upper sprocket, Fig. 1 .
The location of the record-play head
inside the sound drum offers several
advantages over any other location.
The constancy of film motion is opti-

mum at this point and the distance
from sound to picture can be maintained
exactly the same as that standardized
for photographic and proposed for
magnetic tracks.

Anyone familiar with the behavior of
16mm acetate stock can readily under-
standard the difficulties encountered in
obtaining good physical contact be-

del Valle and Putzrath: Optical-Magnetic Projector

313

Fig. 2. Amplifier function selector switch. Signal level indicator
is shown just below switch knob.

Fig. 3. Sound drum partially removed to show magnetic head assembly

tween the magnetic head and the track
on a film that may be anything but
flat. In the RCA 400 magnetic pro-
jector good physical contact has been
obtained between film and head con-
sistent with low head wear and low
film deformation. The head is mounted
on the free end of a hinged, spring-
loaded arm which also automatically
compensates for head wear.

In order to obtain maximum tracking
of the head (Fig. 3) against the film,
it was found necessary to provide four
distinct adjustments for the record-
play head: azimuth, lateral, pressure
and bearing adjustments. For the
purpose of adjusting the magnetic gap
for azimuth, or perpendicularity in
relation to direction of travel of the
film, the magnetic head has been de-

314

April 1952 Journal of the SMPTE Vol. 58

Fig. 4. Erase head and rewind lever.
Above — On "rewind" position. Below — On "Operate" position.

signed to fit a circular cavity in the
arm, the center of which should lie
in the exact center of the magnetic
gap longitudinally and transversely.
It has been found that this adjustment
can be performed accurately with the
aid of a high-power toolmaker's micro-
scope.

For the purpose of centrally locating
the magnetic sound track in relation to

the width of the pole piece (or lateral
location of film) the same adjustment is
used as was originally provided for
photographic sound track, i.e., axially
moving the roller which guides the
film into the sound drum. The photo-
graphic track is adjusted by means of a
split threaded bushing located at the
anchoring point of the exciter-lamp
bracket.

del Valle and Putzrath: Optical-Magnetic Projector

315

316

AprU 1952 Journal of the SMPTE Vol.58

The amount of pressure between
head and film is controlled by merely
turning a set screw. A nylon insert,
through which this setscrew passes,
locks the screw in position.

The last adjustment of the magnetic
head is to obtain the best possible
alignment of the magnetic head film-
bearing surface in relation to the curva-
ture of the sound-drum periphery. This
is obtained by rotating an eccentric
bushing which is located at the fulcrum
point of the arm supporting the magnetic
head. This adjustment is performed
by rotating the eccentric bushing until
maximum output from a high-frequency
recorded track is obtained.

The erase head has also been mounted
on a hinged arm (Fig. 4) but for a
different purpose. Perhaps one of the
weaknesses of magnetically recorded
sound is the possibility of unintentional
erasure of the recorded signal. In the
RCA 400, rewinding of film is accom-
plished by merely threading the tail
end of the film into the upper reel,
tripping the rewind lever, and starting
the projector.

To render the projector as nearly
foolproof as possible, interference was
intentionally provided between the re-
wind lever and the erase head. In other
words, upon completion of a recording,
the erase head must be moved out of
the way to permit the rewinding of the
film. This automatically removes the
erase head from the film-threading
path. One can erase only when the
erase head is deliberately pushed down
into position and the film threaded
through it. Besides this precaution, it
is also necessary to turn the function-
selector switch in the amplifier to the
Record-Erase position and to insert a
plug in the input jack. If any one of
these last two operations is not per-
formed, the erase head will not be
energized. For efficient erasure of the
recorded signal, it is also essential that
good physical contact be maintained
between the sound track and the

magnetic gap. For this purpose, we
have provided a small plastic shoe
with very light pressure which holds
the film directly against the magnetic
gap of the erase head.

Two guide rollers (Fig. 4) have been
provided ahead of the erase head.
These rollers maintain the film at a
constant angle as it enters the erase
head independent of the reel diameter.

As mentioned before, the projector
reproduces photographic track. This
is accomplished by merely placing the
selector switch (Fig. 1) in the Optical
Sound position and the track selector on
Optical. When the track-selector knob
is moved to this position, a microswitch
completes the circuit for the exciter
lamp and at the same time the magnetic
head is retracted to prevent it from
making contact with the film. These
two precautions not only make it
mandatory to turn the track-selector
knob to obtain exciter-lamp excitation,
but also make it impossible to scratch
the track as it passes over the magnetic
head when reproducing photographic
sound.

The amplifier described in this paper
is somewhat similar to the one used in
the 16mm Senior RCA 400 Projector.
The original amplifier employs one
pentode and two triode voltage ampli-
fiers, one triode phase inverter, and
push-pull pentodes in the output with
an inherently stable feedback circuit.
The input is taken from either a photo-
electric tube or a microphone. The
amplifier delivers 10 watts into a 6-,
15- or 250-ohm load. A volume control
and a tone control (which tilts the
frequency response about an 800-cycle
center frequency) are provided. The
polarizing potential for the phototube
is regulated by means of a glow lamp.
An rf oscillator supplies power to the
exciter lamp.

The new amplifier model (Fig. 5)
meets all the performance requirements
of the standard projector and, in addi-
tion, has all the facilities necessary for

del Valle and Putzrath: Optical-Magnetic Projector

317

the recording, reproducing and erasing
of sound on the magnetic-coated film.
The modification of the amplifier has
been accomplished mainly in two steps:

(1) The gain of the amplifier proper
has been increased by substituting a
pentode for the triode voltage amplifier,
resulting in the following tube com-
plement:

1 — 5879 voltage amplifier,
1 — 6J7 voltage amplifier,
1 — 6SL7GT voltage amplifier, phase

inverter,
3 — 6V6GT push-pull output stage,

rf oscillator,

1 — 5Y3GT rectifier, and

2 — NE-2 voltage regulator, recording

level indicator.

(2) A 9-pole, 4-position switch has
been used to permit the selection of
any one of the previously mentioned
projector functions.

For the reproduction of magnetic
sound (S2 in position 1) the record-
play head is connected to the primary
of the input transformer, the secondary
winding of which is connected to the
grid of the first voltage amplifier. The
turns-ratio of this transformer was
chosen so that the resonance between
the inductance of the record-play head
and the distributed capacity of the
transformer secondary falls slightly be-
yond the useful audio range of the
system. A special load is connected
to the plate circuit of the first voltage
amplifier giving the required low-
frequency compensation. The signal
then goes through the regular amplifier
path, including the pentode and triode
amplifiers with their volume and tone
controls, the phase inverter, and the
push-pull output stage. The amplifier
load is the speaker. To avoid possible
erasure of the magnetic film no plate
power is applied to the oscillator tube.
However, a dummy load maintains a
constant load on the power supply.

For the recording of sound on the
magnetic film (S2 in position 3) the grid

of the input is connected to the micro-
phone. The signal follows the regular
amplifier path except that the tone
control circuit is disconnected, insuring
a "flat" recording characteristic. The
speaker load is disconnected to avoid
accidental acoustic feedback. In its
place, a dummy load is connected
across a 250-ohm output winding. A
suitable voltage divider across this
load feeds the record-play head through
the compensation network (R-39 and
C-20). In order to avoid accidental
erasure the oscillator receives plate
power only while a microphone plug is
inserted. The oscillator load is con-
nected from the primary side of the
transformer and is formed by a series-
parallel combination of the erase head,
the record-play head, C-20, R-39 and
R-40. Thus, the mixing of the audio
and biasing currents for the record-play
head occurs between the head and the
compensating network. The switch sec-
tion which was used to disconnect the
speaker load now completes the circuit
of the recording-level indicator, an
NE-2 tube. The resistive network
associated with this indicator is adjusted
so that the indicator flashes at a signal
level slightly below the overload point
of the film.

Several problems were encountered
in the design, layout, and location of
the amplifier and its associated com-
ponents along the film path. In order
to avoid distortion and high hiss-level,
it is imperative that no residual magnet-
ism be left in the record-play head.
Thus, means must be provided to
decrease the bias current in this head
to a small value before it is entirely
removed. In this model a step-by-step
bias attenuation is accomplished auto-
matically when the amplifier is switched
from the magnetic recording position.
In particular, S2-J (being a shorting-
type switch) temporarily parallels the
oscillator tube and the dummy load
that otherwise takes its place. Thus,
a reduction in B-supply voltage is

318

April 1952 Journal of the SMPTE VoL 58

effected, decreasing the recording-head
current. S2-G and S2-H (also of the
shorting type) temporarily load the
oscillator tank primary and recording
windings with R-44 and the exciter
lamp respectively also effecting a de-
crease in bias current. Similarly S2-C
shunts the record-play head with R-38.
Since these means of reducing the bias
current in the record-play head will
occur successively in some random
sequence, the head will be left in an
essentially de-magnetized state at the
time when the biasing current is com-
pletely removed. If the microphone
jack is removed while S2 is in the
Record-Erase position, the capacities as-
sociated with the oscillator will permit
exponential decay of the amplitude of
oscillation.

Because the amplifier is used for
four different functions, careful layout
of the wiring and switching is required.
Since one head is used for recording
as well as playback, connections be-
tween the amplifier output and input
are required. To gain some measure of
isolation in this circuit the available
intermediate switching contacts are es-
sentially grounded. There is also a
tendency for the oscillator signals to be
electrostatically coupled into the ampli-
fier proper. Careful wiring again limits
these signals to levels sufficiently low to
avoid adverse effects on the amplifier
operation.

The external magnetic fields of the
power transformer, the motor and the
projection lamp are of sufficient magni-
tude to introduce hum into the circuit
components and wiring. The effect of
these hum generators is somewhat
reduced by adjusting their physical
location as far as feasible. Thus, the
motor and power transformer are located
as far away as possible from the record-
play head. The motor can be rotated
axially and the transformer, being
supported in a special mounting, can
be physically adjusted to give minimum
hum interference.

Fig. 6. Hum-canceling coil located on
the main frame of the projector.

It is also possible to minimize the
hum pickup at the disturbed points.
Thus, a well-shielded input transformer
is used and rotated for minimum hum
pickup. All the low-level leads are
tightly twisted and the layout of the
switch wafers made to minimize open
loops in the wiring. Ground loops are
completely eliminated.

The residual hum is eliminated by
the use of a hum-bucking coil (Fig. 6) in
series with the record-play head. It
was not possible to obtain a single
minimum hum-bucking coil adjustment
for the two conditions of projection
lamp "on" and "off." However, a
compromise coil position was found
which gives satisfactory overall per-
formance.

The performance of the electrical
system may be summarized as follows:
Since the output stage of the amplifier
remains essentially unchanged, the power
output rating is identical to that of the
original amplifier. Also, overall charac-

del Valle and Putzrath: Optical-Magnetic Projector

319

1 •

'

x

1 1

-~^

^^ — '

*- -X,

i

s^

\

100

1000 10.000

FREQUENCY. CYCLES PER SECOND

Fig. 7. Overall frequency response of
the system for a signal recorded and
played back.

teristics of the amplifier for optical
playback and public address remain
unchanged. During magnetic record-
ing, it is possible to have 35-db attenua-
tion in the volume control before the
input stage overloads. The amplifier
output networks are adjusted so that
the amplifier distortion will always be
small compared to that of the recorded
signal. Thus, optimum signal-to-noise
ratio is obtained during recording.
With modulated film, the amplifier has
approximately 15-db gain reserve dur-

5 6 7 8 9 10 12

POWER OUTPUT- WATTS

Fig. 8. Distortion curve of the amplifier.

ing playback. Under these conditions,
the signal-to-noise ratio is 50 db with
the tone-control in the flat position.
The overall frequency response (Fig. 7)
of the system, for a signal recorded and
played back on this projector, is flat
within 5 db from 100 to 6000 cycles/sec.

The general specifications of the RCA
400 magnetic projector are given in
Figs. 7 and 8 and in Table I. Figure
9 shows the complete unit.

The versatility of application of the
basic development becomes apparent

Fig. 9. The RCA-400 magnetic projector ready for operation.
320 April 1952 Journal of the SMPTE Vol.58

Table I. Specifications of the RCA 400.

Amplifier power output (400
cycles, 5% distortion) . .

Film speed (24 frames/sec) .

Frequency response (mag-
netic)

Signal-to-noise ratio (mag-
netic)

Input impedance

Record-play head:

Gap length

Pole piece width ....

Inductance (1000 cycles) .
Erase head :

Gap length (double) . .

Pole piece width ....

Inductance (1000 cycles) .
Exciter lamp frequency . .
Cost of applying magnetic

track to film (tentative) .

10 w

7. 2 in.

100-7200
cycles'

50 db
100,000
ohms

0.0005 in.
0.084 in.
3.4mh

0.007 in.
0.125 in.
14.0mh
50,000 cycles

3i cents/ft

Projector

Speaker

Weight . . .
Dimensions :

. . . 45 Ib

26 Ib

Length . .
Width . . .

. . . 201 in.
, . . 9 in.

19f in.
9 in.

Height . . .

. . 15 in.

15f in.

with the enumeration of some of its
potentialities. One particular advantage
of magnetic recording is that the sound
track is independent of the film emulsion
or developing processes. The sound
itself can be added either before or
after the film has been developed for
picture, resulting in great flexibility of
editing. Lip synchronization can be
obtained in a few trials.

Besides the conventional method of
recording sound on a standard 100-mil
track, some variations have been tested
which present decided advantages for
certain applications. For example, half
of the width of an optically recorded
track can be coated with iron oxide
material. Although the output of
both tracks is cut by 50% and their
signal-to-noise ratio decreased, great
practical advantages can be realized.
For instance, the tremendous wealth

of knowledge that has been accumulated
in this country in instructional 16mm
films can be released immediately to our
friends overseas. They in turn can
make their own translations and re-
cordings at a very nominal cost per
print. This is but one of the many
possible applications of a 50/50 track.

Another variation that has been tried
with success is the double-film system.
This idea, it is believed, is the one for
which the amateur has really been wait-
ing. It consists simply of running
through the projector two films, one
being the old double-perforated film
containing the picture, the other being
a clear film carrying the magnetic track.
This system offers all the desirable
characteristics of a standard track,
though a given reel size will accommo-
date only one-half the usual film length.
This, of course, is insignificant consider-
ing the complicated arrangements that
the more aggressive amateur is using
today.

Another system which may have
practical possibilities is that of edge-
coating the old, double-perforated film.
Though this method would be somewhat
simpler to use than the one described
above, it presents three disadvantages:
(1) High amplitude variations are
present in the recorded signal due to the
proximity of the sound track to the
sprocket holes. (2) The limited track
width will result in an only moderate
signal-to-noise ratio. (3) A specially
positioned head would have to be pro-
vided for on the projector.

The cost of producing sound on
16mm film with this multi-use equipment
has been estimated to be about one-
third of the cost of achieving comparable
results photographically. In addition,
film waste due to recording errors is
eliminated. Thus, small commercial
studios, schools and colleges, sales and
advertising organizations, governmental
agencies, training specialists in medical,
military, industrial, religious and law
enforcement fields and especially the

del Valle and Putzrath: Optical-Magnetic Projector

321

amateur movie makers and photo-
graphers will benefit greatly by this
development. To such users this new
recorder-projector means high-quality
sound, greater flexibility, and greater
operating convenience with savings in
time, film and processing costs.

References

1. Marvin Camras, "Magnetic sound for
motion pictures," Jour. SMPE, 48: 14-28,
Jan. 1947.

2. Marvin Camras, "Magnetic sound for
8-mm projection," Jour. SMPE, 49:
348-356, Oct. 1947.

Discussion

Loren L. Ryder: In the interest of stand-
ardization with respect to frequency
characteristics, I wonder if you are in a
position to make available the frequency
characteristics of this recorder-reproducer
at this time. It's quite possible that the
work that you have done may set a pat-
tern which should be followed. Further,
it may be to the advantage of all if at an
early date there is a semblance at least
of standardization so that the product to
be reproduced on your equipment or
handled with other equipment might be
interchangeable. Is that information
available?

O. B. Gunby: Since the authors of this
paper aren't here and detailed information

on the frequency characteristic is not avail-
able in Hollywood at the present time,
your question will have to be referred to
them. However, I have a slide here that
shows the frequency response used in
making this demonstration film.

Lloyd Goldsmith: Again I'm speaking
as chairman of the Sound Committee for
the Society and I'd like to report that
at our Tuesday morning meeting it was
brought out that our Subcommittee on
Magnetic Recording is attempting to
standardize, or at least act as a clearing
house, for information on the frequency
response and the pre- and postequalization
in these magnetic recorder-reproducer
projectors for the benefit of all of the
manufacturers. Accordingly, Glenn Dim-
mick has already circulated this informa-
tion to the Subcommittee with respect to
the RCA projector and I will be very glad
to make it available to Mr. Ryder. Also,
Ampro has indicated their division of pre-
and postequalization, and I'm sure that
before very long there will be agreement
on recording-reproducing characteristics
to allow complete interchange of magnetic
film made on this type of projector.

Mr. Gunby: The slide is now ready for
presentation. You will notice that it
gives only the overall frequency response.
It probably doesn't completely answer
Mr. Ryder's question, but the information
referred to by Mr. Goldsmith and which
can be obtained from the Subcommittee
on Magnetic Recording will likely provide
the additional data requested.

322

April 1952 Journal of the SMPTE Vol. 58

Twin-Drum Film-Drive Filter System
for Magnetic Recorder-Reproducer

By CARL E. HITTLE

Use of two drums in tight-loop type of film-drive filter system solves the
problem of film support in magnetic recorder-reproducer utilizing two sepa-
rate magnetic head assemblies. Performance of filter system is analyzed.

M,

.ANY VALUABLE contributions have
been made to the art of sound recording
by the design of film-drive mechanisms
and a wealth of engineering principles
covering these endeavors can be found
in literature.1 It might be considered,
however, that many of the film-drive
mechanisms described in the references
were designed for specific applications
using photographic film as the recording
medium, and were not particularly
suited for the magnetic type of medium.
The theory of recording with the latter
has established certain requirements
for the film-drive mechanisms. These
were met only with some degree of
compromise with many of the other
photographic-type drives employed in
what might be referred to as the interim
period for the acceptance of magnetic
recording by the motion picture industry.
Magnetic recording, having proven to
be a useful tool, has dictated the need
for a more comprehensive design of the
overall equipment as well as of the

Presented on October 18, 1951, at the
Society's Convention at Hollywood, by
Carl E. Hittle, Radio Corporation of
America, Engineering Products Dept.,
RCA Victor Div., 1560 N. Vine St., Holly-
wood 28, Calif.

system components. The purpose of
this paper therefore is to describe a
film-drive mechanism especially designed
for magnetic recording and one which
makes use of the many advantages
attributed to the magnetic medium.

Features of the film-drive mechanism
especially designed for magnetic re-
cording may be more fully appreciated
when illustrated against a background
of those of the basically photographic
types which were converted for use of
magnetic film.

The conversion of a photographic-
type recorder to magnetic is shown
schematically in Fig. 1. In this in-
stance, the photographic-type sound
drum was foreshortened so that the por-
tion of the magnetic film from the sound
track location to the nearest outside
edge of the magnetic film would extend
beyond the drum much in the same
fashion as would be required for con-
verting to photographic sound re-
producing. The single magnetic head
was mounted so that the recording gap
portion of the head would contact the
coated surface of the magnetic film at
the required location. The film-drive
filter system of this recorder was a tight-
loop system utilizing a magnetic drive

April 1952 Journal of the SMPTE Vol. 58

323

SPROCKET

ROLLER

Fig. 1. Recorder having single
magnetic head in drum.

for the sound drum as described by
Collins.1

As may be observed from Fig. 1
since the coated surface of the film was
toward the inside of the film loop between
sprocket and drum, space limitations
permitted the mounting of only the one
magnetic head shown at the drum. In
this particular adaptation the one head
was used for recording the sound track
and later for reproducing with no
facilities available for monitoring the
recorded track at the time of recording.

When used for photographic recording
prior to the conversion, this equipment
at least equalled any other commercially
available equipment in providing flutter-
free film motion. The quality of film
motion, when used for magnetic re-
cording, was for practical purposes the
equivalent of that when used as a
photographic recorder. However, ex-
perience with the equipment as a
magnetic recorder indicated the need
of a more desirable location for the
magnetic head since head wear tended

ROLLER

Fig. 2. Recorder having two retractable
magnetic heads external from drum.

to be uneven due to difference in pressure
between film and head across the width
of the head. The partial view in Fig. 1
shows in exaggerated form the position
the film tends to assume with respect to
the head and drum. Necessarily, the
head at the magnetic-gap section must
protrude slightly beyond the film sup-
porting surface of the drum to provide
the desired contact pressure between
film and head. Unit area pressure tends
to be relatively high at the edge of the
head adjacent to the drum and to
diminish as the opposite edge is ap-
proached. This tendency may be re-
duced to some extent initially by a
slight rotation of the head.

A second illustration of a basically
photographic type of recorder modified
for use of magnetic film is shown
schematically in Fig. 2. The film-
pulled drum type, tight-loop filter
system with damping applied by means
of a dashpot connected to one sprung
roller arm was retained since it too
provided excellent film motion. In this
instance (as well as in the remainder of
the systems to be described) the film
threading was such that the coated
surface faced to the outside of the film
loop between sprocket and drum. With
only a slight change in the film path,

324

April 1952 Journal of the SMPTE VoL 58

RECORD
HEAD

MONITOR
HEAD

DRU

SPROCKET

space was made available for mounting
both a record head and a monitor head.
Since film motion tends to be of best
quality at the sound drum or imme-
diately following the drum with ref-
erence to direction of film travel, the
record head was mounted adjacent to
the drum on the film take-up side with
the monitor head mounted as closely
following as facilities would permit.
The two head mountings were so de-
signed that by actuation of detent pins
the magnetic heads could be rotatably
retracted from film contact position to
eliminate possibility of abrasive damage
to the film emulsion from head contact
when using the equipment for photo-
graphic recording. This permitted al-
most immediate change-over from one
recording method to the other.

One application of this equipment
has been in television studios where
double-film systems are used for making
kinescope recordings of television pro-
grams. Operational economies are
realizable due to the versatility of
operation of this type of equipment.
The purchase and use of such equipment

OIL
DASH POT

Fig. 3. Twin-drum
recorder with drums
in horizontal plane,
single film sprocket.

becomes a money saving investment for
television studios and others concerned
with high-quality sound recording since
either photographic or magnetic medium
may be used for the original recording
with magnetic recording available for
making protection "takes" when two of
these units are available.

Sound quality attainable with such
equipment is at least as good with
magnetic film as with photographic
film. Reproduction using the record
head is superior to that using the monitor
head principally because of the better
film motion obtainable at the record-
head location. For this reason, the
equipment is provided with switching
facilities to permit the record head to
function as a "playback" or reproducing
head when best quality reproduction is
desired.

Representative of the different ap-
proach used in designing a film-drive
mechanism specifically for magnetic
recording and reproducing is the mec-
hanism shown schematically in Fig. 3.
Since freedom of design permitted, an
addition was made to the basic film-

C. £. Hittle: Film-Drive Filter

325

DRUM

ROLLER

SPROCKET

ROLLER

OIL DASH POT

DRUM

Fig. 4. Twin-drum recorder with drums in vertical plane, one film sprocket.

drive mechanism described in the last
illustration given. This addition, made
to provide film motion at the monitor
head as nearly equal to that at the
record head as possible, was a second
impedance drum. The geometry of
the film path, considering for the
moment just the film sprocket and the
two impedance drums, is basically an
equilateral triangle in shape with the
sprocket at the apex.

A sprung roller added to each of the
two equal sides of the basic triangle
serves two purposes. Each serves to
alter the film path in such a manner
as to increase the film wrap about its
adjacent drum to the degree desired
for film-pulled drum operation. The
two sprung roller arm assemblies with
associated tie spring are also essential
elements in the twin-drum film-drive
filter system. The relatively light
sprung roller arms tend to absorb any
disturbances introduced into the film
motion through the film-drive mecha-
nism. This results from the fact that

the film is held in tight contact with the
rotating drum surface and the much
greater inertia effects of the rotating
flywheel mass of the impedance drums
make these elements relatively insensi-
tive to such film motion disturbances.
Damping of any tendencies of the
sprung roller arms and the drums to
oscillate is provided by the oil-type
dashpot which is linked mechanically
to one of the sprung arms. The oil
used in the dashpot is a selected grade of
temperature-stable silicone.

Film tensioning is furnished through
the force exerted by the tie or center
spring connected to the two sprung or
tensioning arms when the film is threaded
properly in the film path shown.
Ground springs shown in the illustration
are used principally for mechanical
purposes and have little effect as func-
tional members of the filter system.
Without the ground springs, the sprung
roller arms tend to rotate to one or the
other extremity of their arc of travel,
depending on direction of film motion.

326

April 1952 Journal of the SMPTE Vol. 58

Fig. 5. Twin-drum recorder with drums in horizontal plane, two film sprockets.

Performance has justified the de-
parture from the usual photographic-
type film path to that which has just
been described utilizing the two im-
pedance drums. Flutter content of a
recording reproduced by the record head
is less than 0.1% rms total with less
than 0.05% rms being 96-cycle flutter.2
Flutter content using the monitor head
for reproducing is less than 0.15% rms
total with less than 0.05% rms being
96-cycle flutter.

It might be well at this point to
mention two features relative to the use
of the twin-drum assemblies. From
outside our organization has come the
suggestion that the inertia effect of the
two impedance drums would have to
differ by appreciable amounts to permit
satisfactory performance free of beat-
frequency disturbances from the two
drums having equal size and weight.
Our experience indicates through the
consistent low flutter performance ob-
tained that drum assemblies of equal
inertia effect are satisfactory in the
filter system which we are now using.

The second feature relates to the position
of the magnetic head with respect to
drum to provide optimum film motion
performance. Tests were made to
determine if head position had the same
relative effect on observed flutter as
compared with that of single-drum
magnetic recorders. These tests showed
that head location with respect to drum
along the path of the tensioned film
between the two drums had no bearing
on the quality of film motion. Since a
latitude of choice of head location
existed, locations for the record and
monitor heads were chosen which per-
mitted maximum useful head life to be
obtained with the original factory
head setting. These locations also pro-
vided protection for the heads at their
most critical section, the film contact
area at the magnetic gap, due to the
close proximity of this area to the drum.
The mechanism shown in Fig. 4 is
the basic mechanism of Fig. 3 rotated
90° with the dashpot appropriately
relocated and is intended primarily
for standard relay cabinet rack mounting

C. E. Hittle: Film-Drive Filter

327

as part of a permanent magnetic re-
cording system in studios. Equipments
produced to date using the physical
arrangement of components as shown
have been of the triple-track type, as
described by Singer and Pettus.3 (It
is equally adaptable for single-track
magnetic equipments.) Quality of film
motion even though six magnetic heads
are in contact with the film simul-
taneously is equally as good as with the
single-track equipment represented by
Fig. 3.

Further illustration of the way in
which the basic design of the twin-drum
film-drive filter system has been adapted
to meet varying space needs is shown
in Fig. 5. In this instance, design
requirements of compactness without
sacrifice of quality of performance had
to be met. Use of the second sprocket
facilitated the attainment of both size
and weight reduction. As may be seen
by comparing Fig. 5 and Fig. 3, the
filter system is essentially the same in
the two illustrations. Film motion using
the mechanism shown in Fig. 5 has
proven to be equally as good as that
described for the mechanisms of Figs.
3 and 4. Equipment utilizing the film-
drive mechanism shown in Fig. 5 is
described in the paper by Singer and
Ward immediately following in this
Journal.

The drum shaft assembly, sprung or
tension roller assembly, and sprocket
assembly of the twin-drum film-drive
filter system remain essentially un-
changed throughout the variations of the
system previously described. The same
basic filter system is used on 35mm,
17jmm and 16mm equipments.

Field performance has furnished proof
of the soundness of design of the twin-
drum film-drive filter system, adequate
solution of the problem of film support
in the critical region of the magnetic
head, and the advantageous choice of
magnetic head locations.

References

1. M. E. Collins, "A deluxe film recording
machine," Jour. SMPE, 48: 148-156,
Feb. 1947; "Lightweight recorders for
35- and 16-mm film," ibid., 49: 415-424,
Nov. 1947.

2. Proposed American Standard, Z57.1 /68,
Method for Determining Flutter Con-
tent of Sound Recorders and Repro-
ducers, American Standards Assn., 70 E.
45 St., New York City.

3. Kurt Singer and J. L. Pettus, "A build-
ing-block approach to magnetic re-
cording equipment design," scheduled
for publication soon in the Journal.

Discussion

D. J. White: The thing that I found
most interesting about this discussion of the
dual drum filter mechanism was the ref-
erence that the speaker made to the
"outside" sources who had called the
attention of the industry to the fact that
different inertias and different masses in
the two inertia wheels make a definite
and distinct difference in the characteris-
tics of motion. As the originators of the
dual flywheel motion path, we at Magna-
gram feel that we are somewhat qualified
to make the statement that our experience
has proved there is definitely a difference
when the proper ratio between the two
inertia wheels is achieved.

The first machine, which we introduced
in May of 1948 to the Society's 63rd
Semiannual Convention, employed dual
inertia wheels of the character just de-
scribed by the Speaker. Since that time,
we've conducted numerous experiments
and we have determined conclusively that
we can reduce overall low-frequency modu-
lation by making calculated changes in
the mass and inertia of the two flywheels.
For your information we refer to this as
"Synkinetic" motion.

Carl E. Hittle: As mentioned during the
reading of the paper, we can speak only
from the experience which we have had
with our type of system and I can only
reiterate the fact that, in the relatively few
years that we have been playing with this
type of equipment, we have not en-
countered any difficulty due to the fact
that our flywheels are of equal mass.

328

April 1952 Journal of the SMPTE Vol. 58

A Technical Solution of Magnetic
Recording Cost Reduction

By KURT SINGER and H. CONNELL WARD

This new portable magnetic recording channel, designed primarily for 17 i
mm film provides high-quality operation and all of the needed facilities for
production recording. By operating at 45 fpm a considerable economy in
film cost is realized and the size and weight of the recorder are reduced. The
recorder is also adaptable for 16mm or 35mm film. A new amplifier system
utilizing miniature tubes and small components is provided as part of the
equipment.

o

'N MAY 18, 1948, the first studio-type
magnetic recording channel was de-
scribed at this Society's Convention at
Santa Monica, Calif.1; and, on October
27 of the same year the first portable
magnetic recording equipment was pre-
sented at the Society's Convention at
Washington, B.C.2 Both of them used
perforated 35mm magnetic film. Their
demonstrations gave convincing proof
that synchronous magnetic recording was
a useful tool in the production of sound
motion pictures.

These equipments were produced by
the addition of the magnetic recording
elements to existing photographic record-
ing facilities, and, in some cases, the
modified equipments were capable of

Presented on October 18, 1951, at the
Society's Convention at Hollywood, Calif.,
by Kurt Singer and H. Connell Ward,
Radio Corporation of America, RCA
Victor Div., Engineering Products Dept.,
1560 N. Vine St., Hollywood 28, Calif.

using both photographic and magnetic
film as the recording media. With a
number of modified channels in service,
information was obtained concerning the
features and facilities that should be in-
cluded in an equipment designed for
magnetic recording only. The first of
these was the PM-62 portable system.
Shortly afterwards, the PM-63 and PM-
66 rack-type arrangements were made for
triple-track and single-track recording.
These are described in detail in a paper
entitled, "A Building-Block Approach to
Magnetic Recording Equipment De-
sign" by Kurt Singer and J. L. Pettus (to
be published soon in the Journal).

Despite the accomplishments of these
channels, many have felt that the antici-
pated saving was not sufficient to warrant
converting to magnetic recording.
There has also been the need for smaller,
lighter-weight magnetic recording facili-
ties capable of recording continuously a
30-min program for television applica-
tions. By previous standards, it would

April 1952 Journal of the SMPTE Vol. 58

329

RECORDING OR

REPRODUCING
HEADS IN LINE

-OI50-*

0.050-k-*|« 0.189

— 0.339 ±0004 .

0689

I6MM

Fig. 1. Proposed magnetic film track standards for 35mm, 17£mm and 16mm.

be necessary for the recorder to use 3000-
ft film reels for television and 1000-ft
reels for location purposes. From addi-
tional investigations grew the idea of
split 35mm film operating at "split"
35mm speed or 45 fpm for all original
recording. The use of split 35mm film
had already gained acceptance as evi-
denced by the 17£mm magnetic track
location proposed by the Motion Picture
Research Council3 and later appearing as
a proposed American standard.4 Tests
showed that 45 fpm provided a reason-
able reserve in frequency range over that
normally used in sound motion picture
production and the ratio between this
new speed and the standard speed for
release film was made very simple. By
splitting a 1000-ft roll of 35mm film
which has a recording time of approxi-
mately 11 min at 90 fpm into two rolls
and cutting it into 500-ft lengths, we have
44 min of recording time at 45 fpm.

This automatically gives a film cost sav-
ing of 75% and approximately the same
amount of saving in film storage space.
In addition, the reduction in the initial
film capacity represented a considerable
reduction of recorder weight and volume.

As companion unit for this new mag-
netic recorder, a new amplifier system
capable of being operated from either of
two types of power supplies has been de-
signed.

The above reviews briefly the steps
leading up to the RCA PM-64 Portable
Magnetic Recording Equipment which
will be described in detail below.

Equipment Features and Adaptations

Many combinations of film width and
speeds, motor systems and power sup-
plies are available. For 35mm and 17^
mm width, recording speeds of 90 fpm
or 45 fpm are available. For 16mm, a
speed of 36 fpm is employed. Any of the

330

April 1952 Journal of the SMPTE Vol. 58

Fig. 2. PR-42 Portable Magnetic

Recorder, 17£mm equipped with

500-ft reels.

commonly used motor systems may be
used with any combination of film width
and speed.

This equipment adheres to the pro-
posed American standard for track loca-
tions (Fig. 1).

The associated audio amplifiers meet
the established studio requirements for
high-grade recording systems. High-
level mixing is provided for two micro-
phones with four steps of dialogue equali-
zation. The mixer may use either direct
or magnetic monitoring and facilities are
available for him to communicate with
the recordist or boom man. The system
may be operated from a-c mains or stor-
age batteries.

Recorder Structure

The recorder as seen in Fig. 2 has
three basic parts: the front cover, the
rear cover and the center section. A
large plastic window in the front cover
permits film observations while 500-ft
reels are in use and the cover closed.
By loosening two thumbscrews, the front
cover can be removed to permit the use of
film reels larger than 500 ft. The rear
cover is similar to the front cover. Lo-

Fig. 3. PR-42 Portable Magnetic

Recorder, 17?,mm equipped with

1 500-ft reels, covers removed.

cated inside its back is a mounting for
additional belts and sound -absorptive
material for noise control. The center
section of the case is a cast magnesium
box having an open end. The closed or
front end forms the panel for mounting
the drive, take-up and feed mechanisms,
footage counter and electrical compo-
nents associated with the bias oscillator
and playback amplifier. At either end
are panels containing additional compo-
nents for external electrical connections
and controls (Fig. 3). This arrange-
ment facilitates the adaptability of the
recorder to different motor systems, con-
trol circuits and external electrical con-
nections and allows the recorder to be
operated completely enclosed.

The recorder may be modified for
operation with 1 500-ft instead of 500-ft
reels in about 8 min by removing the
three thumbscrews which mount the
take-up and feed mechanisms and by re-
mounting that mechanism in its new lo-
cation (see Fig. 4). Except for belts,
all items integral to this change are part
of the recorder.

For transportation, two flush-type
handles are attached to the ends of the
center section.

Singer and Ward: Magnetic Recording Cost Reduction

331

Fig. 4. PR-42 Portable Magnetic Recorder, rear cover opened.

Fig. 5. PR-42 Film-
Drive Mechanism,
front view.

332

April 1952 Journal of the SMPTE Vol. 58

Drive Mechanism

The basic element of the drive mech-
anism is a mounting plate that contains
the drive motor, mechanical filter system,
magnetic heads, motor controls, guide
rollers and other items integral to its
operation. The mounting plate, as 'seen
in Fig. 5, is a cast magnesium panel cap-
able of being mounted for either portable
or studio applications. The drive mech-
anism, using magnesium for all cast parts,
has an overall weight of approximately
26 Ib.

The drive motor is one of a new series
of motors especially designed for mag-
netic recording equipment. The motor,
which is mounted to the rear of the mech-
anism plate, contains a precision gear re-
duction unit to allow the drive sprocket
to be coupled directly to the output
shaft. Also coupled to the shaft, but ex-
tending to the rear, is an overrunning
clutch assembly with driving media for
the holdback sprocket, the take-up mech-
anism and the footage counter. Inter-
changeable gear reduction units provide
for ratios varying between 10:1 and
125:9, depending upon the film speeds
required and the type of motors used.

The film drive consists of two 32-tooth
sprockets in a symmetrical path which
includes two tension or filter rollers and
two drum assemblies. Located in the
film path, between the drums, are the
magnetic record and monitor head as-
semblies. The film drive is described in
detail in the paper "Twin-Drum Film-
Drive Filter System for Magnetic Re-
corder-Reproducer" by Carl E. Hittle,
immediately preceding this paper in this
Journal.

A combination drive and holdback
sprocket5 was designed to replace the
standard sprocket because it best fulfills
the needs of the tight-loop system and will
allow the recorder to be operated in
either direction without sprocket
changes. The sprocket pitch and tooth
shape are dimensioned so that the face of
only one tooth is driving or holding back
the film at any given time. The sprocket

will accommodate films with 0.6%
shrinkage and 0.02% expansion. The
area of contact between film and sprocket
has been reduced from 90 ° in the earlier
sprocket to 60° maximum for the new
sprocket.

The tension or filter rollers are
mounted on shafts attached to the ends
of the roller arms. These arms are in
turn pivoted from a fixed point. The
arms are tied together by a spring and are
separately grounded by additional
springs. Figure 6 shows the mechanical
filter system schematic. An oil-type
dashpot is attached to the other end of
the right roller arm by suitable linkages.
The damping medium is a selected grade
of silicone oil. The entire system is near
critically damped with a resonant fre-
quency of 1| cycle/sec.

The drum assemblies, mounted in cast
tubular housings identical with those of
the holdback sprocket, have dynamically
and statically balanced flywheels serving
as inertia elements. Both drums are
film driven.

The magnetic head6 may be used for
either single- or multiple-track appli-
cations. For single-track applications,
there is a two-piece holder having a ball-
and-socket type of anchorage. Through
the mounting flange screws, the anchor-
age and four opposing setscrews bearing
on the head, it is possible to give longi-
tudinal, lateral and transverse adjust-
ments7 to the head as required. For
16mm and 35mm operations, a hard-
ened stainless-steel shoe is placed in
alignment with the magnetic heads,
thereby maintaining an even plane of
film across the magnetic head and an
even pressure at the gap, thus minimizing
wear. For 17£mm operation, the mag-
netic head is only 5£ mils from the
center line of the film. Instead of using
a shoe, the drums are tapered and flanged
allowing the film to be guided from the
perforated edge, thus assuring track lo-
cation, uniform contact and minimum
wear.

The motor-control-switch mechanism

Singer and Ward: Magnetic Recording Cost Reduction

333

REC"90 HEAD

Fig. 6. Mechanical Filter Schematic. OIL DASH POT

Fig. 7. PR-42 Film-Drive Mechanism with motor and flywheels removed.
334 April 1952 Journal of the SMPTE Vol. 58

is used for switching and for adjustments
of the ground springs connected to the
filter roller arms. Referring to Fig. 7, it
may be seen that the filter ground springs
have been arranged so that they are auto-
matically tensioned through a bell-crank
arrangement for the respective running di-
rections of the mechanism. The damped
filter roller ground spring is connected to
the bell crank by an adjustable anchorage
which permits quick adjustment of the
filter roller balance. The undamped fil-
ter roller spring likewise has an adjustable
anchorage, but is used for manufacturing
convenience only. The center tie spring
is permanently attached to the filter roller
arms. The mechanical linkage, on which
the undamped adjustable anchorage is
mounted, also carries fingers that actuate
microswitches that in turn control re-
motely mounted relays in the motor cir-
cuit. The motor-control switch consists
of a bar moving vertically in an elongated
slot. It is positively locked in the "off"
position and released by pushing in on the
bar to either of the operating positions.

The sprocket shoes are basically safety
devices, since the film has a mechanically
predetermined amount of wrap on each
sprocket. Both shoes are held for normal
clearance from the sprocket by a positive
locking detent arrangement and spring
tension. The drive sprocket shoe (Fig.
7) is mechanically linked with the filter
rollers in such a fashion that when it is
opened, the undamped filter roller is
locked in its rest or innermost position
and the damped filter roller is displaced
from its rest position to a predetermined
location. The film may then be
threaded through the film-drive system in
tight fashion. Upon releasing the drive-
sprocket shoe, the filter rollers are freed
to their normal positions and the film
loop is thus automatically formed.

High permeability shielding around
the motor and the magnetic heads is used
to prevent hum pickup. Ball bearings
have been used at all points except on the
motor shafts.

Take-up and Feed Mechanisms

It has long been the desire of all asso-
ciated with the motion picture industry
to acquire an efficient mechanical
take-up and feed mechanism. As a
result, the take-up and feed mechanisms
were designed with constant torque to
guard against their failure during the
operating cycle of a roll of film. Tests
showed that with constant torque the
film tension between the film reels and
the take-up or holdback sprockets varied
by ratios of 7:1 or 1:7, respectively,
throughout the length of a 1 500-ft roll of
film wound on a 2-in. core. By varying
the applied torque at the friction-clutch
assembly in the new take-up and feed
mechanisms (Fig. 8), it is possible to have
near-constant film tensioning between
the film reels and sprockets. Tests using
a 1 500-ft roll of film on a 2-in. core indi-
cated that the tension varied only 2
oz throughout the length of the film roll.
The take-up and feed mechanisms are
composed of identical subassemblies;
therefore the mechanical construction
will be discussed in terms of the take-up.
The take-up is driven through an over-
running clutch mechanism located at the
friction-clutch assembly, a precision rub-
ber cog belt and a clutch mechanism
coupled to the rear extension of the out-
put shaft. For varying the amount of
pressure needed during operation, suit-
able mechanical linkages are used to con-
nect the adjustable compression spring to
an extended portion of the sensing roller
arm. This connection allows the spring
compression to vary as required by the
film pull on the sensing roller.

The decreasing weight of the film reel
causes a small amount of overslipping at
the friction-clutch assembly, since the
pressure of the compression spring at a
given instant is not sufficient to assure
take-up of the film. At this given in-
stant, the sensing roller relaxes its posi-
tion in the slot to equal the slacking of
the tension in the film loop from sprocket
to take-up reel.

Singer and Ward: Magnetic Recording Cost Reduction

335

Fig. 8. Take-up Assembly,
parts arrangement.

Fig. 9. MI- 102 7 8 Mixer
Amplifier Case.

Fig. 10. MI-10278 Mixer Amplifier
Case on Pedestal.

By this action the sensing roller,
through its linkages to the compression
spring, causes the spring to be com-
pressed, thus giving the needed addi-
tional pressure at the friction-clutch
assembly. These minute impulses are
continuous throughout the time required
to transport the film through the re-
corder.

Fig. 11. MI-10278 Mixer Amplifier
Case (internal arrangement) .

Also attached to the extended portion
of the sensing roller arm is a spring with
an adjustable anchorage and a dashpot.
The dashpot softens the shock of the
roller arm when the roller moves toward
either of the extremities of the slot at
starting or stopping, or in case of irregu-
larities in the film loop. The dashpot
also keeps the sensing roller from hunting

336

April 1952 Journal of the SMPTE Vol. 58

Fig. 12. Transmission Block Schematic.

during the normal operation of the re-
corder.

As a result of this arrangement, the
film tensions between feed reel, holdback
sprocket, take-up reel and drive sprocket
are near constant at all times. Also,
through the action of the overrunning
clutch mechanisms, the operation of
take-up and feed mechanism may be re-
versed according to the rotational action
of the drive motor.

All recording amplifiers and recording
control circuits are contained in the
mixer amplifier case shown in Fig. 9, in
the closed position ready for transporta-
tion. For operation, the top cover is
removed and a writing surface attached
to a hinge, normally located inside the
cover, is folded to the right side of the
case. This permits the mixer to keep
his log and facilitates his written entries.
On the left side is a hinged bracket which
furnishes support for a sound-powered
telephone set. The mixer case, ready
for operation and on a collapsible pedes-
tal, is shown in Fig. 10.

The internal arrangement of ampli-
fiers and control panel is shown in Fig.

1 1 . Plug-in amplifiers are used through-
out, for maximum serviceability.

The electrical transmission circuit
conforms to the block diagram shown in
Fig. 12. The signal from two micro-
phones is amplified by separate micro-
phone amplifiers whose output is con-
trolled by individual mixer pots, then
combined and further amplified by
means of a voltage amplifier. The signal
is then conducted to the power amplifier,
and hence directed to the recorder. The
case containing the above-described
amplifiers also contains a monitor ampli-
fier whose input is normally bridged
across the output of the recording ampli-
fier. At the recorder the signal is com-
bined with the bias current and fed to
the recording head. In the recorder it-
self are contained a bias oscillator and a
playback amplifier. The output from
this playback amplifier is brought back
to the amplifier case and is available to
the mixer by pressing a button so it can
be compared with the output from the
direct monitor. All amplifiers located
in the mixer amplifier case use the same
tube type, namely, 12AY7. Only three

Singer and Ward: Magnetic Recording Cost Reduction

337

different tube types are used in the entire
channel, each chosen to insure optimum
operating efficiency and freedom from
microphonics and tube noise. For in-
stance, the two lowest-level stages in the
playback amplifier employ RCA 5879
tubes which are selected to be equal in
noise to RCA-1620's. In the bias
oscillator the tube is a 12AU7 which has
worked out very satisfactorily in previous
similar applications.

Let us consider briefly the circuits of
each amplifier. The microphone ampli-
fiers each use one single 12AY7 tube.
The two triode sections of these tubes are
connected in cascade with sufficient feed-
back from input to output to keep dis-
tortion to a minimum and to stabilize the
gain, so that changes in tube character-
istics and component tolerances have
only a negligible influence on the gain or
frequency characteristic. The micro-
phone amplifier input transformers pro-
vide facilities for working from 30-, 50-,
150- or 250-ohm microphones. These
transformers have been especially de-
signed so that one primary connection
permits the interchangeable use of 30- or
50-ohm microphones, whereas another
connection accommodates 150- or 250-
ohm microphones. The change in fre-
quency characteristic when working
from these various impedances is negli-
gible. Means are also incorporated in
the microphone amplifiers to reduce the
gain by 10 db if high-level pickup condi-
tions should make this necessary. A
toggle switch introduces equalization so
that either velocity or pressure micro-
phones can be used satisfactorily.

The voltage amplifier uses one 12AY7
tube connected in similar fashion as
already described in the microphone
amplifier. It also contains a switch for
10-db gain reduction which is normally
in the circuit so that a reserve gain of 10
db is available should it be required. In
addition to the amplifier circuit itself,
there are also provided an 8000-cycle
low-pass filter and a low-frequency boost
circuit. The voltage amplifier chassis

also contains a 400-cycle RC oscillator
which may be used for system lineup
purposes and for the recording of a refer-
ence tone for level adjustment of the
transfer channel.

The power and monitor amplifiers
employ identical circuits, except that a
gain control potentiometer is added in
the monitor amplifier and different
secondary impedances are available at
the output transformers. Three 12AY7
tubes are used to provide the necessary
gain and power handling capacity. The
output stage is a push-pull stage fed from
a conventional phase splitter. The
phase splitter is directly coupled to a
driver stage. Negative feedback by
means of a tertiary winding on the output
transformer is applied to the driver stage
cathode. An additional 12AY7 tube
furnishes voltage gain employing a cir-
cuit similar to the microphone amplifier.

These five amplifiers are all housed to-
gether in an aluminum case (Fig. 10),
the top surface of which contains all con-
trol facilities such as VU meters, mixer
pots, power supply voltage metering
switch, oscillator on-and-off switch, mag-
netic-direct monitoring switch and four
steps of dialogue equalization. In addi-
tion, a pair of jacks has been provided
for monitoring headsets of either 10- or
50-ohm impedance. A separate tele-
phone circuit permits communication
with the recordist and/or boom man.
A talkback microphone is available to
the mixer for slating and directions to the
recordist and boom man during re-
hearsals. A pair of headset jacks has
also been provided for a monitoring
headset for the boom man.

As indicated before, the recorder itself
contains a magnetic monitoring or play-
back amplifier and a bias oscillator.
Two 5879 tubes and two 12AY7 tubes
are used in the playback amplifier. The
push-pull output stage is supplied by a
conventional phase splitter which in
turn is fed by a driver stage. Negative
feedback from the output to the driver
stage cathode is obtained by means of a

338

April 1952 Journal of the SMPTE Vol. 58

Fig. 13. MI-10520 Power Supply.

Fig. 14. MI-10520 Power Supply.

Fig. 15. Overall Channel.

tertiary winding on the output trans-
former. Two 5879 tubes are used in the
first and second stage and stabilized with
negative feedback from second stage out-
put to first stage cathode. A gain con-
trol is located after the second 5879 tube.
The magnetic reproducing head is con-
nected to this amplifier through a special
input transformer. The output of this
amplifier is available at a pair of jacks
for headset monitoring and also brought
back to the mixer amplifier for magnetic
monitoring at the disposal of the mixer.
An equalizer circuit located within the
amplifier provides an overall recording/
reproducing frequency response virtually
flat from 40 to 8000 cycles. The equal-

izer constants can be changed to take
care of film speeds of 90, 45 or 36 fpm.

The bias oscillator utilizes a single
12AU7 tube and furnishes more than
sufficient bias current for all presently
manufactured magnetic emulsions. A
meter which indicates bias current and
which uses a germanium rectifier bridge
circuit is mounted in the recorder front
panel for convenient observation by the
recordist. A bias control which nor-
mally needs no resetting has been located
on the oscillator chassis itself and is
available after opening of the recorder
back cover. The application of bias to
the record head is controlled by a 3-
position switch located on the recorder

Singer and Ward: Magnetic Recording Cost Reduction

339

front panel. In the "play" and "off"
positions no bias is applied to the record
head, whereas in the "record" position,
up to 20 ma of bias current are available.
The bias oscillator chassis also contains a
relay which is actuated whenever the
bias current is turned off as, for instance,
during rehearsals or playback or when
the recorder is operated backwards. It
also disconnects the oscillator from the
output of the recording amplifier and
provides the proper termination. This
has been done as a precautionary meas-
ure in order to avoid accidental con-
tamination of a recording. A separate
relay transfers the recordist's monitor
headset from the output of the playback
amplifier to the recording line which
normally supplies signal to the bias
oscillator. This transfer takes place
automatically whenever the recorder is
not running so that the recordist always
can hear when the mixer wants to com-
municate with him by means of the talk-
back microphone. This arrangement
also permits the recordist to listen to re-
hearsals.

Two 6-conductor cables are used be-
tween the mixer amplifier and the re-
corder. These cables contain signal
transmission circuits, telephone circuits,
buzzer circuits and power circuits. A
separate cable is needed between the re-
corder and the power supply.

Two types of power supplies are pro-
vided. An a-c power supply furnishes
d-c heater current and load and line
regulated B current (Figs. 13 and 14).
In addition, a dynamotor supply will be
available shortly which will permit the
use of storage batteries for location work
and will work in conjunction with the
multiduty motor setup.

The weight of the mixer case complete
with amplifiers and tubes and cover is
31 Ib. The weight of the recorder with
playback amplifier, bias oscillator,
front and back cover is 76 Ib, and
the a-c power supply weighs 29 Ib. All
three units comprising the entire channel
are shown in Fig. 15.

Summary

While this recording channel does not
represent minimum weight and size
facilities, it offers studio quality perform-
ance and provides among many others
the conveniences listed below which
normally would have to be sacrificed to
reduce weight and bulk.

1 . Versatility of film speeds and film
widths, namely, speeds of 90, 45 or 36
fpm; film widths of 35, 17£ or 16mm.

2. Flexibility of drive motors:

a. single-phase, 115-v a-c, 50 and
60 cycles,

b. 3-phase, 220-v a-c, 50-, 60-cycle,

c. multiduty motor which permits
operation from 96-v storage bat-
tery or 208/230-v a-c, and

d. selsyn interlock.

3. One to three tracks.

4. Forward and/or reverse direction
of recording and reproducing.

5. Tight loop threading.

6. Overall recording and reproduc-
ing signal-to-noise ratio between 55 and
60 db can be obtained consistently at
distortion of 2.5%.

References

1. Earl Masterson "35-mm magnetic re-
cording system," Jour. SMPE, 51: 481-
488, Nov. 1948.

2. O. B. Gunby, "Portable magnetic re-
cording system," Jour. SMPE, 52: 613-
61 8, June 1949.

3. Motion Picture Research Council Rec-
ommendation 58. 301 -B.

4. Proposed American Standard, Dimen-
sions for Magnetic Sound Tracks on 35
mm and 17^ mm Motion Picture Film
(First Draft), PH22.86, Jour. SMPTE,
57:72, July 1951.

5. J. S. Chandler, "Some theoretical con-
siderations in the design of sprocket for
continuous film movement," Jour.
SMPE, 37: 164-176, Aug. 1941.

6. M. Rettinger, "A magnetic record-
reproduce head," Jour. SMPTE, 55:
377-390, Oct. 1950.

7. Terms are those defined by N. M.
Haynes, "Magnetic tape and head align-
ments nomenclature," Audio Eng., 33:
22, June 1949.

340

April 1952 Journal of the SMPTE Vol. 58

Constitution of the Society of

Motion Picture and Television Engineers

Name

ARTICLE I

The name of this association shall be
SOCIETY OF MOTION PICTURE
AND TELEVISION ENGINEERS.

ARTICLE II
Objects

Its objects shall be: Advancement in
the theory and practice of engineering
in motion pictures, television, and the
allied arts and sciences; the standard-
ization of equipment and practices em-
ployed therein; the maintenance of a
high professional standing among its
members; and the dissemination of
scientific knowledge by publication.

ARTICLE III

Meetings

There shall be an annual meeting and
such other regular and special meetings as
provided in the Bylaws.

ARTICLE IV

Eligibility for Membership

Any person of good character is eligible
to become a member in any grade for
which he is qualified in accordance with
the Bylaws.

ARTICLE V
Officers

The officers of the Society shall be a
President, an Executive Vice-President,
a Past-President, an Engineering Vice-
President, an Editorial Vice-President, a
Financial Vice-President, a Convention
Vice-President, a Secretary, and a Treas-
urer.

The term of office of all elected officers
shall be for a period of two years.

The President shall not be eligible to
succeed himself in office.

At the conclusion of his term of office
the President automatically becomes Past-
President.

Under conditions as set forth in the
Bylaws, the office of Executive Vice-
President may be vacated before the ex-
piration of his term.

A vacancy in any office shall be filled

for the unexpired portion of the term in
accordance with the Bylaws.

ARTICLE VI
Sections

Sections may be established in accord-
ance with the Bylaws.

ARTICLE VII

Board of Governors

The Board of Governors shall consist
of the President, the Past-President, the
five Vice-Presidents, the Secretary, the
Treasurer, the Section Chairmen, and
twelve elected Governors. An equal
number of these elected Governors shall
reside within the areas included in the
Eastern time zone; the Central time
zone; and the Pacific and Mountain time
zones. The term of office of all elected
Governors shall be for a period of two
years.

ARTICLE VIH
Amendments

This Constitution may be amended as
follows: Amendments may originate as
recommendations within the Board of
Governors, or as a proposal to the Board of
Governors, by any ten members of voting
grade; when approved by the Board of
Governors as set forth in the Bylaws, the
proposed amendment shall then be sub-
mitted for discussion at the annual meet-
ing or at a regular or special meeting
called as provided in the Bylaws. The
proposed amendment, together with the
discussion thereon, shall then be promptly
submitted by mail to all members qualified
to vote, as set forth in the Bylaws. Voting
shall be by letter ballot mailed with the
proposed amendment and discussion to
the voting membership. In order to be
counted, returned ballots must be re-
ceived within sixty (60) days of the
mailing-out date. An affirmative vote of
two thirds of the valid ballots returned,
subject to the above time limitations,
shall be required to carry the amendment,
provided one fifteenth of the duly qualified
members shall have voted within the time
limit specified herein.

341

BYLAWS OF THE SOCIETY OF
MOTION PICTURE AND TELEVISION ENGINEERS

BYLAW I
Membership

Sec. 1. Membership of the Society shall
consist of the following grades : Honorary
members, Sustaining members, Fellows,
Active members, Associate members and
Student members.

An Honorary member is one who has
performed eminent service in the advance-
ment of engineering in motion pictures,
television, or allied arts. An Honorary
member shall be entitled to vote and to
hold any office in the Society.

A Sustaining member is an individual,
company, or corporation subscribing sub-
stantially to the financial support of the
Society.

A Fellow is one who shall be not less than
thirty years of age and who shall by his
proficiency and contributions have at-
tained to an outstanding rank among engi-
neers or executives of the motion picture
or television industries. A Fellow shall be
entitled to vote and to hold any office in
the Society.

An Active member is one who shall be not
less than twenty-five years of age and shall
be or shall have been either one or an
equivalent combination of the following :

(a) An engineer or scientist in motion
picture, television or allied arts. As such
he shall have performed and taken re-
sponsibility for important engineering or
scientific work in these arts and shall have
been in the active practice of his profession
for at least three years, or

(b) A teacher of motion picture, tele-
vision or allied subjects for at least six
years in a school of recognized standing in
which he shall have been conducting a
major course in at least one of such fields,
or

(c) A person who by invention or by
contribution to the advancement of engi-
neering or science in motion picture, tele-
vision or allied arts, or to the technical
literature thereof, has attained a standing
equivalent to that required for Active
membership in (a), or

(d) An executive who for at least three
years has had under his direction impor-
tant engineering or responsible work in the
motion picture, television or allied indus-

tries and who is qualified for direct super-
vision of the technical or scientific fea-
tures of such activities. An Active member
shall be entitled to vote and to hold any
office in the Society.

An Associate member is one who shall be
not less than eighteen years of age, and
shall be a person who is interested in the
study of motion picture or television tech-
nical problems or connected with the
application of them. An Associate mem-
ber is not privileged to vote, to hold office
or to act as chairman of any committee,
although he may serve upon any commit-
tee to which he may be appointed; and,
when so appointed, shall be entitled to the
full voting privileges on action taken by
the committee.

A Student member is any person regis-
tered as a student, graduate or under-
graduate, in a college, university, or other
educational institution of like scholastic
standing, who evidences interest in motion
picture or television technology. Member-
ship in this grade shall not extend more
than one year beyond the termination of
the student status described above. A
student member shall have the same privi-
leges as an Associate member of the Soci-
ety.

Sec. 2. All applications for membership
or transfer should be made on blank forms
provided for the purpose, and shall give a
complete record of the applicant's educa-
tion and experience. Honorary and Fel-
low grades may not be applied for.

Sec. 3. (a) Honorary membership may
be granted upon recommendation of the
Honorary Membership Committee when
confirmed first by a three-fourths majority
vote of those present at a meeting of the
Board of Governors, and then by a four-
fifths majority vote of all voting members
present at any regular meeting or at a
special meeting called as stated in the by-
laws. An Honorary member shall be ex-
empt from the payment of all dues.

(b) Upon recommendation of the Fellow
Award Committee, when confirmed by a
three-fourths majority vote by those pres-
ent at a meeting of the Board of Gover-
nors, an Active member may be made a
Fellow.

342

(c) An Applicant for Active membership
shall give as references at least two mem-
bers of the grade applied for or of a higher
grade. Applicants shall be elected to
membership by a three-fourths majority
vote of the entire membership of the ap-
propriate Admissions Committee. An
applicant may appeal to the Board of
Governors if not satisfied with the action
of the Admissions Committee, in which
case approval of at least three-fourths of
those present at a meeting of the Board
of Governors shall be required for election
to membership or to change the action
taken by the Admissions Committee.

(d) An applicant for Associate member-
ship shall give as reference one member of
the Society, or two persons not members
of the Society who are associated with the
motion picture, television, or allied indus-
try. Applicants shall be elected to mem-
bership by approval of the Chairman of
the appropriate Admissions Committee.

(e) An applicant for Student member-
ship shall be sponsored by a member of the
Society, or by a member of the staff of the
department of the institution he is attend-
ing, this faculty member not necessarily
being a member of the Society. Applicants
shall be elected to membership by approval
of the Chairman of the appropriate Admis-
sions Committee.

Sec. 4. Any member may be suspended
or expelled for cause by a majority vote
of the entire Board of Governors, provided
he shall be given notice and a copy in writ-
ing of the charges preferred against him,
and shall be afforded opportunity to be
heard ten days prior to such action.

BYLAW II

Officers

Sec. 1. An officer or governor shall be
an Honorary member, Fellow, or an Ac-
tive member.

BYLAW HI

Board of Governors

Sec. 1. The Board of Governors shall
transact the business of the Society in ac-
cordance with the Constitution and By-
laws.

Sec. 2. The Board of Governors may act
on special resolutions between meetings,
by letter ballot authorized by the Presi-
dent. An affirmative vote from a majority

of the total membership of the Board of
Governors shall be required for approval
of such resolutions.

Sec. 3. A quorum of ten members of the
Board of Governors shall be present to
vote on resolutions presented at any meet-
ing. Unless otherwise specified, a majority
vote of the Governors present shall con-
stitute approval of a resolution.

Sec. 4. A member of the Board of Gover-
nors may not authorize an alternate to act
or vote in his stead.

Sec. 5. Vacancies in the offices or on the
Board of Governors shall be filled by the
Board of Governors until the annual elec-
tions of the Society.

Sec. 6. The Board of Governors, when
filling vacancies in the offices or on the
Board of Governors, shall endeavor to
appoint persons who in the aggregate are
representative of the various branches or
organizations of the industries interested
in the activities of the Society to the end
that there shall be no substantial predom-
inance upon the Board, as the result of its
own action, of representatives of any one
or more branches or organizations of such
industries.

Sec. 7. The time and place of all except
special meetings of the Board of Governors
shall be determined by the Board of
Governors.

Sec. 8. Special Meetings of the Board of
Governors shall be called by the President
with the proviso that no meeting shall be
called without at least seven days prior
notice to all members of the Board by
letter or telegram. Such a notice shall
state the purpose of the meeting.

BYLAW IV

Administrative Practices

Sec. 1. Special rules relating to the
administration of the Society and known
as Administrative Practices shall be es-
tablished by the Board of Governors
and shall be added to or revised as neces-
sary to the efficient pursuit of the Society's
objectives.

BYLAW V
Committees

Sec. 1. All committees, except as other-
wise specified, shall be formed and ap-
pointed in accordance with the Adminis-
trative Practices as determined by the
Board of Governors.

343

Sec. 2. All committees, except as other-
wise specified, shall be appointed to act
for the term served by the officer charged
with appointing the committees or until
he terminates the appointment.

Sec. 3. Chairmen of the committees
shall not be eligible to serve in such ca-
pacity for more than two consecutive
terms.

Sec. 4. Standing Committees of the
Society to be appointed by the President
and confirmed by the Board of Governors
are as follows:

Honorary Membership Committee

Journal Award Committee

Nominating Committee

Progress Medal Award Committee

Public Relations Committee

Samuel L. Warner Memorial Award
Committee

Sec. 5. There shall be an Admissions
Committee for each Section of the Society
composed of a chairman and three mem-
bers of which at least two shall be members
of the Board of Governors.

Sec. 6. There shall be a Fellow Award
Committee composed of all the officers
and section chairmen of the Society under
the chairmanship of the Past-President.
In case the chairmanship is vacated it shall
be temporarily filled by appointment by
the President.

BYLAW VI

Meetings of the Society

Sec. 1. The location and time of each
meeting or convention of the Society shall
be determined by the Board of Governors.

Sec. 2. The grades of membership en-
titled to vote are defined in Bylaw I.

Sec. 3. A quorum of the Society shall
consist in number of ^ of the total of
those qualified to vote as listed in the
Society's records at the close of the last
fiscal year before the meeting.

Sec. 4. The annual meeting shall be held
during the fall convention.

Sec. 5. Special meetings may be called
by the President and upon the request of
any three members of the Board of Gover-
nors not including the President.

Sec. 6. All members of the Society in any
grade shall have the privilege of discussing
technical material presented before the
Society or its Sections.

BYLAW VII
Duties of Officers

Sec. 1. The President shall preside at
all business meetings of the Society and
shall perform the duties pertaining to that
office. As such he shall be the chief execu-
tive of the Society, to whom all other offi-
cers shall report.

Sec. 2. In the absence of the President,
the officer next in order as listed in Article
V of the Constitution shall preside at
meetings and perform the duties of the
President.

Sec. 3. The seven officers shall perform
the duties separately enumerated below
and those defined by the President:

(a) The Executive Vice-President shall
represent the President, and shall be re-
sponsible for the supervision of the general
affairs of the Society as directed by the
President.

The President and the Executive Vice-
President shall not both reside in the geo-
graphical area of the same Society Section,
but one of these officers shall reside in the
vicinity of the executive offices. Should
the President or Executive Vice-President
remove his residence to the same geo-
graphical area of the United States as the
other, the office of Executive Vice-Presi-
dent shall immediately become vacant and
a new Executive Vice-President shall be
elected by the Board of Governors for the
unexpired portion of the term.

(b) The Engineering Vice-President
shall appoint all technical committees. He
shall be responsible for the general initia-
tion, supervision, and co-ordination of the
work of these committees.

(c) The Editorial Vice-President shall be
responsible for the publication of the
Society's Journal and all other Society
publications.

(d) The Financial Vice-President shall
be responsible for the financial operations
of the Society, and shall conduct them in
accordance with budgets prepared by him
and approved by the Board of Governors.

(e) The Convention Vice-President
shall be responsible for the national con-
ventions of the Society. He shall arrange
for at least one annual convention to be
held in the fall of the year.

Sec. 4. The Secretary shall keep a record
of all meetings; and shall have the re-

344

sponsibility for the care and custody of
records, and the seal of the Society.

Sec. 5. The Treasurer shall have charge
of the funds of the Society and disburse
them as and when authorized by the Finan-
cial Vice-President. He shall be bonded
in an amount to be determined by the
Board of Governors, and his bond shall
be filed with the Secretary.

Sec. 6. Each officer of the Society, upon
the expiration of his term of office, shall
transmit to his successor a memorandum
outlining the duties and policies of his
office.

BYLAW VIII

Society Elections

Sec. 1. All officers and governors shall be
elected to their respective offices by a
majority of ballots cast by voting members
in the following manner:

Nominations shall first be presented by
a Nominating Committee appointed by
the President, consisting of nine members,
including a Chairman. The committee
shall be made up of two Past-Presidents,
three members of the Board of Governors
not up for election, and four other voting
members, not currently officers or gover-
nors of the Society. Nominations shall
be made by three-quarters affirmative
vote of the total Nominating Committee.

Not less than three months prior to the
Annual Fall Meeting, the Board of Gov-
ernors shall review the recommendations
of the Nominating Committee, which shall
have nominated suitable candidates for
each vacancy.

Such nominations shall be final unless
any nominee is rejected by a three-
quarters vote of the Board of Governors
present and voting. The Secretary shall
then notify these candidates of their
nomination. From the list of acceptances,
not more than three names for each va-
cancy shall be selected by the Board of
Governors and placed on a letter ballot.
A blank space shall be provided on this
letter ballot under each office, in which
space the name of any voting member
other than those suggested by the Board
of Governors may be voted for. The bal-
loting shall then take place. The ballot
shall be enclosed with a blank envelope
and a business reply envelope bearing the
Secretary's address and a space for the

member's name and address. One set of
these shall be mailed to each voting mem-
ber of the Society, not less than forty days
in advance of the Annual Fall Meeting.

The voter shall then indicate on the
ballot one choice for each vacancy, seal the
ballot in the blank envelope, place this in
the envelope addressed to the Secretary,
sign his name and address on the latter,
and mail it in accordance with the instruc-
tions printed on the ballot. No marks of
any kind except those above prescribed
shall be placed upon the ballots or enve-
lopes. Voting shall close seven days be-
fore the opening session of the annual fall
convention.

The sealed envelope shall be delivered
by the Secretary to a Committee of Tellers
appointed by the President at the annual
fall convention. This committee shall
then examine the return envelopes, open
and count the ballots, and announce the
results of the election.

The newly-elected officers and governors
of the Society shall take office on January
1, following their election.

BYLAW IX
Dues and Indebtedness

Sec. 1. The annual dues shall be fifteen
dollars ($15) for Fellows and Active mem-
bers, ten dollars ($10) for Associate mem-
bers, and five dollars ($5) for Student
members, payable on or before January 1 ,
of each year. Current or first year's dues
for new members in any calendar year
shall be at the full annual rate for those
notified of election to membership on or
before June 30 ; one half the annual rate
for those notified of election to membership
in the Society on or after July 1 .

Sec. 2. (a) Transfer of membership to a
higher grade may be made at any time
subject to the requirements for initial mem-
bership in the higher grade. If the trans-
fer is made on or before June 30, the an-
nual dues of the higher grade are required.
If the transfer is made on or after July 1,
and the member's dues for the full year
have been paid, one half of the annual dues
of the higher grade is payable less one
half the annual dues of the lower grade.

(b) No credit shall be given for annual
dues in a membership transfer from a
higher to a lower grade, and such transfers
shall take place on January 1, of each year.

345

Sec. 3. Annual dues shall be paid in ad-
vance.

Sec. 4. Failure to pay dues may be con-
sidered just cause for suspension.

BYLAW X

Publications

Sec. 1. The Society shall publish a tech-
nical magazine to consist of twelve
monthly issues, in two volumes per year.
The editorial policy of the Journal shall be
based upon the provisions of the Constitu-
tion and a copy of each issue shall be sup-
plied to each member in good standing
mailed to his last address of record.
Copies may be made available for sale at
a price approved by the Board of Gover-
nors.

BYLAW XI

Local Sections

Sec. 1. Sections of the Society may be
authorized in any locality where the voting
membership exceeds twenty. The geo-
graphic boundaries of each Section shall
be determined by the Board of Governors.
Upon written petition for the authoriza-
tion of a Section of the Society, signed by
twenty or more voting members, the
Board of Governors may grant such
authoriza tion .

Section Membership

Sec. 2. All members of the Society of
Motion Picture and Television Engineers
in good standing residing within the geo-
graphic boundaries of any local Section
shall be considered members of that Sec-
tion.

Sec. 3. Should the enrolled voting mem-
bership of a Section fall below twenty, or
should the technical quality of the pre-
sented papers fall below an acceptable
level, or the average attendance at meet-
ings not warrant the expense of maintain-
ing that Section, the Board of Governors
may cancel its authorization.

Section Officers

Sec. 4. The officers of each Section shall
be a Chairman and a Secretary-Treasurer.
The Section chairmen shall be ex-officio
members of the Board of Governors and
shall continue in such positions for the
duration of their terms as chairmen of the
local Sections. Each Section officer shall
hold office for one year, or until his suc-
cessor is chosen.

Section Board of Managers

Sec. 5. The Board of Managers shall con-
sist of the Section Chairman, the Section
Past-Chairman, the Section Secretary-
Treasurer, and six voting members. Each
manager of a Section shall hold office for
two years. Vacancies shall be filled by
appointment by the Board of Managers
until the annual election of the Section.

Section Elections

Sec. 6. The officers and managers of a
Section shall be voting members of the
Society. All officers and managers shall
be elected to their respective offices by a
majority of ballots cast by the voting
members residing in the geographical area
of the Section. Not less than three
months prior to the annual fall convention
of the Society, nominations shall be pre-
sented to the Board of Managers of the
Section by a Nominating Committee ap-
pointed by the Chairman of the Section,
consisting of seven members, including a
chairman. The committee shall be com-
posed of the present Chairman, the Past-
Chairman, two other members of the
Board of Managers not up for election, and
three other voting members of the Section
not currently officers or managers of the
Section. Nominations shall be made by
a three-quarters affirmative vote of the
total Nominating Committee. Such nom-
inations shall be final, unless any nominee
is rejected by a three-quarters vote of the
Board of Managers, and in the event of
such rejection the Board of Managers will
make its own nomination.

The Chairman of the Section shall then
notify the candidates of their nomination.
From the list of acceptances, not more than
three names for each vacancy shall be
selected by the Board of Managers and
placed on a letter ballot. A blank space
shall be provided on this letter ballot
under each office, in which space the name
of any voting member other than those
suggested by the Board of Managers may
be voted for. The balloting shall then
take place. The ballot shall be enclosed
with a blank envelope and a business reply
envelope bearing the local Secretary-
Treasurer's address and a space for the
member's name and address. One of these
shall be mailed to each voting member of
the Society residing in the geographical

346

area covered by the Section, not less than
forty days in advance of the annual fall
convention.

The voter shall then indicate on the
ballot one choice for each office, seal the
ballot in the blank envelope, place this in
the envelope addressed to the Secretary-
Treasurer, sign his name and address on
the latter, and mail it in accordance with
the instructions printed on the ballot.
No marks of any kind except those above
prescribed shall be placed upon the ballots
or envelopes. Voting shall close seven
days before the opening session of the
annual fall convention. The sealed enve-
lopes shall be delivered by the Secretary-
Treasurer to his Board of Managers at a
duly called meeting. The Board of Man-
agers shall then examine the returned enve-
lopes, open and count the ballots, and an-
nounce the results of the election.

The newly-elected officers and managers
shall take office on January 1, following
their election.

Section Business

Sec. 7. The business of a Section shall be
conducted by the Board of Managers.

Section Expenses

Sec. 8. (a) At the beginning of each
fiscal year, the Secretary-Treasurer of each
section shall submit to the Board of Gover-
nors of the Society a budget of expenses
for the year.

(b) The Treasurer of the Society shall
deposit with each Section Secretary-
Treasurer a sum of money for current ex-
penses, the amount to be fixed by the
Board of Governors.

(c) The Secretary-Treasurer of each
Section shall send to the Treasurer of the
Society, quarterly or on demand, an item-
ized account of all expenditures incurred
during the preceding period.

(d) Expenses other than those enu-
merated in the budget, as approved by the
Board of Governors of the Society, shall
not be payable from the general funds of
the Society without express permission
from the Board of Governors.

(e) The Section Board of Managers
shall defray all expenses of the Section not
provided for by the Board of Governors,
from funds raised locally.

(f) The Secretary of the Society shall,

unless otherwise arranged, supply to each
Section all stationery and printing neces-
sary for the conduct of its business.

Section Meetings

Sec. 9. The regular meetings of a Section
shall be held in such places and at such
hours as the Board of Managers may desig-
nate. The Secretary-Treasurer of each
Section shall forward to the Secretary of
the Society, not later than five days after
a meeting of a Section, a statement of the
attendance and of the business transacted.

Constitution and Bylaws

Sec. 10. Sections shall abide by the
Constitution and Bylaws of the Society
and conform to the regulations of the
Board of Governors. The conduct of Sec-
tions shall always be in conformity with
the general policy of the Society as fixed
by the Board of Governors.

BYLAW XII

Student Chapters

Sec. 1. Student Chapters of the Society
may be authorized in any college, univer-
sity, or technical institute of collegiate
standing. Upon written petition for the
authorization of a Student Chapter, signed
by twelve or more Society members, or
applicants for Society membership, and
the Faculty Adviser, the Board of Gover-
nors may grant such authorization.

Chapter Membership

Sec. 2. All members of the Society in
good standing who are attending the desig-
nated educational institution shall be
eligible for membership in the Student
Chapter, and when so enrolled they shall
be entitled to all privileges that such Stu-
dent Chapter may, under the Constitution
and Bylaws, provide.

Sec. 3. Should the membership of the
Student Chapter fall below ten, or the
average attendance of meetings not war-
rant the expense of maintaining the
organization, the Board of Governors may
cancel its authorization.

Chapter Officers

Sec. 4. The officers of each Student
Chapter shall be a Chairman and a
Secretary-Treasurer. Each Chapter officer
shall hold office for one year, or until his

347

successor is chosen. Where possible,
officers shall be chosen in May to take
office at the beginning of the following
school year. The procedure for holding
elections shall be prescribed in Administra-
tive Practices.

Faculty Adviser

Sec. 5. A member of the faculty of the
same educational institution shall be
designated by the Board of Governors as
Faculty Adviser. It shall be his duty to
advise the officers on the conduct of the
Chapter and to approve all reports to the
Secretary and the Treasurer of the Society.

Chapter Expenses

Sec. 6. The Treasurer of the Society shall
deposit with each Chapter Secretary-
Treasurer a sum of money, the amount to
be fixed by the Board of Governors. The
Secretary-Treasurer of the Chapter shall
send to the Treasurer of the Society at the
end of each school year or on demand an
itemized account of all expenditures in-
curred.

Chapter Meetings

Sec. 7. The Chapter shall hold at least
four meetings per year. The Secretary-
Treasurer shall forward to the Secretary

of the Society at the end of each school
year a report of the meetings for that year,
giving the subject, speaker, and approxi-
mate attendance for each meeting.

BYLAW XIII
Amendments

Sec. 1. Proposed amendments to these
Bylaws may be initiated by the Board of
Governors or by a recommendation to
the Board cf Governors signed by ten
voting members. Proposed amendments
may be approved at any regular meeting
of the Society at which a quorum is present,
by the affirmative vote of two-thirds of
the members present and eligible to vote
thereon. Such proposed amendments
shall have been published in the Journal
of the Society, in the issue next preceding
the date of the stated business meeting of
the Society at which the amendment or
amendments are to be acted upon.

Sec. 2. In the event that no quorum of
the voting members is present at the time
of the meeting referred to in Sec. 1, the
amendment or amendments shall be re-
ferred for action to the Board of Gover-
nors. The proposed amendment or amend-
ments then become a part of the Bylaws
upon receiving the affirmative vote of
three-quarters of the entire membership
of the Board of Governors.

348

Officers

of the Society

April, 1952

HERBERT BARNETT

Executive Vice-President
1951-52

PETER MOLE

President
1951-52

EARL I. SPONABLE

Past-President

1951-52

FRED T. BOWDITCH

Engineering Vice-President

1952-53

JOHN G. FRAYNE

Editorial Vice-President

1951-52

FRANK E. GAHILL, JR.

Financial Vice-President

1952-53

WILLIAM G. KUNZMANN

Convention Vice-President

1951-52

ROBERT M. CORBIN
Secretary
1951-52

349

BARTON KREUZER
Treasurer, 1952-53

THOMAS T. MOULTON
Governor, 1951-52

r^ T^^.

MALCOLM G. TOWNSLEY
Governor, 1951-52

FRANK E. CARLSON
Governor, 1951-52

WILLIAM B. LODGE
Governor, 1951-52

NORWOOD L. SIMMONS
Governor, 1951-52

FRED G. ALBIN
Governor, 1952-53

GEORGE W. COLBURN
Governor, 1952-53

ELLIS W. D'ARCY
Governor, 1952-53

OSCAR F. NEU
Governor, 1951-52

JOSEPH E. AIKEN
Governor, 1952-53

350

JOHN K. MILLIARD

Governor, 1952-53

AXEL G. JENSEN
Governor, 1952-53

C. E. HEPPBERGER
Governor, 1952

VAUGHN C. SHANER
Governor, 1952

E. M. STIFLE
Governor, 1952

OFFICERS AND MANAGERS OF SECTIONS

ATLANTIC COAST: Chairman, E. M. Stifle; Secretary-Treasurer, H. C. Milholland;

Managers. E. A. Bertram, H. A. Ghinn, F. N. Gillette, Richard Hodgson, D. B.

Joy, John G. Stott.
CENTRAL: Chairman, C. E. Heppberger; Secretary-Treasurer, J. L. Wassell;

Managers, E. E. Bickel, W. C. Eddy, I. F. Jacobsen, K. M. Mason, R. H. Ray,

M. G. Townsley.
PACIFIC CO AST: Chairman, Vaughn C. Shaner; Secretary-Treasurer, P. C. Caldwell;

Managers, F. G. Albin, A. C. Blaney, L. G. Dunn, A. M. Gundelfinger, W. F.

Kelley, R. E. Lovell.

STUDENT CHAPTER OFFICERS

NEW YORK UNIVERSITY:

Under Reorganization

UNIVERSITY OF SOUTHERN CALI-
FORNIA: Chairman, Donald Stern;
Secretary-Treasurer, Arthur Schneider

DONALD STERN
Chairman, 1952

351

Treasurer's Report January 1 — December 31, 1951

CASH

Cash on Deposit, Regular Account, Chase National Bank,

January 1, 1951 $30,093.92

Net Receipts (27,857.40)

Cash on Deposit, Regular Account, December 31, 1951 $2,236.52

Cash on Deposit, Payroll Account, Chase Na-
tional Bank, January 1, 1951

Deposits 41,400.00

Total ... 41,400.00

Disbursements 41,302.00

Cash on Deposit, Payroll Account, December 31, 1951 98.00

Petty Cash Fund 200.00

Total Cash on Deposit and on Hand 2,534.52

INVESTMENTS

Savings Accounts, January 1, 1951 . . . .
Add : Interest Credited . . . . •'.. . . .

31,419.71
927.63

Total
Less: Account Closed

32,347.34
5,138.43

Savings Accounts, December 31, 1951 . .
U.S. Government Bonds (at cost) ....

27,208.91
60,000.00

Total Investments

Total Cash and Investments, December 31, 1951 .

87,208.91
$89,743.43

Respectfully submitted,
FRANK E. CAHILL, Jr., Treasurer

Summary of Financial Condition — Dec. 31, 1951

ASSETS (What Your Society Owns]

Cash on Hand and in Bank $ 2,534.52

Savings Accounts 27,208.91

U.S. Government Bonds (at cost) 60,000.00

Accounts Receivable 21,719.86

Test Film Inventory 53,019.81

Test Film Equipment (memo value) 1 . 00

Office Furniture and Equipment (memo value) 1 . 00

Prepaid Expenses 63 . 00

Total Assets $164,548.10

LIABILITIES (What Tour Society Owes]

Accounts Payable $ 22 , 640 . 07

Due to Customers 860.85

Membership Dues Received in Advance 12,687.85

N.Y.C. Sales Tax Payable 14.38

Reserve for 1955 Five-Year Index . . 500.00

Total Liabilities $ 36,703.15

MEMBERS' EQUITY (What Tour Society Is Worth) 127,844.95

Total Liabilities and Members' Equity $164,548.10

352

Statement of Income and Expenses

January 1 — December 31, 1951

Test Film Operations

Test Film Sales $133,746.17

Cost of Test Films Sold 79,148.48

Net Income From Test Film Operations $54,597.69

Publications Operations

Publications Income $ 20,774.36

Cost of Publications 45,467.44

Net Loss From Publications Operations (24,693.08)

Other Operations

Other Operations Income $ 388 . 49

Cost of Other Operations 710.56

Net Loss From Other Operations (322 . 07)

Other Income

Membership Dues $ 60,511.51

Interest Earned 2,454.70

Miscellaneous Income . 101.11

Total Other Income 63,067.32

Total Operating Income $92,649c. 86

Operating Expenses

Engineering $ 13,026.44

Administrative 59,866.31

Officers 108.85

Sections and Chapters 2,700.00

Affiliations 1,385.00

Conventions 1,230.06

Total Operating Expenses 78,316.66

Net Operating Income $14,333.20

Other Deductions

Depreciation of Test Film Equipment $ 3 , 729 . 85

Excess in Reserve for 1950 Five-Year Index . . . (453.53)

Provision for 1 955 Five-Year Index 500 . 00

Total Other Deductions 3 , 776 . 32

Excess of Income Over Expenses $10,556.88

The foregoing financial statements were prepared from the records of the Society for the
year 1951 and reflect the results of operations for that year. The records and financial
statements were audited for the year ended December 31, 1951, by Wilbur A. Smith,
Certified Public Accountant, New York City, and are in conformity with that audit.

RALPH B. AUSTRIAN, Financial Vice-President

353

Membership Report

For Year Ended December 31, 1951

Hon. Sust. Pel. Act. Assoc. Stud. Total

Membership, January 7, 1951 ... 4 79 198 931 1887 184 3283

New Members 2 171 291 67 531

Reinstatements . 10 20 6 36

4 81 198 1112 2198 257 3850

Resignations -2 -2 -15 -27 -5 -51

Deceased -1 -3 -5 -8 -17

Delinquent -3 -2 -72 -194 -23 -294

3 76 191 1020 1969 229 3488
Changes in Grade :

Active to Fellow . ... 16—16

Associate to Active . . 114—114

Student to Associate . . 14—14

Active to Associate —4 4

Membership, December 31,1951 . . 3 76 207 1114 1873 215 3488

Nonmember Subscription Report

For Year Ended December 31, 1951

Subscriptions, January 1, 1951 575

New Subscriptions and Previous Cutoffs 892

1467

Cutoffs and Expirations 439

Subscriptions, December 31, 1951 1028

354

Awards

In accordance with the provisions of the
Administrative Practices of the Society
and the regulations for granting , the
Journal Award, the Progress Medal
Award, the Samuel L. Warner Memorial

Award and the David SarnofF Gold Medal
Award, a list of names of previous re-
cipients and the reasons for the awards
are published annually in the Journal as
follows :

Journal Award

The Journal Award Committee shall
consist of five Fellows or Active mem-
bers of the Society, appointed by the
President and confirmed by the Board
of Governors. The Chairman of the
Committee shall be designated by the
President.

At the fall convention of the Society
a Journal Award Certificate shall be
presented to the author or to each of
the authors of the most outstanding
paper originally published in the Journal
of the Society during the preceding
calendar year.

Other papers published in the Journal
of the Society may be cited for Honorable
Mention at the option of the Committee,
but in any case should not exceed five in
number.

The Journal Award shall be made on
the basis of the following qualifications :

(1) The paper must deal with some
technical phase of motion picture engi-
neering.

(2) No paper given in connection
with the receipt of any other Award of
the Society shall be eligible.

(3) In judging of the merits of the
paper, three qualities shall be considered,
with the weights here indicated: (a) tech-
nical merit and importance of material,
45%; (b) originality and breadth of in-
terest, 35%; and (c) excellence of presen-
tation of the material, 20%.

A majority vote of the entire Com-
mittee shall be required for the election
to the Award. Absent members may
vote in writing.

The report of the Committee shall be
presented to the Board of Governors at
their July meeting for ratification.

These regulations, a list of the names
of those who have previously received

the Journal Award, the year of each
Award, and the titles of the papers shall
be published annually in the Journal of
the Society. In addition, the list of
papers selected for Honorable Mention
shall be published in the Journal of the
Society during the year current with the
Award.

The recipients are listed below by year,
with the date of Journal publication
given after the title.

1934, P. A. Snell, "An introduction to the
experimental study of visual fatigue,"
May 1933.

1935, L. A. Jones and J. H. Webb,
"Reciprocity law failure in photographic
exposure," Sept. 1934.

1936, E. W. Kellogg, "A comparison of
variable-density and variable-width sys-
tems," Sept. 1935.

1937, D. B. Judd, "Color blindness and
anomalies of vision," June 1936.

1938, K. S. Gibson, "The analysis and
specification of color," Apr. 1937.

1939, H. T. Kalmus, "Technicolor ad-
ventures in cinemaland," Dec. 1938.

1940, R. R. McNath, "The surface of the
nearest star," Mar. 1939.

1941, J. G. Frayne and Vincent Pagliarulo,
"The effects of ultraviolet light on
variable-density recording and print-
ing, June 1940.

1942, W. J. Albersheim and Donald Mac-
Kenzie, "Analysis of soundfilm drives,"
July 1941.

1943, R. R. Scoville and W. L. Bell,
"Design and use of noise-reduction
bias systems," Feb. 1942 (Award made
Apr. 1944).

1944, J. I. Crabtree, G. T. Eaton and
M. E. Muehler, "Removal of hypo and
silver salts from photographic materials
as affected by the composition of the
processing solutions," July 1943.

355

1945, C. J. Kunz, H. E. Goldberg and
C. E. Ives, "Improvement in illumina-
tion efficiency of motion picture
printers," May 1944.
946, R. H. Talbot, "The projection life
of film," Aug. 1945.

1947, Albert Rose, "A unified approach
to the performance of photographic film,
television pickup tubes, and the human
eye," Oct. 1946.

1948, J. S. Chandler, D. F. Lyman and
L. R. Martin, "Proposals for 16-mm
and 8-mm sprocket standards," June
1947.

1949, F. G. Albin, "Sensitometric aspect
of television monitor-tube photog-
raphy," Dec. 1948.

1950, Frederick J. Kolb, Jr., "Air cooling
of motion picture film for higher
screen illumination," Dec. 1949.

1951, A. B. Jennings, W. A. Stanton and
J. P. Weiss, "Synthetic color-forming
binders for photographic emulsions,"
Nov. 1950.

The present Chairman of the Journal
Award Committee is Frederick J. Kolb, Jr.

Progress Medal Award

The Progress Medal Award Committee
shall consist of five Fellows or Active
members of the Society, appointed by the
President and confirmed by the Board of
Governors. The Chairman of the Com-
mittee shall be designated by the Presi-
dent.

The Progress Medal may be awarded
each year to an individual in recognition
of any invention, research or develop-
ment which, in the opinion of the Com-
mittee, shall have resulted in a significant
advance in the development of motion
picture technology.

Any member of the Society may recom-
mend persons deemed worthy of the
Award. The recommendation in each
case shall be in writing and in detail as
to the accomplishments which are thought
to justify consideration. The recom-
mendation shall be seconded in writing
by any two Fellows or Active members
of the Society, who shall set forth their
knowledge of the accomplishments of the
candidate which, in their opinion, justify
consideration.

A majority vote of the entire Com-
mittee shall be required to constitute an
Award of the Progress Medal. Absent
members may vote in writing.

The report of the Committee shall be
presented to the Board of Governors at
their July meeting for ratification.

The recipient of the Progress Medal
shall be asked to present a photograph of
himself to the Society and, at the discre-
tion of the Committee, may be asked to
prepare a paper for publication in the
Journal of the Society.

These regulations, a list of the names of
those who have previously received the
Medal, the year of each Award and a
statement of the reason for the Award
shall be published annually in the Journal
of the Society.

Awards have been made as follows:

1935, E. C. Wente, for his work in sound
recording and reproduction, Dec. 1935.

1936, C. E. K. Mees, for his work in
photography, Dec. 1936.

1937, E. W. Kellogg, for his work in sound
reproduction, Dec. 1937.

1938, H. T. Kalmus, for his work in de-
veloping color motion pictures, Dec.
1938.

1939, L. A. Jones, for his scientific re-
searches in photography, Dec. 1939.

1940, Walt Disney, for his contributions
to motion picture photography and
sound recording of feature and short
cartoon films, Dec. 1940.

1941, G. L. Dimmick, for his development
activities in motion picture sound re-
cording, Dec. 1941.

No Awards were made in 1942 and 1943.

1944, J. G. Capstaff, for his research and
development of films and apparatus
used in amateur cinematography, Jan.
1945.

No Awards were made in 1945 and 1946.

1947, J. G. Frayne, for his technical
achievements and the documenting of
his work in addition to his contributions
to the field of education and his inspira-
tion to his fellow engineers, Jan. 1948.

1948, Peter Mole for his outstanding
achievements in motion picture studio

35$

lighting which set a pattern for lighting
techniques and equipment for the
American motion picture industry,
Jan. 1949.

1949, Harvey Fletcher for his outstand-
ing contributions to the art of record-
ing and reproducing of sound for motion
pictures, Oct. 1949

1950, V. K. Zworykin, for his outstanding

contributions to the development of
television, Dec. 1950.
1951, Earl I. Sponable, for outstanding
contributions to technical advancement
of the motion picture art, particularly
with respect to sound on film, color and
large-screen television, Dec. 1951.

The present Chairman of the Progress
Medal Award Committee is D. B. Joy.

Samuel L. Warner Memorial Award

Each year the President shall appoint
a Samuel L. Warner Memorial Award
Committee consisting of a chairman and
four members. The chairman and com-
mittee members must be Active Members
or Fellows of the Society. In consider-
ing candidates for the Award, the com-
mittee shall give preference to inventions
or developments occurring in the last five
years. Preference should also be given
to the invention or development likely
to have the widest and most beneficial
effect on the quality of the reproduced
sound and picture. A description of the
method or apparatus must be available
for publication in sufficient detail so that
it may be followed by anyone skilled in the
art. Since the Award is made to an in-
dividual, a development in which a group
participates should be considered only if
one person has contributed the basic idea
and also has contributed substantially to
the practical working out of the idea. If,
in any year, the committee does not con-
sider any recent development to be more
than the logical working out of details
along well-known lines, no recommenda-
tion for the Award shall be made. The
recommendation of the committee shall be
presented to the Board of Governors at
the July meeting.

The purpose of this Award is to en-
courage the development of new and im-
proved methods or apparatus designed
for sound-on-film motion pictures, in-
cluding any step in the process.

Any person, whether or not a member
of the Society of Motion Picture and Tele-
vision Engineers, is eligible to receive the
Award.

The Award shall consist of a gold medal
suitably engraved for each recipient. It

shall be presented at the Fall Convention
of the Society, together with a bronze
replica.

These regulations, a list of those who
previously have received the Award, and
a statement of the reason for the Award
shall be published annually in the Journal
of the Society. The recipients have been :

1947, J. A. Maurer, for his outstanding
contributions to the field of high-quality
16-mm sound recording and reproduc-
tion, film processing, development of
16-mm sound test films, and for his in-
spired leadership in industry standardi-
zation (citation published, Jan. 1948).

1948, Nathan Levinson, for his outstand-
ing work in the field of motion picture
sound recording, the intercutting of
variable-area and variable-density
sound tracks, the commercial use of
control track for extending volume
range, and the use of the first sound-
proof camera blimps (citation pub-
lished, Jan. 1949).

1949, R. M. Evans, for his outstanding
work in the field of color motion picture
films, including research on visual
effects in photography and develop-
ment work on commercial color proc-
esses (citation published, Oct. 1949).

1950, Charles R. Fordyce, for his efforts
in and achievement of the develop-
ment of triacetate safety base film
(citation published, Dec. 1950).

1951, Earl I. Sponable, for years of re-
search and development in recording
of sound on film (citation published
Dec. 1951).

The present Chairman of the Samuel L.
Warner Memorial Award Committee is
Glenn L. Dimmick.

357

David Sarnoff Gold Medal Award

The David Sarnoff Gold Medal Award
Committee, appointed by the President,
shall consist of five Fellows, Honorary
Members or former recipients of some
formal Society Award, each of whom shall
be qualified to judge the importance or
value of current work in some technical
phase of the broad field of television engi-
neering, whether in research, development,
design, manufacture, operation, or in any
similar phase of theater television.

The award shall consist of a gold medal,
together with a bronze replica and a
citation, stating the recipient's qualifica-
tions.

The David Sarnoff Gold Medal may
be awarded each year to any qualified
individual, whether or not currently
a member of this Society, in recognition of
recent technical contributions to the art of
television, to encourage the development
of new techniques, new methods and new
equipment which hold promise for the
continued improvement of television, pref-
erence to be given for work having reached
completion within the preceding five
years.

Recommendations of the Committee
and a report of its deliberations shall be

presented to the Board of Governors three
months in advance of the time for pres-
entation (at the July meeting of the
Board, for presentation at the Fall Con-
vention). Any member of the Society
may recommend persons deemed worthy
of the Award. The recommendation in
each case shall be in writing and in detail
as to the accomplishments which are
thought to justify consideration.

These regulations, a list of the names of
those who have previously received the
medal, the year of each Award and a state-
ment of the reason for the Award shall be
published annually in the Journal of the
Society. The first recipient is:

1951, Otto H. Schade, for his outstanding
accomplishments in the fields of tele-
vision and motion picture science and
engineering, in outlining the potentiali-
ties of television and film systems as to
fidelity of photography and reproduction
of images (citation published Dec.
1951).

The present Chairman of the David
Sarnoff Gold Medal Award Committee
is Pierre Mertz.

HONORARY MEMBERS

Lee de Forest
Edward W. Kellogg

A. S. Howell
V. K. Zworykin

The distinction of Honorary Membership in the Society is awarded to
living pioneers whose basic contributions when examined through the
perspective of time represent a substantial forward step in the recorded
history of the arts and sciences with which the Society is most concerned.

SMPTE HONOR ROLL

Louis Aime Augustin Le Prince
William Friese-Greene
Thomas Alva Edison
George Eastman
Frederic Eugene Ives
Jean Acme Le Roy
C. Francis Jenkins
Eugene Augustin Lauste
William Kennedy Laurie Dickson

Edwin Stanton Porter
Herman A. DeVry
Robert W. Paul
Frank H. Richardson
Leon Gaumont
Theodore W. Case
Edward B. Craft
Samuel L. Warner
Louis Lumiere
Thomas Armat

Elevation to the Honor Roll of the Society is granted to each distinguished
pioneer who during his lifetime was awarded Honorary Membership or
whose work was recognized subsequently as fully meriting that award.

358

1952 Nominations

Candidates for election to national office
of the Society are now being considered
by the Nominating Committee. The
eleven vacancies which will occur at the
end of 1952 and are to be filled by {his
year's election are the offices of President,
Executive Vice-President, Editorial Vice-
President, Convention Vice-President,
Secretary, two Governors from the West, two
Governors from the Central area, and two

Governors from the East. Names of the
incumbents will be found on the inside
back cover of each issue of the Journal.

Members in the Honorary, Fellow and
Active Grades are invited by the Chairman
of the Nominating Committee to submit
their suggestions for candidates at the
earliest possible dates. Address them to
Earl I. Sponable, Movietonews, Inc., 460
W. 54th St., New York 19, N.Y.

Papers on Photographic Instrumentation

Instrumentation is the subject of this
year's symposium of the Society of Photo-
graphic Engineers, to be held on June 4
and 5 at the Naval Ordnance Laboratory,
White Oak, Md., according to information
from SPE President Edward K. Kaprelian.
The symposium will cover equipments,

materials and techniques involved in the
recording of data. Papers relating to
high-speed cinematography will not be
presented. Information about possible
instrumentation papers will be welcomed
by the symposium chairman, D. Max
Beard, 4304 S. Capitol, Washington 20,
D.C.

Book Reviews

Television Engineering
(Second Edition)

By D. G. Fink. Published (1952) by
McGraw-Hill, 330 W. 42 St., New York
36. i-xiv + 690 pp. + 12 pp. appendix +
19 pp. index. 512 illus. 6 X 9 in.
Price $8.50.

Mr. Fink is one of those all too rare
individuals — an engineer who can write.
His previous books have been noted for
their clear, lucid style and one would be
disappointed if this one were not up to his
previous standards. As a matter of fact,
it is, if anything, superior to his earlier
books in this respect and he has succeeded
in turning out a text book for television
engineering which is extremely clear and
well written.

The book covers the entire field of
television engineering starting with the
fundamentals and progressing to a fairly
detailed description of commercial tele-
vision transmitting studio and receiving
equipment. Two chapters of the book
are devoted to an especially good descrip-

tion of color television which includes a
consideration of color fundamentals and
an objective study of the various systems
which have been proposed for the trans-
mission of television pictures and color.
Television engineering covers such a wide
variety of subject matter, drawing as it
does upon combinations of practically all
of the physical sciences, that any attempt
to cover the entire system in one book will
inevitably result in treatment which will
seem superficial to the specialists. For
example, in his discussion of radio wave
propagation, Mr. Fink barely mentions
the important work which was done by
the FCC Ad Hoc Committee in connection
with the determination of a terrain factor
which describes the deviation of the
median signal intensity from the smooth
earth value because of the irregularities
in the earth surface. Again, his discussion
of the definition obtainable from the va-
rious components in the television system
is entirely in terms of the resolving power
of the various components. He must be
ignoring the important work of Schade and

359

others who have shown that this is not an
adequate criterion for picture definition.

The treatment of such a wide variety
of subject matter probably leads inevitably
to errors of fact which occur from time to
time in the book. For example, an
equivalent circuit which is supposed to show
the input impedance of a balun is shown
in Fig. 283; this circuit has a series LG
circuit presumably resonant at the center
of the frequency band shunted across the
input terminals, so that input impedance
of this frequency can be a short circuit.
Again, on page 326, there is the following
description of defraction of energy past
the horizon: "Defraction occurs when the
instant energy, following tangentially on
the rim of the obstacle, is re-radiated from
absorbing points on the rim." Even aside
from the contradiction in terms involved
in re-radiation from an absorbing object,
this is surely not an accurate description
of the phenomena of defraction.

The criticisms of the book described
above were meant to illustrate the in-
evitable difficulties which arise in covering
so much territory in one volume and not
to deprecate what, in general, represents
a very excellent job in doing what it was
intended to do. The beginning student of
television engineering or the specialist
attempting to obtain a broad background
in fields other than his own will find the
book well organized, readable, and, with
a few exceptions such as those noted above,
accurate. — Mclntosh & Inglis, Consulting
Radio Engineers, 777 14th St., N.W.,
Washington 5, D.G.

Prism and Lens Making
(Second Edition)

By F. Twyman. Published (1952) by
Hilger & Watts Ltd., 98 St. Pancras Way,
London, N.W. 1. Distributed in U.S.A.
by the Jarrell-Ash Co., 165 Newbury
St., Boston, Mass, i-viii + 590 pp. +
27 pp. appendix -f 5 pp. bibliography -f
7 pp. index. 260 illus. 5$ X 8$ in.
Price $11.25.

Although this is called a second edition
of Twyman's 1942 book on prism and lens
making, it is so much larger than the
original (629 pages against the former
178) that it might almost be regarded as
a new work. Where the previous treat-

ment was stilted and severe, the new is
easy to read and full of anecdotes and
illustrative material of every kind. Indeed,
the number of references to both ancient
and recent authorities is extraordinary,
and the writing is in the best tradition of
Rayleigh or Dennis Taylor.

The chief charge against the previous
edition was that only the procedures and
techniques in use by Adam Hilger Ltd.
were described. This was not very
surprising as Mr. Twyman is the emeritus
Managing Director of Hilger's, but in the
new edition this is no longer the case.
The author has gone to the greatest trouble
to ascertain the methods used by other
manufacturers (mainly, however, in Eng-
land), and has described them impartially.
This of course increases the value of the
book very greatly, since Hilger's produc-
tion is small in quantity but wide in
variety and of the highest quality, while
in some other companies the need for large-
scale or mass production of lower-grade
lenses has led to the development of entirely
different manufacturing procedures.

In addition to a survey of the regular
methods for the grinding, polishing, center-
ing and cementing of lenses and prisms,
several new chapters have been added
dealing with such subjects as optical
crystals and plastics and the manufacture
of optical elements from them, microscope
objectives, large astronomical objectives
and mirrors, the surface treatment of
lenses, spectacle lenses, and an excellent
summary of the methods available for the
generation of nonspherical surfaces. Al-
most 100 pages are devoted to the testing
of optical work, both on the individual
surfaces and on the completed systems.
The tests of Fizeau, Foucault, Newton,
Hartmann, Zernicke, and others are fully
described, and in a separate chapter the
applications of the author's well-known
interferometers receive extensive treat-
ment. The nature of glass and its anneal-
ing, and workshop tests for optical glass,
are well covered.

Among the useful appendices there is a
glossary of equivalent terms used in the
optical industry in English, French and
German. There is an extensive bibliog-
raphy, and a good index. The paper
and printing are excellent, but the review
copy as received was poorly bound and

360

the cover was already falling off. Mis-
prints are negligibly few. This excellent
book can be very strongly recommended
to all who have a close connection with the
optical industry, or any occasion to grind
and polish a lens. — R. Kingslake, Optical
Design Dept., Hawk-Eye Works, Eastman
Kodak Co., Rochester 4, N.Y.

dogmatic over-simplifications by hedging
with carefully worded reservations. One
must regretfully state, however, that the
book's worthy aim of explaining the nature
of film art to the general public falls very
short of its fulfillment. — George L. George,
Screen Directors Guild, 133 E. 40 St.,
New York 16.

Dynamics of the Film

By Joseph and Harry Feldman. Published
(1952) by Hermitage House, 8 W. 13 St.,
New York 11. 241 pp. + 3 pp. bibliog-
raphy + 2 pp. periodicals listing + 7 pp.
index. Illustrated. 5| X 8 in. Price
$3.50.

The main risk in attempting to
"popularize" a difficult subject, especially
in the field of aesthetics, lies in depriving
it of all human and artistic warmth and in
reducing it to a mere mechanical strata-
gem.

In this pitfall is precisely where the
Messrs. Feldman have landed. Their
book, intended purposely "for the BIG
audience of movie-goers," fails to convey
the meaning and essence of a film's overall
dramatic impact. It is a case of not seeing
the forest for the trees, and their analysis
of the basic elements of a film constitutes
a reductio ad absurdum of the approach they
have chosen.

To some extent, they seem aware of their
predicament. They try to tone down their

Standards for single-line diagrams for

use in both power and communication
work combined in one volume in The
American Standard Graphical Symbols
for Single (One) Line Electrical Engineer-
ing Diagrams, Z32. 1.1 -1951, published by
the American Standards Association, 70
E. 45 St., New York 17, at $1.40 per copy.
This standard coordinates and modifies
the single-line diagrams contained in the
American Standard Graphical Symbols
for Electrical Power and Control, Z32.3-
1946, and for Telephone, Telegraph and
Radio Use, Z32.5-1944.

The American Institute of Electrical
Engineers and the American Society of
Mechanical Engineers were sponsors of
the new standard, which contains 81
sections covering symbols for almost all
electrical engineering work in the fields of
power and communication. Sample dia-
grams show the use of the single line
drawing in illustrations of a laboratory
sound system, a microwave test setup
telephone repeater and line equipment,
and power equipment.

Test films are the customary tool for checking picture and sound performance in theaters,
service shops, in factories and in television stations. Twenty-seven different test films
in 16mm and 35mm sizes are produced by the Society and the Motion Picture Research
Council. Write to Society Headquarters for a free catalog.

Six American Standards have been added to the Motion Picture Set of 60 which the
Society has had available for sale. To holders of the present set the Society has made
available the six new standards: PH22.11-1952, PH22.24-1952, PH22.73-1951, PH22.74-
1951, PH22.76-1951 and PH22.82-1951. The price is $1 plus 3% sales tax on deliveries
in New York City.

The new set of 66 standards in a heavy three-post binder with an index is available at
$14.50 plus 3% sales tax on deliveries in New York City; foreign postage is $.50 extra.

All standards in sets only are available from Society Headquarters. Single copies of
any particular standard must be ordered from the American Standards Association,
70 East 45th St., New York 17, N.Y.

361

New Members

The following members have been added to the Society's rolls since those last published.
The designations of grades are the same as those used in the 1950 MEMBERSHIP DIRECTORY.

Honorary (H) Fellow (F) Active (M) Associate (A) Student (S)

Angarola, Salvatorc, SRT— TV Studios.
Mail: 90-50—53 Ave., Elmhurst, L.I.,
N.Y. (S)

Arora, O. P., University of Southern Cali-
fornia. Mail: 1183£ W. 29 St., Los
Angeles 7, Calif. (S)

Bartleson, C. James, Jr., Photographic
Color Technician, Pavelle Color, Inc.
Mail: 7018 Colonial Rd., Brooklyn,
N.Y. (A)

Booth, John H. L., University of Southern
California. Mail: Ste. C, 2730 S.
Normandie, Los Angeles 7, Calif. (S)

Bray, Frederic L., Engineer, Du-Art Film
Laboratories. Mail: 353 Pin Oak La.,
Westbury, L.I., N.Y. (A)

Catanzaro, Carl J., SRT— TV Studios.
Mail: 27-19— 24 Ave., Astoria, L.I., N.Y.
(S)

Colman, Robert, University of Southern
California. Mail: 1732£ W. 20 St.,
Los Angeles, Calif. (S)

Deutch, Irving, New Inst. for Film &
Television. Mail: 2110 Newkirk Ave.,
Brooklyn, N.Y. (S)

Dickinson, William A., Electronics Engi-
neer, Sylvania Electric Products, Inc.,
Seneca Falls, N.Y. (M)

Doha, Stephen, Jr., Telephone Engineer,
Bell Telephone Laboratories, Inc., Mur-
ray Hill, NJ. (A)

Erlinger, Joseph A., Foreman, Camera
Shop, Warner Brothers. Mail: 1212
S. Crescent Heights Blvd., Los Angeles
35, Calif. (A)

Everest, F. Alton, Associate Director,
Moody Institute of Science. Mail:
11428 Santa Monica Blvd., Los Angeles
25, Calif. (A)

Goren, Lewis, SRT— TV Studios. Mail:
124 E. 146 St., New York 51, N.Y. (S)

Gregory, John R., New Institute for Mo-
tion Pictures. Mail: 64-12 — 65 PI.,
Middle Village 79, N.Y. (S)

Hall, Frank, Clinical-Surgical Photogra-
pher, Dept. of Veterans Affairs, Sunny-
brook Hospital. Mail: 1068 St. Clair
Ave., W., Toronto, Ont., Canada. (A)

Hanson, Charles L., Jr., Photographic
Technician, Arthur D. Little, Inc.
Mail: 54 Hammond St., Cambridge 38,
Mass. (A)

Harber, Richard G., University of South-
ern California. Mail: 7843 Flight
Ave., Los Angeles 45, Calif. (S)

Hollzer, Herbert M., University of
Southern California. Mail: 820 S.
Mansfield, Los Angeles 36, Calif. (S)

Howland, Walter A., Optical Engineer,
J. A. Maurer, Inc. Mail: 179 Sadler
Rd., Bloomfield, NJ. (A)

Jacobsen, Michael M., Sound Engineer,
A/S Palladium Film. Mail: Gustav
Adolfs Gade 5, Copenhagen 0, Den-
mark. (A)

Jamieson, Hugh V., Jr., Production Man-
ager, Partner, Jamieson Film Co.
Mail: 3825 Bryan, Dallas, Tex. (M)

Kayser, Paul W., Foreign Manager,
Westrex Corp. Mail: 299 S. Middle-
town Rd., Pearl River, N.Y. (A)

Kirk, Michael, Film Editor, WOSM-TV.
Mail: 2139 Gen. Taylor, New Orleans
15, La. (M)

Linton, C. Bruce, University of Southern
California. Mail: 401 Adlena Dr.,
Fullerton, Calif. (S)

Long, Maurice L., University of Southern
California. Mail: 3202 W. 43 PI.,
Los Angeles 8, Calif. (S)

Maclsaac, Donald M., Sound Editor,
Syracuse University, Audio-Visual Cen-
ter. Mail: 304 Farmer St., Syracuse,
N.Y. (A)

Madore, Douglas, Actor, Director, Free-
lance. Mail: 6088 Selma Ave., Holly-
wood 28, Calif. (S)

Mell, Labe B., General Manager, Reela
Films, Inc., 17 N.W. Third St., Miami,
Fla. (A)

Mendelwager, Jerome, SRT — TVStudios.
Mail: 1016 Boulevard, Bayonne, NJ.
(S)

Nesbitt, Charles D., Motion Picture Tech-
nical Representative, E. I. du Pont de
Nemours & Co. Mail: 3289 N. Cali-
fornia Ave., Chicago, 111. (M)

Netervala, Minoo, University of Southern
California. Mail: 1190 W. Adams
Blvd., Los Angeles 7, Calif. (S)

Noriega, Joseph, Motion Picture Pro-
ducer, Reforma 77, Apt. 1107, Mexico
City, Mexico. (M)

Oleson, Robert, University of Southern
California. Mail: 207| S. Hoover,
Los Angeles, Calif. (S)

Pascal, Captain Samuel, Hq. Sqd., 131
A.B. Gp., George Air Force Base, Victor-
ville, Calif. (A)

Patelis, George, SRT— TV Studios. Mail:
87-72—253 St., Bellerose, N.Y. (S)

362

Pritzlaff, Kipp, University of Southern
California. Mail: 14340 Dickens,
Sherman Oaks, Calif. (S)

Quiroga, Alex S., TV Engineer, ABC-TV.
Mail: 3757£ Monon St., Hollywood,
Calif. (M)

South, David F. W., University of South-
ern California. Mail: 5353 W. Third
St., The Art Center School, Los Angeles,
Calif. (S)

Swenson, Russell, University of Southern
California. Mail: 682 W. 35 St., Los
Angeles 7, Calif. (S)

Ward, Julius C., Electronic Engineer,
General Precision Laboratory, 7 Man-
ville La., Pleasantville, N.Y. (A)

Wheeler, Charles F., Assistant Cameraman,
Free-lance. Mail: 2557 Westwood
Blvd., Los Angeles 64, Calif. (A)

Wilt, Chester, Development Engineer,
Eastman Kodak Co. Mail: 4007 St.
Paul Blvd., Rochester 17, N.Y. (A)

Win, Maung Nay, University of Southern
California. Mail: 1130 W. 37 St.,
Los Angeles, Calif. (S)

Wong, Willie, SRT— TV Studios. Mail:
66 Cedar Dr., Farmingdale, N.Y. (S)

CHANGES IN GRADE

Bauman, Harold W., (A) to (M)
Demetros, Nicholas K., (A) to (M)
Duvall, Delmer P., (A) to (M)
Gemeinhardt, George C., (A) to (M)
Gillet, Albert, (A) to (M)
Helhena, Leslie E., (A) to (M)
Kemp, Jay S., (A) to (M)
Krulish, John A., (A) to (M)
Manley, Fred A., (A) to (M)
McGough, William A., (A) to (M)
Newmayer, Richard H., (A) to (M)
Pittaro, Ernest M., (A) to (M)
Schwarz, Sigmund, (A) to (M)
Searle, Milton H., (A) to (M)
Smith, H. Beresford, (A) to (M)
Sparks, R. F., (A) to (M)
Szeglin, Stephen J., (A) to (M)
Wesson, Rufus, (A) to (M)

Meetings

The Atlantic Coast Section of the SMPTE will meet on April 16, 7:30 P.M., at the
Henry Hudson Hotel, New York City, when Robert Dressier of Paramount Pictures
Corp.'s Chromatic Television Laboratories will present a paper and a demonstration on
electrooptic sound recording on film.

71st Semiannual Convention of the SMPTE, April 21-25, The Drake, Chicago

Other Societies

American Physical Society, May 1-3, Washington, D.C.
Acoustical Society of America, May 8-10, New York

Society of Photographic Engineers, Symposium on Instrumentation, June 4-5, Naval

Ordnance Laboratory, White Oak, Md.

American Institute of Electrical Engineers, Summer General Meeting, June 23-27,

Hotel Nicollet, Minneapolis, Minn.

American Physical Society, June 30-July 3, Denver, Colo.

National Audio-Visual Association, Convention and Trade Show, Aug. 2-5, Hotel Sher-
man, Chicago

Photographic Society of America, Annual Convention, Aug. 12-16, Hotel New Yorker,

New York

American Institute of Electrical Engineers, Pacific General Meeting, Aug. 19-22, Hotel

Westward Ho, Phoenix, Ariz.

Illuminating Engineering Society, National Technical Conference, Aug. 27-30, Wash-
ington, D.C.

International Society of Photogrammetry, Conference, Sept. 4-13, Hotel Shoreham,

Washington, D.C.

363

New Products

Further information about these items can be obtained direct from the addresses given.
As in the case of technical papers, the Society is not responsible for manufacturers' state-
ments, and publication of these items does not constitute endorsement of the products.

A new ultra-high-speed camera, de-
signed to take pictures at speeds up to
100,000 frames /sec, has been developed
at Battelle Institute, Columbus, Ohio. It
will be described by C. D. Miller of the
Battelle technical staff in a paper to be
presented before an early convention of
SMPTE. The camera is an extensively
modified version of one developed by Mr.
Miller some years ago while employed by
the National Advisory Committee for
Aeronautics. The earlier camera was
described in the November 1949 Journal
and in the reprint, High-Speed Photography,
Vol. 2.

The new camera is being used at Battelle
in studies of knock in spark-ignited piston
engines, and will be available for other
high-speed research at Battelle as desired
by industry or government. The camera
operates under conditions of steady light,
with direct photography, Schlieren pho-
tography or shadowgraphs. It operates
by optical compensation, exposes six feet
of standard 8-mm film in a single burst,
with resolution reported better than 30
lines per millimeter. Exposed film is
ready for projection as a motion picture
immediately after development, without
the need of a reprinting and registering
procedure.

Back issues of the Journal available: The following Volumes are available upon a
reasonable offer to Alfred S. Norbury, 3526 Harrison St., Kansas City 3, Mo.

Vol. 44 (Jan.-June 1945)
Vol. 45 (July-Dec. 1945)
Vol. 47 (July-Dec. 1946)
Vol. 48 (Jan.-June 1947)

Vol. 49 (July-Dec. 1947)
Vol. 50 (Jan.-June 1948)
Vol. 51 (July-Dec. 1948)

Vol. 52 (Jan.-June 1949)
Vol. 56 (Jan.-June 1951)
Vol. 57 (July-Dec. 1951)

364

A Silent Magnetic Splicer has been
developed and patented by Unusual Films
at Bob Jones University in Greenville,
S.C., which says that it is for the fast and
durable splicing of magnetic film. A
diagonal butt splice with Minnesota
Mining and Manufacturing Tape No". 41
that will outlast normal film has been
achieved. A single frame can be removed
and restored; a splice made in this manner
can be broken and put together again
without loss of a frame; and trims and
waste material can be reclaimed and used
repeatedly until too short to be of any
value. Designed specifically for magnetic
film in accordance with existing film
standards, the Silent Splicer needs no
blooping. While there is some disad-
vantage, the University says, in not being
able to see the striations, with a little
practice and familiarization with a sound
reader one can locate sync closer than half
a frame. Film must be handled carefully
and all heads must be demagnetized regu-
larly, if clear sound is to be maintained.

'Common Causes of Damage to 35mm
Release Prints" has just been issued in an
extensively revised edition by the Eastman
Kodak Company as a means of helping
laboratories, exchanges, and theaters to
keep motion picture release prints in
better condition.

The booklet discusses such possible
sources of damage as failure to provide
adequate storage facilities, improper lab-
oratory methods, inadequate inspection in
the exchanges, careless handling in the
projection room and worn or imperfectly
adjusted projectors. Also covered are
such general but equally important sub-
jects as making good splices, methods of
lubrication of release prints, directions for
determining the correct tension of pro-

Position Wanted

The Sound Splicer is designed so that
one side of the machine is for cutting of
film, the other side for registration, per-
foration and application of the tape. It
is available for 16mm film either double
or single perforated.

jector parts, and methods of making other
simple projector adjustments.

Some of the material that appears in
this new data book has been issued by
Kodak in previous booklets covering the
same general field, but all of the old
material has now been brought up to date
and a discussion of how properly to
identify the new safety base material now
used for release prints has also been added.

Written in four sections — the film, the
processing laboratory, the exchange and
the theater — and liberally illustrated with
many comparison photographs, "Common
Causes of Damage to 35mm Release
Prints" can be obtained without charge
on request to the Motion Picture Division,
Eastman Kodak Company. The data
book is punched for binding in the Kodak
Photographic Notebook.

Sound mixer and transmission engineer: 5 yr experience 35mm magnetic and optical,
16mm optical and disc recording systems. As mixer has experience stage recording and
re-recording; in transmission has installed a recording channel complete from design to
operation, also maintenance. Will accept position any geographic location. Write
L-30, c/o Fifer, 143 Church St., Phoenixville, Pa.

365

Committees of the Society

As of March 15, 1952

Administrative Committees

ADMISSIONS. To pass upon all applications for membership, applications for transfer, and
to review the Student and Associate membership list periodically for possible transfer to the
Associate and Active grades, respectively. The duties of each committee are limited to applica-
tions and transfers originating in the geographic area covered.

E. A. Bertram, Chairman, East, DeLuxe Laboratories, 850 Tenth Ave., New York 19, N.Y.

C. R. Keith W. B. Lodge L. A. Bonn

Bertel J. Kleerup, Chairman, Central, Society for Visual Education, 1345 W. Diversey Park-
way, Chicago 14, 111.

E. E. Bickel Lloyd Thompson M. G. Townsley

N. L. Simmons, Chairman, West, Eastman Kodak Co., 6706 Santa Monica Blvd., Hollywood
38, Calif.

T. T. Moulton E. H. Reichard Petro Vlahos

BOARD OF EDITORS. To pass upon the suitability of all material submitted for publication,
or for presentation at conventions, and publish the JOURNAL.

Arthur C. Downes, Chairman, 2181 Niagara Dr., Lakewood 7, Ohio

D. M. Beard A. M. Gundelfinger Pierre Mertz N. L. Simmons
G. M. Best C. W. Handley C. D. Miller R. T. Van Niman
L. B. Browder A. C. Hardy J. A. Norling J. H. Waddell

C. R. Fordyce C. R. Keith H. W. Pangborn D. R. White

L. D. Grignon G. E. Matthews

EUROPEAN ADVISORY COMMITTEES. To act as liaison between the general Society and
European firms, individuals, and organizations interested in motion picture and television
engineering. To report to the Society on such affairs in Europe, on new technical develop-
ments, and to assist the Papers Committee in soliciting papers for publication in the JOURNAL.

I. D. Wratten, Chairman (British Division), Kodak, Ltd., Kingsway, London, England
R. H. Cricks W. M. Harcourt L. Knopp A. W. Watkins

L. Didiee, Chairman (Continental Division), Association Francaise des Ingenieurs et Techni-
cians du Cinema, 92 Champs-Elysees, Paris (8e), France

R. Alia J. Cordonnier G. Mareschal J. Vivi6

R. Bocquel S. Feldman M. Terms M. Yvonnet

M. Certes J. Fourrage

FELLOW AWARD. To consider publications of Active members as candidates for elevation to
Fdlow, and to submit such nominations to the Board of Governors.

Earl I. Sponable, Chairman, Movietonews, Inc., 460 W. 54 St., New York 19, N. Y.

Herbert Barnett R. M. Corbin Barton Kreuzer V. C. Shaner

F. T. Bowditch J. G. Frayne W. C. Kunzmann E. M. Stifle

F. E. Cahill C. E. Heppberger Peter Mole

366

HISTORICAL AND MUSEUM. To collect facts and assemble data relating to the historical
development of the motion picture and television industries, to encourage pioneers to place
their work on record in the form of papers for publication in the JOURNAL, and to place in
suitable depositories equipment pertaining to the industry.

E. A. Bertram, Chairman, DeLuxe Laboratories, Inc., 850 Tenth Ave., New York 19, N.Y.

(Under Organization)

HONORARY MEMBERSHIP. To search diligently for candidates who through their basic
inventions or outstanding accomplishments have contributed to the advancement of the motion
picture industry and are thus worthy of becoming Honorary members of the Society.

Gordon Chambers, Chairman, Eastman Kodak Co., 343 State St., Rochester 4, N.Y.
Carroll H. Dunning Philo T. Farnsworth Barton Kreuzer Loren L. Ryder-

JOURNAL AWARD. To recommend to the Board of Governors the author or authors of the
most outstanding paper originally published in the JOURNAL during the preceding calendar
year to receive the Society's Journal Award.

F. J. Kolb, Jr., Chairman, Eastman Kodak Co., 343 State St., Rochester 4, N.Y.
Paul Arnold A. N. Goldsmith Joseph H. Spray

MEMBERSHIP. To solicit new members and to arouse general interest in the activities of the
Society and its publications.

A. Raymond Gallo, General Chairman, Quigley Publications, 1270 Sixth Ave., New York 20,
N.Y.

J. B. McCullough, Vice-Chairman, Motion Picture Association, 28 W. 44th St., New York 18,
N.Y.

Col. Samuel R. Todd, V ice-Chairman, Bureau of Electrical Inspection, Room 707, City Hall,
Chicago 2, 111.

L. D. Grignon, V ice-Chairman, Twentieth Century-Fox, Box 900, Beverly Hills, Calif.

Don Prideaux, Lamp Department, General Electric Company, 601 W. Fifth St., Los

Angeles 13, Calif.
John W. Duvall, E. I. du Pont de Nemours & Co., 6656 Santa Monica Blvd., Hollywood

, 38, Calif.
H. S. Walker, Chairman, 4040 St. Catherine St., W., Montreal, Quebec.

(Under Organization)

Member Delegates

V. D. Armstrong Carlos H. Elmer C. L. Jeffers W. M. Sheahan

H. C. Barr C. R. Fordyce C. R. Long John M. Sims

P. E. Brigandi D. C. Gilkeson L. R. Martin S. P. Solow

Harry Bnieggemann G. R. Groves W. C. Miller R. L. Sutton

G. A. Chambers Sol Halprin G. C. Misener J. E. Volkmann

R. W. Conant R. W. Harmon C. G. Nopper Allison V. Ziegler

J. W. Cummings Bruce Howard J. A. Ouimet

C. R. Daily B. J. Howell G. F. Rackett

A. R. Davis Hugh Jamieson J. W. Servies

NOMINATIONS. To recommend nominations to the Board of Governors for annual election of
officers and governors.

Earl I. Sponable, Chairman, Movietonews, Inc., 460 W. 54 St., New York 19

F. T. Bowditch N. D. Golden J. K. Hilliard C. H. Percy

G. L. Carrington D. E. Hyndman F. E. Cahill R. C. Warn

367

PAPERS. To solicit papers and provide the program for semiannual conventions, and make
available to local sections for their meetings papers presented at national conventions.

Edward S. Seeley, Chairman, Altec Service, 161 Sixth Ave., New York 13, N.Y.
Joseph E. Aiken, Vice-Chairman, 116 No. Galveston St., Arlington, Va.

F. G. Albin, Vice-Chairman, American Broadcasting Co., Station KECA-TV, 4151 Prospect

Ave., Hollywood, Calif.

G. G. Graham, Vice-Chairman, National Film Board of Canada, John St., Ottawa, Canada
W. H. Rivers, Vice-Chairman, Eastman Kodak Co., 342 Madison Ave., New York 17, N.Y.
G. W. Colburn, V ice-Chairman, 164 N. Wacker Dr., Chicago 6, 111.

John H. Waddell, Vice-chairman, Wollensak Optical Co., 850 Hudson Ave., Rochester, N.Y.

D. Max Beard E. W. D'Arcy L. Hughes Herbert Pangborn

A. C. Blaney W. P. Dutton P. A. Jacobson Ben Plakun

Richard Blount Farciot Edouart William Kelley Edward Schmidt

R. P. Burns F. L. Eich George Lewin N. L. Simmons

Philip Caldwell Charles Handley E. C. Manderfeld S. P. Solow

F. O. Calvin R. N. Harmon Glenn Matthews J. G. Stott
J. P. Corcoran Scott Helt Pierre Mertz W. L. Tesch
P. M. Cowett C. E. Heppberger Harry Milholland S. R, Todd

G. R. Crane J. K. Billiard W. J. Morlock M. G. Townsley

PROGRESS. To prepare an annual report on progress in the motion picture and television in-
dustries.

C. W. Handley, Chairman, 1960 West 84 St., Los Angeles 44, Calif.

J. E. Aiken T. J. Gibbons W. F. Kelley B. F. Perry

W. L. Bell G. H. Gordon R. E. Lewis E. H. Reichard

P. G. Caldwell G. R. Groves W. A. Mueller W. L. Tesch
J. W. Duvall

PROGRESS MEDAL AWARD. To recommend to the Board of Governors a candidate who by
his inventions, research, or development has contributed in a significant manner to the advance-
ment of motion picture technology, and is deemed worthy of receiving the Progress Medal
Award of the Society.

David B. Joy, Chairman, National Carbon Division, 30 E. 42 St., New York 17, N.Y.
Max Batsel F. H. Mclntosh G. H. Mitchell D. R. White

DAVID SARNOFF AWARD. To recommend to the Board of Governors a candidate who has
done outstanding work in some technical phase of the broad field of television or in any similar
phase of theater television, whether in research, development design, manufacture or operation.

Pierre Mertz, Chairman, Bell Telephone Laboratories, Inc., 463 West, St. New York 14, N.Y.
R. L. Carman T. T. Goldsmith O. B. Hanson W. B. Lodge

SUSTAINING MEMBERSHIP. To solicit new sustaining members and thereby obtain adequate
financial support required by the Society to carry on its technical and engineering activities.

Earl I. Sponable, Chairman, Movietonews, Inc., 460 W. 54 St., New York 19, N.Y.

D. B. Joy S. P. Solow

SAMUEL L. WARNER AWARD. To recommend to the Board of Governors a candidate who
has done the most outstanding work in the field of sound motion picture engineering, in the
development of new and improved methods or apparatus designed for sound motion pictures,
including any steps in the process, and who, whether or not a Member of the Society of Motion
Picture and Television Engineers, is deemed eligible to receive the Samuel L. Warner Memorial
Award of the Society

Glenn L. Dimmick, Chairman, RCA Victor Division, Front and Cooper Sts., Camden, N.J.
Lloyd Goldsmith John Hilliard John Maurer Otto Sandvik

368

Engineering Committees

COLOR. To make recommendations and prepare specifications for the operation, maintenance,
and servicing of color motion picture processes, accessory equipment, studio lighting, selection
of studio set colors, color cameras, color motion picture films, and general color photography.
(File C 1)

H. H. Duerr, Chairman, Ansco, Binghamton, N.Y.

R. H. Bingham A. A. Duryea A. M. Gundelfinger W. E. Pohl

H. E. Bragg R. M. Evans W. W. Lozier G. F. Rackett

O. O. Ceccarini J. G. Frayne A. J. Miller L. E. Varden

R. O. Drew L. T. Goldsmith C. F. J. Overhage J. P. Weiss

FILM DIMENSIONS. To make recommendations and prepare specifications on those film
dimensions which affect performance and interchangeability, and to investigate new methods
of cutting and perforating motion picture film in addition to the study of its physical properties.
(File FD 2}

E. K. Carver, Chairman, Eastman Kodak Co., Kodak Park Works, Rochester 4, N.Y.
J. E. Aiken W. G. Hill N. L. Simmons W. J. Wade

E. A. Bertram A. J. Miller M. G. Townsley D. R. White
A. M. Gundelfinger W. E. Pohl

FILM-PROJECTION PRACTICE. To make recommendations and prepare specifications
for the operation, maintenance, and servicing of motion picture projection equipment, projec-
tion rooms, film-storage facilities, stage arrangement, screen dimensions and placement, and
maintenance of loudspeakers to improve the quality of reproduced sound and the quality of
the projected picture in the theater. (File FPP 3)

R. H. Heacock, Chairman, Radio Corporation of America, RCA Victor Div., Camden 2, N.J.

C. S. Ashcraft H. F. Heidegger M. D. O'Brien Ben Schlanger

F. E. Cahill C. F. Horstman Paul Ries J. W. Servies
L. W. Davee H. T. Matthews Harry Rubin S R. Todd

FILMS FOR TELEVISION. To make recommendations and prepare specifications on all
phases of the production, processing and use of film made for transmission over a television
system excluding video transcriptions. (File FTV 4)

R. L. Garman, Chairman, General Precision Laboratory, Inc., 63 Bedford Road, Pleasant-

vffle, N.Y.

R. O. Drew Pierre Mertz R. C. Rheineck C. L. Townsend

Richard Hodgson H. C. Milholland H. J. Schlafly L. F. Transue

S. E. Howse G. C. Misener N. L. Simmons T. G. Veal

R. Johnston R. M. Morris J. G. Stott H. E. White
H. R. Lipman

HIGH-SPEED PHOTOGRAPHY. To make recommendations and prepare specifications
for the construction, installation, operation, and servicing of equipment for photographing and
projecting pictures taken at high repetition rates or with extremely short exposure times.
(File HSP 5)

H. E. Edgerton, Chairman, Dept. of Electrical Engineering, Massachusetts Institute of
Technology, Cambridge 39, Mass.

Richard O. Painter, Vice-Chairman, General Motors, Proving Ground Section, Milford, Mich.

E. A. Andres, Sr. C. H. Elmer Brian O'Brien Morton Sultanoff
H. C. Barr Eleanor Gerlach D. H. Peterson J. H. Waddell

D. M. Beard C. D. Miller Earl Quinn Roy Wolford
C. S. Eraser Eugene Miller M. L. Sandell C. W. Wyckoff

F. E. Carlson Kenneth Morgan Kenneth Shaftan

369

LABORATORY PRACTICE. To make recommendations and prepare specifications for the
operation, maintenance, and servicing of motion picture printers, processing machines, in-
spection projectors, splicing machines, film-cleaning and treating equipment, rewinding
equipment, any type of film-handling accessories, methods, and processes which offer in-
creased efficiency and improvements in the photographic quality of the final print. (File LP 6)

J. G. Stott, Chairman, Du Art Film Laboratories, 245 West 55 St., New York, N.Y.

V. D. Armstrong G. W. Colburn J. A. Maurer V. C. Shaner

H. L. Baumbach I. M. Ewig O. W. Murray J. H. Spray

D. P. Boyle T. M. Ingman W. H. Offenhauser Lloyd Thompson

0. E. Cantor P. A. Kaufman W. E. Pohl Paul Zeff
Gordon Chambers C. F. LoBalbo E. H. Reichard

MOTION PICTURE STUDIO LIGHTING AND PROCESS PHOTOGRAPHY. To make
recommendations and prepare specifications for the operation, maintenance, and servicing of
all types of studio and outdoor auxiliary lighting equipment, tungsten light and carbon-arc
sources, lighting-effect devices, diffusers, special light screens, etc., to increase the general
engineering knowledge of the art; and to make recommendations and prepare specifications on
motion picture optical printers, process projectors (background process), matte processes,
special process lighting technique, special processing machines, miniature-set requirements,
special-effects devices, and the like, that will lead to improvement in this phase of the produc-
tion art. (File MPSL 7)

J. W. Boyle, Chairman, 139? South Doheny Dr., Los Angeles 48, Calif.

Richard Blount C. W. Handley C. R. Long D. W. Prideaux

Karl Freund M. A. Hankins W. W. Lozier Petro Vlahos

OPTICS. To make recommendations and prepare specifications on all subjects connected with
lenses and their properties. (File Op 8)

R. Kingslake, Chairman, Eastman Kodak Co., Hawk Eye Works, Rochester 4, N.Y.
F. G. Back J. W. Gillon G. A. Mitchell L. T. Sachtleben

A. A. Cook Grover Laube A. E. Murray O. H. Schade

C. R. Daily J. L. Maulbetsch W. E. Pohl M. G. Townsley

1. C. Gardner J. A. Maurer

SCREEN BRIGHTNESS. To make recommendations, prepare specifications, and test methods
for determining and standardizing the brightness of the motion picture screen image at various
parts of the screen, and for special means or devices in the projection room adapted to the
control or improvement of screen brightness. (File SB 10)

W. W. Lozier, Chairman, National Carbon Div., Fostoria, Ohio

H. J. Benham A. J. Hatch O. W. Richards G. H. Walter

F. E. Carlson L. B. Isaac Leonard Satz H. E. White

M. H. Chamberlin W. F. Kelley Ben Schlanger A. T. Williams

E. R. Geib F. J. Kolb Allen Stimson D. L. Williams
L. D. Grignon L. J. Patton C. R. Underbill

16MM AND 8MM MOTION PICTURES. To make recommendations and prepare specifi-
cations for 16mm and 8mm cameras, 16mm sound recorders and sound-recording practices,
16mm and 8mm printers and other film laboratory equipment and practices, 16mm and 8mm
projectors, splicing machines, screen dimensions and placement, loudspeaker output and
placement, preview or theater arrangements, test films, and the like, which will improve the
quality of 16mm and 8mm motion pictures. (File SE 11)

M. G. Townsley, Chairman, Bell & Howell Co., 7100 McCormick Rd., Chicago 45, 111.
H. W. Bauman E. W. D'Arcy H. J. Hood A. G. Petrasek

W. C. Bowen G. A. del Valle W. W. Lozier L. T. Sachtleben

F. L. Brethauer J. W. Evans D. F. Lyman H. H. Strong

F. E. Brooker C. R. Fordyce W. C. Miller Lloyd Thompson

F. E. Carlson John Forrest J. W. Moore Willett Wilson

Carl Claras R. C. Holslag W. H. Offenhauser

370

SOUND. To make recommendations and prepare specifications for the operation, maintenance,
and servicing of motion picture film, sound recorders, re-recorders, and reproducing equip-
ment, methods of recording sound, sound-film processing, and the like, to obtain means of
standardizing procedures that will result in the production of better uniform quality sound in
the theater. (File So 12)

J. K. Hilliard, Chairman, Altec Lansing Corp., 9356 Santa Monica Blvd., Beverly Hills, Calif.

G. L. Dimmick, Vice-Chairman, RCA Victor Division, Camden. N.J.

F. G. Albin E. W. D'Arcy J. P. Livadary G. E. Sawyer

H. W. Bauman R. J. Engler J. A. Maurer W. L. Thayer

R. J. Beaudry R. M. Fraser K. M. Macllvain M. G. Townsley

A. C. Blaney J. G. Frayne W. C. Miller Harold Walker

D. J. Bloomberg L. T. Goldsmith G. C. Misener J. P. Weiss
Harry Brueggemann L. D. Grignon Otto Sandvik W. W. Wetzel

F. E. Cahill

STANDARDS. To survey constantly all engineering phases of motion picture production, dis-
tribution, and exhibition, to make recommendations and prepare specifications that may be-
come proposals for American Standards. This Committee should follow carefully the
work of all other committees on engineering and may request any committee to investigate and
prepare a report on the phase of motion picture engineering to which it is assigned. (File
St 13)

H. J. Hood, Chairman, Eastman Kodak Co., 343 State St., Rochester 4, N.Y.
Chairmen of Engineering Committees

G. L. Beers E. C. Fritts J. K. Hilliard F. J. Pfeiff
Richard Blount R. L. Garman Rudolf Kingslake J. W. Servies
J. W. Boyle F. N. Gillette W. W. Lozier J. G. Stott

E. K. Carver R. H. Heacock J. A. Norling M. G. Townsley
H. E. Edgerton

Members at Large

F. E. Carlson W. F. Kelley J. A. Maurer D. R. White

STEREOSCOPIC MOTION PICTURES. (File STE 20)

J. A. Norling, Chairman, Loucks and Norling Studios, Inc., 245 West 55 St., New York 19,
N.Y.

G. A. Chambers Clarence Kennedy K. Pestrecov J. T. Rule
J. S. Goldhammer Joseph Mahler F. A. Ramsdell W. H. Ryan
Gerald Graham Wil Marcus J. A. Ruddy R. J. Spottiswoode
L. B. Isaac

TELEVISION FILM EQUIPMENT (JOINT RTMA-SMPTE COMMITTEE). To make
recommendations and prepare specifications on all phases of film equipment as used in the
television broadcast stations. (File TVFE 14)

F. N. Gillette, RTMA, Chairman, General Precision Laboratory, 63 Bedford Road, Pleasant-
ville, N.Y.

E. C. Fritts, SMPTE, Vice-Chairman, Eastman Kodak Co., 343 State St., Rochester 4, N.Y.

F. G. Albin E. H. Lederer G. C. Misener J. H. Roe

A. J. Baracket E. C. Manderfeld R. M. Morris C.L. Townsend

P. F. Brown J. A. Maurer N. F. Oakley M. G. Townsley

Sydney Cramer H. C. Milholland R. C. Rheineck H. E. White

TELEVISION STUDIO LIGHTING. To make recommendations and prepare specifications
on all phases of lighting employed in television studios. (File TVSL 16)

Richard Blount, Chairman, General Electric Co., Nela Park, Cleveland 12, Ohio
H. R. Bell Otis Freeman R. S. O'Brien Malcolm Waring

A. H. Brolly H. M. Gurin Newland Smith W. R. Wilson

D. D. Cavelli Robert Morris Adrian Terlouw

371

TEST FILM QUALITY. To develop and keep up to date all test film specifications, and to
supervise, inspect and approve methods of production and quality control of all test films sold
by the Society. (File TFQ 16)

F. J. Pfeiff, Chairman, Altec Service Corp.. 161 Sixth Ave., New York 13, N.Y.

R. M. Corbin W. F. Kelley Joseph Spray M. G. Townsley

Russell Drew J. A. Maurer J. G. Stott

THEATER TELEVISION. To make recommendations and prepare specifications for the con-
struct ion, installation, operation, maintenance, and servicing of equipment for projecting
television pictures in the motion picture theater, as well as projection-room arrangements
necessary for such equipment, and such picture-dimensional and screen-characteristic mat-
ters as may be involved in high-quality theater-television presentations. (File TTV 17}

G. L. Beers, Chairman, Radio Corporation of America, RCA Victor Div., Camden, N.J.

Ralph Austrian N. L. Halpern P. J. Larsen Otto Sandvik

F. E. Cahill Richard Hodgson W. W. Lozier Ed Schmidt

R. L. Garman C. F. Horstman R. H. McCullough A. G. Smith

E. P. Genock D. E. Hyndman G. C. Misener E. I. Sponable

A. N. Goldsmith L. B. Isaac Harry Rubin J. E. Volkmann

E. D. Goodale A. G. Jensen L. L. Ryder

THEATER ENGINEERING. To make recommendations and prepare specifications of engi-
neering methods and equipment of motion picture theaters in relation to their contribution
to the physical comfort and safety of patrons, so far as can be enhanced by correct theater de-
sign, construction, and operation of equipment. (File TE 18)

J. W. Servies, Chairman, National Theatre Supply, 92 Gold St., New York 7, N.Y.
F. W. Alexa C. M. Cutler L. E. Pope Seymour Seider

Henry Anderson James Frank Leonard Satz J. E. Troy

Charles Bachman Aaron Nadell Ben Schlanger Emil Wandelmaier

E. J. Content E. H. Perkins

SMPTE Representatives to Other Organizations

AMERICAN STANDARDS ASSOCIATION
Standards Council, D. E. Hyndman

SECTIONAL COMMITTEES

Standardization of Letter Symbols and Abbreviations for Science and Engineering, Z10, S. L.

Chertok

Motion Pictures, PH22

D. R. White, Chairman, Photo Products Dept., E. I. du Pont de Nemours & Co., Parlin, N.J.

F.T.Bowditch H. J. Hood Pierre Mertz

Acoustical Measurements and Terminology, Z24, H. F. Olson

INTER-SOCIETY COLOR COUNCIL

R. M. Evans, Chairman, Eastman Kodak Co., 59 Kodak Park, Rochester 4, N.Y.
F. T. Bowditch H. E. Bragg A. M. Gundelfinger W. H. Ryan

M. R. Boyer L. E. Clark H. C. Harsh

UNITED STATES NATIONAL COMMITTEE OF THE INTERNATIONAL COMMIS-
SION ON ILLUMINATION

R. E. Farnham, Chairman, General Electric Co., Nela Park, Cleveland 12, Ohio
Herbert Barnett H. E. White

372

I

Techniques for Effective High-Speed
Photography and Analysis

By RICHARD O. PAINTER

High-speed photographic methods used with commercially available moving
film cameras are reviewed in this paper. Careful planning in regard to
field size, reference lines, background and timing as well as other factors
can contribute a great deal toward producing films which furnish a maximum
of useful information. Special techniques employed in analysis can likewise
simplify the task of data reduction. Some of these methods are illustrated.

H,

.IGH-SPEED PHOTOGRAPHY has been
used as an engineering tool by
General Motors Proving Ground for
over thirteen years. The experience
gained during this period has been
almost entirely with commercially avail-
able rotating-prism cameras covering
the speed range from about 150 to 15,000
pictures per second. In a corporation
as huge as General Motors and with
such diversified products, the range of
subjects to which high-speed photog-
raphy might be applied is naturally
very large. Much has been written
concerning specialized techniques for
high-speed photography such as schlie-
ren, shadowgraph, X-ray and others.
I shall not dwell on elaborate tech-
niques such as these, but will relate some
of the more common problems faced in
the application of high-speed photog-
raphy and how in our experience these
problems have been met.

A 16mm camera with a maximum
speed of about 2000 pictures per second
was used for our early high-speed photo-
graphic work. Our lighting equipment

Presented on October 16, 1951, at the
Society's Convention at Hollywood, Calif.,
by Richard O. Painter, General Motors
Proving Ground, Milford, Mich.

consisted only of conventional photo-
flood lamps in aluminum reflectors.
There was only one exposure procedure
to follow when using higher camera
speeds: "Open the lens fully, use as
many lights as space permits, and bring
them as close to the subject as possible."
It was obvious that room for improve-
ment existed in the lighting equipment.
When projector spotlights first became
available we were quick to make use of
them. Overvoltage for these units was
obtained through autotransformer boxes
provided with a switch for selecting line
voltage or high voltage, a tap switch
for adjusting the high-voltage output in
10-v steps and a meter for reading the
output voltage to the lights. These
units and the projector spotlights were
the solution to the lighting problem and
are in fact still in use for most of our work.
They are especially handy in compensat-
ing for line voltages that may be higher
or lower than normal and are now pro-
vided with relays for shunting the line
switch and throwing the lights on and
off simultaneously with the camera
operation. Experience has shown that
the light intensity builds up rapidly
enough to give adequate exposure from
the very start of the film.

May 1952 Journal of the SMPTE VoL 58

373

REFERENCE (LINE) VOLTAGES
(APPLIED TO LAMPS WHEN
LIGHT MEASUREMENTS ARE MADE) 1

Fig.

220

210

200

170

I 2 3 4 5 6 7 8 9 10 II 12 13
LIGHT INTENSITY RATIO

1. Light intensity ratios for projector spotlights, G.E. Type Par/SP-
150-w, 120-v.

374

Fig. 2. Comparison-type brightness meter.
May 1952 Journal of the SMPTE Vol.58

The same autotransformer boxes
proved to be of help in the problem of
measuring the high light intensities
involved in high-speed photography.
The ordinary exposure meter is not
usually considered of much use in
measuring high light intensities. How-
ever, with several supplementary devices,
we have used such meters with much
success. The meters are provided with
perforated screens giving a multiplying
factor of 5 on their indication. In
addition, light measurement is carried
out with line voltage applied to the
lamps. This voltage is noted on the
voltmeter mounted in the autotrans-
former boxes, and from the set of curves
shown in Fig. 1 an additional multiply-
ing factor is determined, to be applied
to the light reading when the lamps
are operated on higher voltage. This
factor usually runs between 5 and 9
times and is dependent only upon
lamp voltage at which light measure-
ment is made and voltage at which the
lamps are operated when the film is
exposed. These ratios have been care-
fully determined by measurement with
three different light-measuring devices
and have been found to be very reliable.
It can be seen that the employment of
the exposure-meter filter and light-
ratio method allows us to increase the
light intensity which can be measured
with conventional equipment by ap-
proximately 25 to 50 times. Light levels
this high are suitable for high-speed
photography up to the maximum speeds
possible with rotating-prism cameras
and allow plenty of margin so that
smaller lens openings may be used.

Light measurement is frequently made
on a substitute surface of paper having
about 25 to 50% of the reflectivity of
pure white. This is usually wise when
the subject area is largely dark but does
have small light-colored or highly
reflective sections which must not be
overexposed. Figure 2 is an interior
view of a brightness meter which is
very helpful when it becomes necessary

to measure the light level on small areas
or parts located in deep recesses where
the conventional meter cannot be used.
The subject is observed through the
upper lefthand eyepiece and is viewed
through a split-field prism having a
hexagonal pattern in the center which
receives its illumination from the light
source below through an optical wedge
or circular gradient filter. The subject
is brought into focus with the lens tube
at the right and with the current to the
lamp set to the value indicated on the
meter, the back of which may be seen
at lower center, the optical wedge is
rotated until brightness of the hexagonal
pattern and brightness of the desired
section of the field are matched. If the
subject is too bright for a match, a
multiplication of the range by a factor of
10,100 or 1000 is possible by inserting
either one or both filters located at the
upper center. The brightness level is
read through the lower lefthand eye-
piece from a scale surrounding the
optical wedge. The range of this instru-
ment is from 0.000005 to 110 c/sq in.
The highest light level which can be
measured with this instrument is there-
fore about 10 times the upper limit of
the conventional meter and with the
application of the light-ratio method
previously mentioned, the measurement
of any light level we may care to use or
be able to attain appears to be possible.
This meter was developed by Matthew
Luckiesh and A. H. Taylor of the
General Electric Lighting Research
Laboratory. A brightness meter having
somewhat similar features and manu-
factured by Salford Electrical Instru-
ments, Ltd. of Great Britain was de-
scribed in the New Products section,
page 184, of the August 1951 Journal.

There is a strong tendency in high-
speed photography to neglect the plan-
ning and preparation of the subject for
the picture to be taken. We are often
working with engineers from various
divisions on mechanisms with which we
are not familiar. It is wise in such a

R. O. Painter: High-Speed Techniques

375

Fig. 3. Commonly used lenses and extensions.

Fig. 4. Direct plotting of motions with time and motion study projector.
376 May 1952 Journal of the SMPTE Vol.58

case to determine just what parts of the
mechanism must be seen and what
information will be needed from the
film. If movements are to be measured,
a scale is necessary and if only small
movements are expected, every effort
should be made to reduce the field
covered by the camera. One basic
rule we try to follow is to restrict the
camera field to only that which is
absolutely required to show the action.
There are those who will try hard to
have the entire engine included when
only the action of a valve spring is under
study. It should be obvious to even the
most enthusiastic camera designer that
movements of a few thousandths of an
inch cannot be shown in a field five
inches wide and yet just that sort of
thing is expected by many people who
want high-speed motion pictures taken.

To take care of the various field sizes
which may be required when camera-to-
subject distances are not a matter of
choice, we use a variety of lenses and
lens extensions. These are shown in
Fig. 3. At the bottom of the picture
are lens-plate shims of various thick-
nesses used to permit lens mounting at
distances from film between normal and
that possible with the shortest bayonet
extension seen above to the right. In
the lower righthand corner of the
picture is a bayonet adapter used to
mount the lenses shown on another
camera not so equipped originally.
Lenses shown are 254-mm, 105-mm,
2-in. and 35-mm. Although many
other lenses are available for these
cameras, we have found that those
shown take care of a very high percentage
of our work adequately. Above the
lenses are extensions of various lengths,
some variable, some fixed. Any exten-
sion up to about 12 in. can be obtained
by using these tubes singly or in combi-
nation. The use of a long extension
in connection with the 254-mm lens
will provide as much as 5 times magni-
fication of the image on 16mm film.
Such magnifications have been useful

in photographing contact action, small
vibratory movements and spot and
projection welding. When close-up
photography is being performed, it
is of course necessary to correct the lens
aperture used for the amount of lens
extension employed as this might other-
wise reduce the exposure considerably.

When a subject is being prepared for
high-speed photography it is well to
plan beyond the basic uses to which the
film is to be put and anticipate the need
for additional information from the
pictures. It is always wise to provide
an accurate time base for the film either
in the field covered, if frequent reference
must be made to it, or at least along the
edge of the film as furnished by the
timing units provided for these cameras.
Dimensional information should be given
somewhere in the field of view by scales
located close to the plane of motion or
by marks placed a known distance apart
on the subject. If the motion of a part
of the subject relative to a stationary
object or to another part is to be deter-
mined, a scale should be appropriately
mounted to show this motion directly.
Often the only requirement of a film
is to show the character of a motion in
comparing various conditions; however,
the provision of scales and timing for
more specific evaluation is usually good
foresight.

Figure 4 shows a time and motion
study projector being used to plot the
action of a high-speed camera subject.
Such projectors designed for time and
motion study of production operations
by methods-engineering and standards
groups have features which suit the job
of analyzing our films very well. This
machine projects a very bright clear
single frame and has a hand crank which
moves the film four frames per turn.
In addition a frame counter is provided
which can be reset to zero on any frame
desired. This makes plotting on any
desired picture interval easy, allows a
precise accounting of motion relative
to frame count and simplifies the de-

R. O. Painter: High-Speed Techniques

377

Fig. 5. Camera and mirror mounted to photograph suspension action on car.

Fig. 6. Underside of car showing mirror and lights used to photograph
suspension action.

378 May 1952 Journal of the SMPTE Vol. 58

termination of picture-taking rate. This
projector also features governor-con-
trolled projection rate which is adjust-
able from 800 to 1200 pictures per
minute. This feature makes it possible
to quickly make time measurements
directly from the film. This is done by
adjusting the projector speed so that
the elapsed time between timing marks,
or an interval on the time scale as
checked by a stopwatch, is a convenient
multiple of the actual time represented.
It is then a simple matter to use the
stopwatch to check time in the motion
of the subject and divide this by the pre-
determined multiple to get true time.
Reasonable accuracy may be obtained
in this manner provided reference is
made to the time base along the same
section of film as is being studied.

The arrangement in Fig. 4 is used to
plot subject motions directly on graph
paper. The projection distance is ad-
justed to make the dimensions scale
1 : 1 or some convenient ratio. In this
case all motions of interest were hori-
zontal and timing marks were provided
on the film at 0.001 -sec intervals.
Horizontal motions were plotted against
time on the vertical scale and the
projector elevating mechanism was used
to position the picture for plotting while
a stationary reference line on the subject
was used in following one of the vertical
chart lines.

The photography of moving vehicles
and various components on them such
as suspension members, accounts for
about 20% of our film footage. Figure
5 shows a camera mounted in the trunk
of a car using a mirror to obtain a view
of the rear spring and shock absorber.
Mirrors can be very useful in getting
views from otherwise difficult angles.
Mirror shake usually gives no trouble
since at the frame speeds employed it
results in only a slow drift in the pro-
jected picture. Figure 6 is a view
underneath the same car with the gaso-
line tank removed to allow a better
view of the subject. The lights used in

this case are fog-lamp units having an
elongated beam of light quite suitable
for illuminating the spring. Batteries
were used to power the lights and an
overvoltage of about 75% was applied
to increase the illumination intensity.
Camera power was likewise derived
from batteries, although a motor gener-
ator a-c source is frequently used for
this purpose.

Another method of photographing
moving vehicles which we use involves
setting up the camera alongside the
roadway and either following the vehicle
by panning the camera or allowing the
vehicle to pass through the field of the
camera. Still another method em-
ployed with success involves a car to
carry the camera and run alongside the
subject vehicle while the picture is
taken. The camera is sighted on the
subject through the use of a simple
sighting arrangement and the camera
operation is initiated when the subject
is properly centered in the sight. This
method of taking moving-vehicle films
has been especially successful and since
the time involved in running the film
through the camera is only a few seconds
no great difficulty exists in keeping the
subject in view once it is lined up.

At one time it was desired to photo-
graph the action of a tire tread as a
car was accelerated from a standstill.
The best view and perhaps the only
view possible could be taken from just
one position, underneath a plate-glass
surface on which the tire rested. Figure
7 shows the tire on a heavy piece of
plate glass. The camera setup below
the plate glass is shown in Fig. 8. Al-
though the coefficient of friction be-
tween plate glass and rubber was
naturally somewhat different from that
between normal road surfaces and
rubber, the general behavior of the tire
tread was quite informative.

The application of high-speed photog-
raphy to the study of combustion in a
gasoline engine is by no means new;
however, steady progress has been made

R. O. Painter: High-Speed Techniques

379

Fig. 7. Wheel on plate glass for tire
tread photography.

Fig. 8. Camera and lights arranged to
photograph tire tread action.

in the methods used in taking pictures
of this sort. Rassweiler and Withrow
of General Motors Research Labora-
tories started on the problem of combus-
tion photography well over fifteen
years ago and a number of papers on
their methods have been presented
before various engineering society meet-
ings.1 This work has been continued
right up to the present time under Mr.
Withrow's direction. Figure 9 shows
one of the quartz-window engines being
used for combustion study. One of
the principal aims of this work has been
to obtain films of the progress of the
flame front during both knocking and
nonknocking combustion and to observe
carbon particle formation. For this
work it is not considered desirable to
use fuel additives for the purpose of
increasing flame brightness, therefore
camera frame rates have been held to
around 1500/sec or 2000/sec in the
interest of good exposure. Sensitive
films such as linagraph pan are usually
employed and fast camera lenses are
required. Lenses of either //1. 9 or
//1. 5 aperture rating are usually used.
In Fig. 9 three light sources can be seen.
These are ultraviolet lamps which have
replaced ordinary incandescent illumi-
nation for delineating the cylinder
outline. The use of visible light for
illumination tended to wash out the ex-
posure of the cylinder area due to the
light reflected from the combustion
chamber. Now used is an outline mask
to which have been applied fluorescent
coatings which glow brightly under
ultraviolet light. The value of these
films depends upon precise determina-
tion of crankshaft angle for each frame.
This information is provided by the
crankshaft position scale seen at the
lower center part of Fig. 9. Just above
the scale is an optical arrangement
which places an aerial image of the scale
in the plane of the quartz window.
Figure 10 is a view of the quartz window
and crankshaft scale as seen from the
high-speed camera position. The cylin-

380

May 1952 Journal of the SMPTE Vol. 58

der outline mask has been coated with a
variety of fluorescent materials such as
vaseline, solium and coatings known as
"da-glow," "glo-craft" and others. The
various materials on the mask were being
compared for intensity of fluorescence.
It is interesting to note that in the first
use of ultraviolet light on the mask
alternating-current sources were em-
ployed and the cyclic variations in mask
illumination were quite disturbing.
Direct-current units are now used to
supply the ultraviolet light.

The action of the subject for a high-
speed motion picture may have asso-
ciated with it certain electrical or other
phenomena which would add much to
the picture if simultaneously recorded.
In some cases this information can be
conveyed through the use of the timing
lamp, but often is too complex for re-
cording by this means. Transducers
such as microphones, vibration pickups
or strain gages, can be of use in recording
this other information on the film. A
method we have used to perform simul-
taneous photography of a subject and
associated characteristics is shown in
Fig. 11. A cathode-ray oscillograph
is used to convert the electrical informa-
tion to a trace suitable for photography.
A small first-surface mirror is positioned
so that it combines the oscillograph
trace with a view of the subject in the
camera field of view. Camera-to-sub-
ject distance and camera-to-oscillograph
screen distance via the reflected path
must be made equal so that a sharp
focus may be had on both. Modern
cathode-ray oscillographs can provide
a trace sufficiently bright for good
exposure when using a high-speed camera
of the commercial variety at maximum
frame speed. The subject of the setup
in Fig. 1 1 was the contact action of an
auto radio vibrator and the input cur-
rent waveshape was displayed on the
oscillograph. There are, of course,
other methods of showing electrical
variations along with mechanical action;
however, it is felt that the versatility,

sensitivity and ease of calibration as
well as trace-positioning simplicity and
flat response of modern cathode-ray
oscillographs should be fully considered
in this application.

The study of valve spring and valve
motions in engine work is simplified
greatly through the use of high-speed
photography. Figure 12 shows equip-
ment used to measure exhaust valve
motion on an overhead valve engine.
This procedure is similar to one used by
Thompson Products Co., and a more
specific description of the results achieved
is contained in a paper by Thoren,
Engemann and Stoddart, presented
before the Society of Automotive Engi-
neers.2

In the illustration shown, an 8mm
high-speed camera was used and
mounted horizontally so as to take ad-
vantage of the 16mm frame width
available when using a special aperture
plate. Camera speeds employed were
as high as 15,000 frames/sec at higher
engine speeds. A vernier indicating
device was rigidly mounted to the
engine head and the sliding scale which
was very light in weight was welded to
the valve-spring retainer cap. The
resulting pictures of the vernier could be
read easily to 0.001 in. of valve move-
ment. The fixed vernier scale was
provided with two holes through which
the flash tube of a Strobolux flash unit
could be seen from the camera position.
This flash was triggered from a contact
arranged to close at top center for the
cam of the valve being studied. In-
terpolation of the film for camshaft
angles between flashes was accom-
plished by using elapsed frames as a
basis. Figure 13 shows valve-motion
curves obtained by this method for an
engine speed of 4600 rpm. This repre-
sents an extreme condition rarely en-
countered in actual service. The solid
line represents the ideal valve-lift curve.
The curve traced in long dashes shows
the motion with a solid valve lifter
and it should be noted that valve open-

•y ' R. O. Painter: High-Speed Techniques

381

382

May 1952 Journal of the SMPTE Vol. 58

Fig. 11. Setup for simultaneous photography of contact action and
oscillograph trace of waveform.

Fig. 12. 8mm camera and equipment used to photograph engine valve motion.
R. O. Painter: High-Speed Techniques

383

0

\B

'"

— -

-^

'

.

s

0* 140* 160* I80» 200* 220*

//

^ "

—x

N^

\

//

/

\

\\

*-—~+

r.

SOLID L

\

.IFTI

iR ^

»600 RPM

1 1

/

\

r

HYC

I. LI

-TE

* 4C

00

RPV

y

\\

/

1

1

\

^

//

\

S*

"^v

k

r^

'

\\

Xj

/^'

-*•""•"•

"\

\

.

V

v>>-

-''

N,

K»

s ,

80'

60«

40'

20°

0- 20° 40°

CAM DEGREES

60'

80«

100'

120*

Fig. 13. Valve motion curves for 4600 engine rpm as plotted from film.

ing greatly overshoots the ideal curve
at top center and the valve does not
follow the calculated curve except as it
begins to open. It is interesting to
note that the valve goes through about
four bounce cycles upon closing before
it settles down. The curve traced in
short dashes is the one obtained with a
hydraulic valve lifter and it is to be
noted that even in this extreme condi-
tion it more closely follows the ideal
curve with less overshoot and evidently
with less impact and bounce upon
closing. The departure of measured
valve motion from the ideal curve is
caused by forces and elastic deflections
within the valve train at higher engine
speeds. This method of checking valve
motion is used to verify improvements in
valve motion brought about by a
method of designing cam contours to
allow for the dynamic forces in the valve
train at high speeds. Engineers familiar
with valve-train design say that no other
method is available for checking valve

motions as accurately as through the
use of high-speed motion pictures.

The methods described here and
many others, have been used by General
Motors Proving Ground over the past
thirteen years with excellent results.
They have been instrumental in many
product improvements and have pointed
the way in a number of instances to
more dependable designs and more
efficient production operations. High-
speed photography seems destined to
play an ever-increasing part in our
design and experimental activities in
the future.

References

1. G. M. Rassweiler and L. Withrow,
"High speed motion pictures of engine
flames," Ind. Eng. Chem., 28: 672-677,
June 1936.

2. T. R. Thoren, H. H. Engemann and
D. A. Stoddart, "Gam design as related
to valve train dynamics," S.A.E. Trans-
actions, 6: 1-14, Jan. 1952.

384

May 1952 Journal of the SMPTE Vol. 58

A Direct-Projection System
For Theater Television

By FRANK N. GILLETTE

The Simplex Model PB-600 contains all facilities required for operation from
any of the standard sources of television signal. It projects on the theater
screen a full-size television picture of the highest quality compatible with
the present state of the art. The system combines simplicity in installation,
convenience in maintenance and reliability in operation. This paper de-
scribes the circuitry and mechanical features of the system components, and
presents the design considerations which governed the development.

-L HE SIMPLEX system consists of three
compact units: a control-panel cabinet,
a high-voltage supply and an optical
barrel. The installation location of these
units is indicated in Fig. 1 .

The control-panel cabinet, Fig. 2,
contains all of the operating controls and
the large majority of alignment and serv-
ice controls. This unit would nor-
mally be installed in the projection
booth, but in the case of a crowded
booth, many alternate locations are pos-
sible.

The high-voltage supply, Fig. 3, can be
installed in almost any convenient loca-
tion. It has no controls, meters or
switches mounted on it and should re-
quire no attention for months at a time.

The optical barrel, Fig. 4, does have
some critical installation requirements.

Presented on October 15, 1951, at the
Society's Convention at Hollywood, Calif.,
by Frank N. Gillette, General Precision
Laboratory, Inc., 63 Bedford Rd., Pleasant-
ville, N.Y.

Since the projection optics have a fixed
focal length and have, moreover, an ex-
tremely wide aperture, the optimum
location for the barrel is fixed, within
rather narrow limits, by the size and
location of the projection screen.

The projection screen itself may be re-
garded as a component of the theater
television system or as a part of the exist-
ing theater equipment. The screen
should be selected for optimum reflection
of incident light into the audience area.
This might occasionally dictate the use
of a beaded screen, but only if the theater
is quite narrow. With a wider theater,
beaded screens would probably be re-
jected because of their very poor per-
formance at large reflection angles.

In many installations it will be ad-
visable to use the screen already present.
This screen will presumably have been
selected because its reflection properties
are suitable for the shape and size of the
theater. Of course, the existing screen
may be too large, requiring masking to a

May 1952 Journal of the SMPTE Vol. 58

385

CONTROL PANEL CABINET

I VOLTAGE SUPPUT

0

Fig. 1. Equipment locaton,
Simplex Model PB-600.

386

Fig. 2. Control-panel cabinet, Simplex Model PB-600.
May 1952 Journal of the SMPTE VoL 58

Fig. 3. High-voltage power supply,
Simplex Model PB-600.

smaller size during periods of television
presentation.

The optical system is designed to pro-
vide a picture 1 5 ft high and 20 ft wide at
a throw distance of 62 ft. The system
does permit some variation of picture
size and throw distance, but there are
also some unyielding restrictions on such
variation. The many inquiries we re-
ceive on the subject of picture size and
throw distance indicate the advisability
of calling specific attention to the nature
and source of these limitations.

Figure 5 shows the optical elements of
the system. The picture is formed on the
face of the cathode-ray tube shown at T.
Light from the tube face is collected by
the mirror at M and directed toward the
projection screen at S. The corrector
plate is inserted at P for the purpose of
correcting aberrations, principally the
spherical aberration of the mirror.

The design of the entire optical system
is fundamentally controlled by the cath-
ode-ray tube, in this case a Type 7NP4.
For good focus over the entire picture
area, it is necessary that the curve of the
mirror be essentially concentric with the
curve of the tube face. It is further
necessary that the tube face be located

Fig. 4. Optical barrel,
Simplex Model PB-600.

approximately at the focal point of the
mirror. Since the focal length of a
spherical mirror is equal to one-half its
radius of curvature, the foregoing condi-
tions result in a mirror having a radius of
curvature twice that of the cathode-ray
tube and a system having a focal length
equal to the radius of curvature of the
tube face.

With the focal length fixed in this way,
there is then a single value of magni-
fication for any chosen throw distance.
Thus, picture size at a fixed throw dis-
tance can be changed only by changing
the size of the picture on the cathode-
ray tube. If the size is increased too
much, the corners will be clipped by the
edge of the tube. If the size is decreased
appreciably, resolution will suffer. In
practice, the dimensions of the picture
can be varied some 10% either way from
the nominal size.

The magnification is, of course, a
linear function of the throw distance, but
throw distance is not readily controlled.
Throw distance is strongly influenced by
the design of the theater and can be ma-
nipulated only by reconstruction of a more
or less extensive nature. If the pre-
ferred installation location provides a

F. N. Gillette: Direct-Projection Theater Television

387

Fig. 5. Optical elements, Simplex Model PB-600.

40' THROW

\

/

\

388

Fig. 6. Effect of focal-length variation, Simplex Model PB-600.
May 1952 Journal of the SMPTE Vol. 58

throw that is too short, use of a smaller
screen is possible and provides the at-
tendant advantage of increased screen
brightness. If the preferred location
gives a throw that is too long, the only
answer is theater modification. Increas-
ing the screen size is not recommended
because the brightness soon becomes un-
acceptably low.

Theater people are quite familiar with
these relationships between picture size,
throw distance and focal length. Un-
fortunately, they are also accustomed to
purchasing projection lenses in many
different focal lengths scattered over such
a wide range as to satisfy almost any re-
quirement of picture size and throw dis-
tance. Quite naturally, they expect to
find a similar flexibility offered in theater
television equipment.

The cost of designing and stocking ex-
pensive optical systems of different focal
lengths is one obvious reason for not of-
fering such flexibility. Other, and per-
haps even more forceful, reasons are in-
dicated in Fig. 6 which shows the effect
of focal-length variation. The middle
drawing, illustrating the 60-ft throw,
shows the components of our present
optical system; the upper drawing shows
a system using a shorter focal length;
the lower drawing shows a system using a
longer focal length. As drawn, the three
systems provide approximately the same
screen brightness.

It will be noted that the diameters of
the optical elements of the system of
longer focal length are considerably
larger than the elements of the "Sim-
plex" system. Not only are such ele-
ments much more expensive than those
used in the present system, they are also
larger than can be manufactured in
quantity by existing equipment. The
system of shorter focal length involves
smaller components which could indeed
be manufactured at reasonable cost.
However, the angular width of the pic-
ture becomes significantly greater. As
this angular width becomes larger, the
optical design problem becomes tre-

mendously more complex. Adequate
correction of optical aberrations in the
corners of the picture becomes virtually
impossible.

Although an optical system of this
type is generally called a Schmidt system,
it differs tremendously from the system
originally developed by Schmidt for use
as an astronomical telescope. Funda-
mentally, a Schmidt system consists of a
spherical mirror, a diaphragm located
at the center of curvature of the mirror,
and a corrector plate also located at the
center of curvature. The diaphragm
serves to eliminate third-order aberra-
tions and the corrector plate provides
compensation for spherical aberration.
The optical quality of this system can in-
deed be very good, provided the design is
restricted to an angular field of some-
thing like 1 ° and an aperture less than
//3. For the "Simplex" system, an
angular field of 23° and a geometrical
aperture of // '0.7 is required. Clearly,
these requirements are well beyond the
limitations of the basic Schmidt de-
sign.

The classical Schmidt formulas have
been applied to the present conditions
with a reasonable degree of success.
However, much better results have been
obtained by approaching the design
problem from a somewhat different
point of view. Louis Raitiere of the
General Precision Laboratory has suc-
ceeded in developing a design approach
which results in a system that differs
slightly, but very significantly, from the
classical Schmidt system. The per-
formance obtained with Raitiere's optical
system has been quite gratifying. A
limiting resolution in the extreme corner
of the field of 2000 television lines per
picture height is observed. This figure,
of course, applies to the optical system
alone and not to the overall system.

The detail-contrast ratio that can be
obtained in any system which works with
a cathode-ray tube as the basic picture
source is never as much as one would de-
sire. The contrast ratio is still further de-

F. N. Gillette: Direct-Projection Theater Television

389

graded by the presence of any dirt on the
optical elements of the system. To re-
duce the rate at which dirt collects on the
optical elements, and consequently to
minimize the necessity for frequent clean-
ing, the optical barrel, shown in Fig. 4,
has been designed. The barrel is com-
pletely enclosed and there is no circula-
tion of outside air through the system.
The cooling air which must be directed
against the face of the cathode-ray tube,
to avoid damage to the tube, is recircu-
lated through the barrel and serves only
to conduct heat from the cathode-ray
tube to the outer walls of the barrel. The
outside of the barrel provides such a
large radiating surface that the resulting
temperature rise is insignificant.

The use of a closed system also permits
quite simple solutions to any problems
arising from excessive humidity. Thus
far no difficulty with arc -over within the
barrel has been encountered, but should
such difficulty develop, no trouble in con-
trolling the humidity within the unit is
anticipated.

The barrel is supported mechanically
at three points. The two pivot points are
located at approximately the center of
gravity and carry the bulk of the weight
of the unit. The third support point is at
the bottom of the front of the barrel. Its
function is to tilt the barrel and to hold
the line of sight, once it is established.
The maximum tilt, permissible from op-
tical considerations, is approximately 7°.
If it is possible to tilt the screen, a greater
tilt of the barrel can be accommodated by
the mechanical adjustment provided.

The barrel opens at the top for clean-
ing and service. The video amplifier and
the alignment controls are located here,
which makes readjustment or tube re-
placement very simple. Of course, there
are no more tubes in the barrel than is
absolutely necessary. Only the final
video amplifier is located here.

The cathode-ray tube is mounted in
the deflection yoke which is in turn held
by a support arm that hangs from the top
of the barrel. The support arm fastens

to a mounting plate from which it can
easily be removed and to which it returns
without disturbance of previously made
alignment adjustments. All alignment
adjustments required by tolerances of the
cathode-ray tube itself are made on the
support-arm assembly. Thus, any oper-
ator who may wish to do so can equip
himself with a spare tube support arm in
which he can mount and align a spare
cathode-ray tube to have it in instant
readiness for replacement in case of tube
failure. To facilitate this operation, all
electrical connections to the cathode-ray
tube and the deflection yoke are carried
up the tube support arm to connectors
that can be quickly disconnected in time
of need. With these provisions, a show
need not be lost for more than three
minutes by failure of the cathode-ray
tube.

The 80-kv power supply is shown in
Fig. 3. This unit provides the anode
voltage for the cathode-ray tube and also
the focus voltage.

The circuit employs a 60-cycle volt-
age doubler using two Type VR3B recti-
fiers. The output voltage is regulated
against variation in both line voltage and
load current by an electronic regulator
which controls a saturable reactor in
series with the primary of the high-volt-
age transformer. The regulation charac-
teristic is essentially flat from zero current
to 2.5 ma. Beyond 2.5 ma, the voltage
drops rapidly with increasing current in
the manner required for protection of
equipment against permanent damage in
case of momentary failure.

The focus voltage is bled from the 80-
kv level to take advantage of the stability
of that level and to provide a focus volt-
age that will remain proportional to the
anode voltage, should any variation in
that level occur. Remote control of the
focus voltage is provided by a high-volt-
age triode used as a shunt across the low
end of the focus bleeder.

The unit is oil filled for maximum re-
liability. It also contains a number of
electrostatic shields and protective spark

390

May 1952 Journal of the SMPTE Vol. 58

1

Fig. 7. Control-panel cabinet servicing, Simplex Model PB-600.

gaps on the low-voltage wiring to ensure
that any breakdown which might occur
inside the unit will have no harmful ef-
fect on external circuits.

The booth equipment consists of the
Control-Panel Cabinet, shown in Fig. 2.
It is a double-relay rack, each rack being
of the standard width to accommodate
19-in. panels.

The rack itself possesses a number of
special features that deserve mention.
The component chassis are strictly con-
ventional, each one consisting of a hori-
zontal chassis with a vertical front panel
of standard 19-in. width. However, the
method of mounting is such as to provide
much greater serviceability than is usu-
ally found in equipment constructed in
this fashion. Each individual chassis is
held in place with two quarter-turn
locks . When these are ,<,£ eleased, the
chassis may be drawn forward on rollers

until it is fully clear of the rack. This
provides quick access to all of the tubes
in the rack without the removal of cover
plates or other ornamentation (see Fig.
7).

Should the wiring side of the chassis re-
quire attention, it is necessary only to lift
the front of the chassis and swing it up-
ward through 90° where it will rest in a
stable position with the wiring facing out-
ward. In either of these positions the
chassis is still connected and still operat-
ing.

These provisions make it possible to
perform all service functions without ac-
cess to the rear of the rack. This same
thought has been carried further. When
all the chassis are removed from the rack,
there remains but an empty shell. As a
first step in installation, this shell can be
bolted, once and for all, in its final posi-
tion even though this places the back of

F. N. Gillette: Direct-Projection Theater Television

391

1

Is

8 *
o 1**

h O

.ss

§ M

8 ^

'H,
SJ

9 CA
^O

£

00

t>

:.,iv.

11

392

May 1952 Journal of the SMPTE Vol.58

the cabinet solidly against a wall. The
conduits and cables can then be affixed
and the chassis installed without further
movement of the cabinet.

The equipment in the racks is so dis-
tributed as to place the monitors a.nd
meters at eye level and the operational
controls at convenient finger-tip level.

The unit in the upper left corner is the
Picture Monitor, containing its own
power supply. The controls on this unit
serve only to adjust the picture on the
8|-in. monitor tube (see Fig. 8).

Below the monitor is a receiver of
rather superior characteristics which pro-
vides off-the-air reception during periods
of test and alignment (see Fig. 9) .

The two units below the receiver are
the vertical and horizontal deflection
chassis (see Fig. 2). These units contain
all of the deflection controls, circuits and
components except the deflection yoke,
which is necessarily located with the pro-
jection tube in the optical barrel. The
deflection circuits and components are
especially designed to permit a long cable
connection to the deflection yoke. With
the cable usually provided, this run can
be 150 ft. With special low-capacity
cable, even longer runs are possible.
This point is mentioned particularly be-
cause this cable run is the only one in the
system that bears any restriction as to
length.

Below the deflection chassis are two
blank panels behind which is located a
line-voltage regulator that stabilizes the
input voltage to various circuits that are
not sufficiently critical to demand elec-
tronic regulation, and to the filament
transformers of more critical circuits.

The bottom panel of the right-hand
rack is also blank. In this space the
saturable reactor, which regulates the 80-
kv supply, is mounted.

The two chassis directly above contain
power supplies which provide the various
plate and bias voltages required by all of
the circuits except the monitors.

Above the power supplies the Hi-
Voltage Control Unit is located. This

unit contains all of the low-voltage ele-
ments associated with the 80-kv supply
except the saturable reactor mentioned
earlier. The panel controls consist of
pushbuttons for controlling power to the
supply and a knob for setting focus-
voltage level.

The remaining two panels in this
rack are shown in more detail in Figs. 10
and 11. Figure 10 shows the program-
selector panel located immediately above
the high-voltage control unit. All of the
signal switching and audio-control func-
tions of the equipment are performed at
this panel.

The system provides for three incom-
ing program channels, each consisting of
an audio and a video line. Normally,
one of these channels will be connected
to the receiver included in the equipment.
The second will take the incoming pro-
gram line. The third might be used for a
parallel safety channel for the main pro-
gram line, for an auxiliary microwave re-
ceiver or possibly for a local signal gener-
ated by pick-up equipment within the
theater.

The switching facilities permit inde-
pendent monitoring of any incoming
audio or video line. The three panel
pushbuttons at the upper left connect
any of the three audio lines to the monitor
headphone jack in the lower right corner
of the panel. The gain control for the
monitor channel is adjacent to the phone
jack. The larger knob sets gain in the
program line to the theater. This con-
trol is used only to set the audio level
from the television equipment to the
level required by the input of the theater
sound system. It is not considered an
operational control.

The pushbuttons at the upper right
switch the input signal to the Picture
Monitor and also to the Waveform
Monitor, yet to be described. The first
three buttons select any of the three
incoming video lines. The fourth
button is labeled "Screen." Its function
will be described in connection with the
Projector Control Panel. An additional

F. N. Gillette: Direct-Projection Theater Television

393

nonlocking pushbutton, located nearby
and labeled "Push to Calibrate," serves
to connect a calibrating signal to the
Waveform Monitor for use in setting
signal levels.

Program switching is done by the
pushbuttons at the lower left of the
panel. They feed any of the three input
channels to the theater system, control-
ling both picture and sound. Interlocked
switching has been used here as another
means of eliminating operator error.

Figure 1 1 shows the Projector Control
Panel which is located at the top of the
right-hand rack. On the Projector
Control Panel are concentrated all but
one of the operational switches and
controls normally used in turning on
and adjusting the theater screen picture.
This panel also contains the waveform
monitor and a multipurpose meter,
both very useful as monitors during
projection and as test instruments during
preliminary setup.

The equipment will normally be
turned on by the four-step procedure
outlined below. However, the equip-
ment contains enough interlocks and
protective circuits to ensure that no
damage will result no matter how care-
lessly the operator handles his turn-on
procedure. In any case the progress
of the operation is indicated by the
condition of four amber lights at the
upper right of the Control Panel. Not
until all four are illuminated will a
picture appear on the projection screen.
As the first step of the procedure, the
main power relay is closed by means
of the motor-starting-type pushbutton
located below the meter. This causes
immediate glowing of the first of the
amber lights.

While the circuits warm up and reach
normal operating conditions, the second
amber light begins to glow, indicating
the presence of deflection fields at the
cathode-ray tube.

As the second step, power is applied
to the Hi -Voltage Supply by means of
the pushbuttons located on its control

panel. In a short time the anode
potential rises to its proper level, causing
the third amber light to glow.

At this point, the entire system is
turned on and the three lights inform
the operator that all interlocks are
closed, all supply voltages are present
and most of the circuits are functioning
in essentially normal fashion. However,
there is still no picture on the projection
screen because the projection tube is
biased well beyond cutoff.

The third step might be considered
optional, but is actually essential to
good showmanship. It consists of using
meter and monitors to preset various
controls to ensure that the picture first
seen by the audience is a good picture.

The test meter is used first to check
the levels of the various supply voltages,
including the 80-kv anode supply. It
is then used to set the operating bias of
the projection tube at the proper level
by means of the "Brightness" control.
Finally, it is turned to the "2MA"
position to serve as a monitor during
the projection period.

The monitors, both picture and sound,
are used first to check on the quality
and the levels of the incoming signals.
Then the picture and waveform monitors
are switched to their "Screen" position.
In this condition, both receive a video
signal brought back from the final
video stage in the optical barrel, which
permits preliminary adjustment, by
means of the "Contrast" control, of the
actual driving signal applied to the
cathode of the cathode-ray tube.

Furthermore, when the Picture Moni-
tor is switched to the "Screen" position,
its horizontal and vertical sweeps are
synchronized directly by pulses obtained
from pickup coils wound into the de-
flection yoke of the projection cathode-
ray tube. Since these pulses are actually
a measure of the magnetic deflection
fields applied to the projection tube, a
normal picture on the Picture Monitor
is a positive indication that the deflection

394

May 1952 Journal of the SMPTE Vol. 58

signals, applied to the projection tube,
have the correct frequencies and essen-
tially the correct amplitudes.

All is now ready for the fourth step.
Turning the "Picture" switch to "On,"
illuminates the fourth amber light,
switches the projection tube from cutoff
to operating bias and presents the picture
in essentially perfect adjustment.

The cathode-ray tube used in this
system is rather expensive and, while
it is an amazingly tough device when
treated properly, it is highly fragile
when mistreated. These remarks apply
equally well to personnel who operate
and maintain the equipment. Conse-
quently, the equipment includes an
elaborate system of interlocks and safety
devices for protection of tubes and
personnel.

The interlock system prevents the
application of primary power to the
high-voltage supply unless all doors
giving access to the anode and focus
voltages are closed and all chassis in the
rack are in place. The protection
system allows beam energy to reach the
tube face only when the following
conditions are satisfied:

1. Proper voltage levels exist in the
+750-, +400-, +285-, -105- and
— 150-v power supplies.

2. The 80-kv supply is up to operating
level but not in excess of 82 kv.

3. Horizontal and vertical deflection
fields have at least 75% of their normal
amplitudes.

4. A substantial stream of air is blow-
ing against the tube face.

The projection system is designed
primarily to prevent damage to the
7NP4 projection tube, but it also serves
to protect the remainder of the system
against bias failure.

Throughout the system, the protection
circuits have been designed to work
directly from the critical quantity and
not from signals which usually, but not
always, denote that quantity. For
example, the circuit which protects
against sweep failure might work with

almost complete safety from various
currents or voltages that are readily
available in the deflection circuits.
Actually, in this case, the critical
quantity is the magnetic field in the gap
of the deflection yoke. Our protection
system includes pickup coils in the yoke
which measure the magnetic fields and
thus give positive and complete protec-
tion against sweep failure.

In designing the system described
above, the goal of providing quality of
performance, exceeding the require-
ments of present television standards,
has been pursued. The degree of
success that has been achieved justifies
a prediction that the Simplex equipment
will not be found wanting whenever
higher performance standards may be
adopted.

Discussion

Anon: You mentioned that there was
no restriction on the physical separation
of the various components in this system
with the exception of one point that I
don't recall now. How about the distance
between the 80-kv power supply and the
optical barrel? Won't you get into trouble
there with high capacity if you have that
distance too great?

F. Ji. Gillette: In that case you're not
in any trouble because of high capacity.
Indeed, the cable capacity can be used as
an essential and valuable part of the
filter system on the 80-kv. If we were
actually obliged to use a very short cable
we would find it necessary to add capacity
in the power supply. We've left space
for this purpose in case we hit such an
emergency, but we're definitely planning
on the capacity of the cable as part of the
filter system.

Anon: That isn't the point l had in
mind. The larger the capacity in the
output circuit of the high voltage power
supply the more lethal the thing becomes.
Dr. Gillette: The thing is lethal without
any doubt. I don't think it's possible to
reduce the capacity to a value which would
not be lethal. The only remedy seems

F. N. Gillette: Direct-Projection Theater Television

395

to be to prevent access to the voltage. such a system. We have operated the

From the point at which it emerges from system in that fashion just to find out how

the supply itself, until it disappears into much further we have to go and we are

the barrel, the high voltage lead is encased satisfied that we need considerably more

in the outer conductor of a coaxial cable filtering in the system than we now have,

and the cable is run in a conduit. We The observed modulation of any image

think that's better than trying to keep point is perhaps two television lines,

capacity down. Most of the modulation comes from ripple

L. D. Grignon: At the Lake Placid 68th on the high voltage supply. A small

Convention you described a proposed amount is very definitely in the deflection

system of 675 scanning lines with a 24/sec system. These ripple effects are easy to

frame rate. Is this such a system? remove, but the equipment we are now

Dr. Gillette: At the moment this is not constructing is not 675-24 equipment.

396 May 1952 Journal of the SMPTE Vol.58

Progress Committee Report

X ROGRESS in the motion picture
studios during 1951 was mainly another
step forward in the application of new
color systems and in the completion of
the installation and operation of mag-
netic recording equipment.

Feature pictures made with Eastman
negative-positive color, Ansco negative-
positive color, Eastman color negative
with Du Pont color positive, and East-
man color negative with SUPERcine-
COLOR color positive are in release.

While the continued economy drive
has tended to restrict the production
value of some pictures, many producers
are experimenting in order to determine
just how much the illusion in a picture
may be enhanced by the increased scope
which is provided by large, spectacularly
illuminated sets and complicated rou-
tines. As a case in point, the concluding
ballet scene in An American In Paris
was filmed after the picture in its original
form was completed ! l

The Telecinema at the Festival of
Britain in London deserves special
mention in the introduction to this
report.2 This theater was built to
demonstrate technical advancements in
the projection of motion pictures, sound,
and large-screen television. It is open
to the public on an admission-fee basis
and at last reports was self-supporting.

Large-screen television is shown and
people arriving in the auditorium proper
see on the screen others in the foyer
who are about to enter. Stereoscopic
color motion picture short subjects of

Submitted, April 22, 1952, at the Society's
Convention at Chicago, by Charles W.
Handley, Committee Chairman.

the Polaroid variety are shown along
with stereophonic sound. The theater
has a maskless screen with an illuminated
surround which changes brightness with
the intensity changes in the picture.

The Motion Picture Research Council
in Hollywood has described its work on
the investigation of stereoscopic motion
pictures stating that thus far the major
producers have not used the presently
available systems on anything but a
novelty basis. These systems have
been described in various articles and
papers.8"8

An announcement was made that
Arch Oboler will start soon on a "three-
dimensional" film in Hollywood which
will utilize the Natural Vision Process for
both taking and showing.9 In this
process the pictures are photographed
with two synchronized cameras. They
are projected with two synchronized
projectors having polarizing filters over
the lenses and are viewed with Polaroid
spectacles. The system will require
four projectors for a full-length feature,
or for production to be made in such a
way that special intermissions may be
used for rethreading. It was stated
that either film may be used for two-
dimensional showing. Demonstration
films have been exhibited to invited
groups by M. L. Gunsburg who is the
head of the Natural Vision Process.

An interesting development during
1951 has been the formation in Great
Britain of a new company known as
High Definition Films Limited, whose
object is to develop electronic camera
and kinescope recording equipment
suitable for use in film studios for pro-

May 1952 Journal of the SMPTE Vol. 58

397

duction of first-feature films. The
intention is to use a television camera
to record the scene on a pickup tube,
this picture then to be photographed on
motion picture film in the conventional
manner. It is planned to develop
equipment having a sequentially scanned
900-line picture in order to obtain the
necessary quality. The company is
staffed largely by engineers recruited
from The British Broadcasting Corpora-
tion television service and is headed by
Norman Collins who was until recently
Controller of B.B.C. Television. Initial
development work on the necessary
equipment is planned to take two years.
The company hopes to rent equipment
to studios requiring it for production
purposes. The proposals outlined by
High Definition Films Limited have
met with a mixed reception from the
film industry and it remains to be seen
to what extent they will be accepted.

The Naval Photographic Center suc-
cessfully completed the experimental
utilization of television equipment to
produce a motion picture training film.
Employing television cameras and tele-
vision studio techniques, the motion
picture negative was exposed by kine-
scope recording. Sound was simul-
taneously recorded on a standard film
recorder. Composite prints from these
negatives were distributed in the usual
manner and are adequately fulfilling
the requirements of a Navy training film.
From the Navy's point of view, the pro-
duction work load does not appear to
warrant employment of such equipment
at this time. In time of full mobiliza-
tion, when production time must be
greatly shortened, it is probable that
serious consideration will be given to
the production of Navy training films
with television equipment.

Motion pictures on tape have been
listed as a possibility for practical future
use.10 Several people are reported to
be working on magnetically recorded
motion pictures in which no optics are
used in the electronic "camera." It is

claimed that a demonstration was made
whereby the picture was recorded on
a J-in. magnetic tape from a home
television reception of a motion picture
film being televised. Images in the
rebroadcast were fuzzy but comparable
to results obtained with early TV
receivers.

Additional television equipment has
been installed in theaters throughout
the country. These installations in-
clude both the direct projection system
and the intermediate system. Con-
siderable publicity has been released on
the Swiss Eidophor system which was
mentioned in the Progress Report of
1950. Preliminary demonstrations for
the press and motion picture industry
people were conducted in Zurich during
mid-November 1951. Preparations are
now being completed for early demon-
strations in New York.

Color Processes. At least three studios
are equipped to do all or a part of color
processing on a limited number of full-
length color features.11

Consolidated Film Industries report
additions and new equipment for color
processing which enable them to handle
the various negative-positive color proc-
esses and also to make 35mm theater
release prints from 16mm Kodachrome
originals, as well as 16mm color re-
ductions on Kodachrome from 35mm
Eastman color negative.

Pa the Laboratories, Inc., have also
installed equipment for the handling of
color processes and additional auxiliary
equipment is being built.

Cinecolor Corporation has stated that
its processing capacity for color coupling
films has doubled during 1951. Also
that SUPERcineCOLOR processing has
been set up in the Cinecolor (G.B.)
laboratory in England.

It was reported that an increasing
demand for color film was noticeable
during 1951 in Great Britain, but
shortage of suitable raw stock reduced
the actual footage used. Technicolor

398

May 1952 Journal of the SMPTE Vol. 58

Table I. Releases on Various Color Processes.

Picture

Studio

Negative Type

Positive Type

Greatest Show on Earth

Golden Girl

The Belle of New York

The Lion and the Horse
The Wild North
Honey Chile
Jack and the Beanstalk

Paramount Technicolor 3350 K Technicolor

Balance

20th Century-Fox Technicolor 3350 K Technicolor

Balance

M.G.M. Technicolor 3350 K Technicolor

Balance

Warner Bros. Eastman Eastman

M.G.M. Ansco Ansco

Republic Eastman Du Pont

Warner (Release) Eastman SUPERcineCOLOR

and Gevacolor were the only 35mm
processes available.

Changes in color sensitivity of the
Technicolor process from a white -light
balance to a color-temperature balance
of 3350 K has been accomplished and
the majority of Technicolor pictures
now in production are being made
with that system. The camera filter
arrangement is changed when shooting
with white light on interiors or when
shooting exteriors.

Table I shows some of the pictures
now in release which were made on the
various negative-positive color processes.

The J. Arthur Rank Organization in
Great Britain have adopted Ektacolor
sheet film for preparing stills for back
projection. The scene is photographed
on a 5-in. X 4-in. Ektacolor negative
and then reduction-printed on Ektacolor
positive to a size 3 in. X 2.2 in., which
is required for the back projection slide.

35mm Photography

A new synthetic base for photographic
film has been developed by the Du Pont
Company.12113 Greater toughness and
dimensional stability are claimed for
this than for other types of safety base
film. Two years will be needed to com-
plete large-scale manufacturing facili-
ties.

In England the J. Arthur Rank
Organization are reported to have used
an improved traveling-matte system on

a large number of pictures. This
process, designed for monochromatic
photography, involves the use of a special
beam-splitter camera and special colored
lighting. The process can be used on
almost any subject and does not suffer
from the limitations imposed by subjects
such as smoke, fine detail, reflections in
glass, etc.

Lighting Equipment and Techniques. The
change in color sensitivity of the Techni-
color process from a white-light to a
color-temperature balance of 3350 K
has brought about a large increase in
the use of incandescent lamps on that
system. On small sets and in places
where the light from the incandescent
lamp is adequate, the illumination may
be largely, or entirely, from the in-
candescent tungsten sources. It is
possible to use this system on either a
3350 K balance or a white-light balance
by changing the filter arrangement at
the camera. During the introduction
of the system the incandescent lamps
were left unfiltered and the carbon arcs
used for effect lighting were filtered to
the lower balance. Recently some
directors of photography have been re-
verting to a white-light basis and filtering
the incandescents for certain large-area
sets like a theater shot in which follow-
spots are used, or a night exterior.14

The foregoing changes have brought
the 10-kw incandescent lamp back into

Progress Committee Report

399

Fig. 1. Paramount 5-kw incan-
descent remote-control lamp.

use as well as a unit which is known
as Type T-5.15

The Paramount Studio's engineering
department has developed a remote-
control lighting system for use with
incandescent lamps (Fig. 1). With this
system, lightweight units mounted in
various places may be moved at almost
any angle, or the focus changed by
motor control from a remote master
station.16 The system was designed for
use on a circus picture where the lamps
had to be mounted on the tent poles;
however, it is being adjusted with the
thought of bringing studio lighting to
an automatically controlled operation
insofar as possible. At the time of this
writing only the one studio had built
any of these motor-drive-controlled units.

Cameras and Accessories. Details of a
new combination aerial combat-recon-
naissance camera have been announced

by the Bell & Howell Go.17 It is a
lightweight, portable, 35mm motion
picture camera designed to Air Force
specifications by the company in co-
operation with Air Force engineers.

35mm Sound Recording

The report on progress in sound re-
cording deals almost entirely with
developments in magnetic recording.

Those motion picture producers who
had started the conversion of their
plants from photographic to magnetic
recording during 1950 completed the
change-over during 1951, with the
result that by the end of last year ap-
proximately 75% of the original pro-
duction recording, music scoring and
dubbing in Hollywood was being done
on magnetic recording equipment.

Magnetic recording has also found
wide use in western continental Europe.
For motion picture recording both
35mm and 17jmm films are used. In
England the adoption of magnetic re-
cording facilities by studios during 1951
proceeded with considerable caution.
It was reported that British Lion Studios
at Shepperton are installing five mag-
netic conversion kits for their Western
Electric photographic recorders and
Associated British Picture Corporation
at Elstree are planning to install five
channels of RCA 17jmm magnetic
sound equipment.

As a result of experience with magnetic
recording during the year, economies
anticipated in 1950 were realized in
1951. Through the use of equipments
designed to record three tracks on the
one strip of film, very worth-while re-
ductions in storage-vault requirements
were made and the cost of dubbing
domestic and foreign versions was ma-
terially reduced.18-19

At the Columbia Studios, storage-
vault requirements were reduced by a
ratio of better than ten to one and the
cost of dubbing foreign versions at this
same studio was reduced two to one.

By the use of magnetic equipment on

400

May 1952 Journal of the SMPTE Vol. 58

original production recording, great
savings have been effected in the quan-
tity of photographic film used by the
industry. This stems largely from the
dractices of . transferring only choice
takes from magnetic to photographic
film for dailies and for editing purposes.
A number of problems have arisen
during the year connected with the use
of magnetic film, one of the most serious
of which has been the problem of edit-
ing.20 A number of solutions have been
offered, some of which have been adopted
with varying degrees of success. Among
the solutions, the following may be
listed as having the most promise:

1. Magnetic recordings may be trans-
ferred to negative photographic film
from which prints are made.

2. They may be transferred to direct-
positive photographic film which can
be used for editing and re-recording.

3. The transfer may be made to a
film carrying both photographic and
magnetic media.

4. The magnetic material itself may
be edited.

The problem of editing the magnetic
material itself remains one of the most
difficult, and while several designs of
editing and splicing equipment have
been made, it appears that a satisfactory
magnetic film splicer and splicing tech-
niques still remain to be produced.

Paramount Studios report that their
magnetic recording program has now
been extended to the first phases of
magnetic sound editing, and that striped
magnetic prints are now replacing direct-
positives for this purpose.

Columbia Studios have introduced
during the year a combination magnetic
and photographic print for editing.

Warner Bros. Studios transfer all
magnetic recordings to direct-positive
for dailies, for editing purposes and for
previews.

Another problem which has received
considerable attention during the year
has been that of magnetic head wear.
Much thought and attention has been

given to this problem and solutions
appear in the design of magnetic re-
cording heads of harder materials or
in the plating of existing heads with
some wear-resisting material such as
chromium. The latter solution has
been adopted by Columbia Studios. It
is probable that this particular solution,
however, will be superseded as newly
designed heads appear and the expensive
procedure of removing heads, replacing
them in correct alignment and putting
them through the plating process will
be discarded.

In the use of triple-track recording
equipment, problems of crosstalk be-
tween tracks and of complete erasure
of all three tracks appeared and were
solved during the year.

New magnetic recording equipments
emerged from the design to the pro-
duction stage during the year 1951.
Among these may be listed a portable
magnetic recorder manufactured by
RCA for Warner Bros. Pictures, Inc.
(Fig. 2), which records a magnetic track
on 17jmm film and operates at a speed
of 45 fpm.21 The use of this film width
and film speed constitutes a fourfold
saving of film cost compared to the
single-track magnetic recordings made
on 35mm film operating at 90 fpm.
Also manufactured by RCA during
1951 and put into service at Warner
Bros, was a direct-positive recording
optical system using a variable-area
Class A push-pull type of sound track.
In addition to providing a positive-type
sound track, this system also provides
anticipatory noise reduction.

Two new magnetic recording and re-
producing systems were introduced to
the motion picture and television in-
dustry in 1951 by Westrex Corp. The
1100 Portable Magnetic System, which
is contained in two units, provided for
high-quality recording or reproduction
with 35mm, 17jmm or 16mm film in
synchronism with picture. New features
of this system include two-way talk-back
between the mixer and recordist, a

Progress Committee Report

401

Fig. 2. RCA portable 17£mm 45-fpm magnetic recorder.

Bros. Pictures. Inc.)

(Courtesy of Warner

new synchronizing bloop unit which
records an audible signal on the mag-
netic film in synchronism with an optical
bloop when the recorder is up to speed
and it is designed for operation with
synchronous, interlock or multiduty
motor systems.

The Westrex RA-1506-A Recorder
shown in Fig. 3 is a cabinet-mounted
equipment containing three independent
recording and reproducing channels.
Crosstalk from adjacent tracks is kept
to a level of —60 db by the introduction
of magnetic decouplers in the multiple-
head structure.

Tape recording of sound made several
gains in 1951. There was substantial
improvement in the magnetic tapes
available with respect to both mechani-
cal and electromagnetic qualities. There
were general improvements in the ma-
chines available both in the home and
professional fields. Tapes operating as
low as 1 in. /sec that reproduce voice
quality have been demonstrated. There
is a trend toward lower tape speeds
with 1\ in. /sec becoming more im-
portant to broadcasters. Fifteen inches
per second is becoming common for

402

May 1952 Journal of the SMPTE Vol. 58

high fidelity, with 30 in./sec in use
only in a few places for such work as
master recordings. Tape duplicating
service has been announced where
several copies can be made from one
original at several times normal record-
ing speed.

In the manufacture of magnetic re-
cording materials, considerable em-
phasis has been placed on uniformity of
coating and the proven results of roll-to-
roll uniformity have aided in the ac-
ceptance of magnetic recording as the
number one recording medium. A
significant contribution was made during
the past year by the Reeves Soundcraft
Corp. through its commercial intro-
duction of 16mm and 35mm film having
magnetic stripes for magnetic recording.
After two or three years of development
work, its availability on a commercial
basis was finally announced in Feb-
ruary 1951. Magnetic striping can also
be added to a customers photographic
film which has already been processed.

Some progress was made during the
year toward standardization of magnetic
track position and track width in 16mm,
17jmm, 35mm and striped 16mm and
35mm films. However, as of the present
date, no final standards have been
adopted and much work remains to
finally settle this debatable issue.

Temporary frequency azimuth test
films have been made available. Final
versions of these test films, theater test
films, and azimuth test films, covering
the full width of the film to accommodate
any track position, are at present in
work.

The Armed Services are making ever-
increasing use of magnetic recording and
it is reported that they have used over
one million feet of striped film on which
the magnetic striping covers one-half
of the photographic sound track. These
films are used where it is necessary to
record information in a foreign language
and still have the English version avail-
able for reproduction.

Fig. 3. Westrex RA-1506-A
triple-track magnetic recorder.

Progress Committee Report

403

To divert from the field of magnetic
recording, the Eastman Kodak Co.
introduced during the past year the
Eastman high-speed positive safety film
Type 5305. This is a film having the
high speed of Eastman sound recording
film Type 1357 and the finer grain of
Eastman release positive film Type
1301, both of which the new film super-
sedes.

The results of a study of the technique
of making sound-track prints on East-
man color print film Type 5381 were
reported.22-23

16mm Photography and
Sound Recording

In the field of 16mm recording, new
equipments have been introduced by
both RCA and Westrex. A number of
1 6mm magnetic recorder-reproducer
equipments (PM-66), operating at 36
fpm and capable of being electrically
interlocked with 35mm cameras or
projectors, were manufactured by RCA.

A 16mm photomagnetic re-recorder,
manufactured by Westrex Corp. (Type
RA-1509-A), is a cabinet-mounted
equipment which provides facilities for
recording and reproducing magnetic
sound track and for reproducing photo-
graphic sound track at synchronous film
speed.

Berndt-Bach Corp. of Hollywood has
built a new 16mm professional-type
camera which provides for a maximum
of 1200 ft of film and is known as the
"Super-1200." It is a sound-on-film
type and is reported to run so quietly
that no external blimp is required
(Fig. 4).2<

Both Consolidated Film Industries
and Pathe Laboratories, Inc., in Holly-
wood have provided for new building
facilities for the developing and printing
of 16mm film. These new facilities
reflect the widespread demand for 1 6mm
prints in the educational, religious and
commercial fields, together with the
increasing demands of the television
industry.

The Air Force has developed a system
of three-dimensional motion picture
photography and projection employing
a Polaroid method for right and left
picture selection. 16mm high-speed,
normal -speed and time-lapse stereoscopic
color motion pictures have been demon-
strated which used a single projector
equipped with a synchronizing drum
polarizer in front of the lens and a
Morgana-type shuttle mechanism.4

Third-dimension converters for 16mm
cameras have been described. The
camera and projector converters, screen,
and Polaroid glasses are being sold as a
package unit by the Nord Company of
Minneapolis (Fig. 5).25

A prototype of an automatic-loading
motion picture camera has been de-
signed and produced for the Naval
Photographic Center by G. J. Badgely.
The insertion of the magazine causes the
feed and take-up sprockets of the camera
to rotate, automatically picking up pre-
determined lengths of film, and forming
the film into loops before and after the
picture aperture. In its development
for television recording, the camera
uses a single motor to drive the mech-
anism. Stabilization of the shutter
is accomplished by a combination of
slipping rim and hysteresis drag. Pro-
visions are made to observe and correct
shutter banding while the camera is in
operation.

The Naval Ordnance Laboratory has
employed the image phototube as a
high-speed camera shutter. Having a
greater efficiency than the Kerr cell,
a light gain is possible. The angle
of view is governed entirely by the lens
system used.

The Springfield Arsenal has designed
a slide rule for analyzing high-speed
motion picture data. It performs several
basic calculations which must be re-
peated often in the evaluation of high-
speed photography in mechanics re-
search. It permits more rapid calcula-
tions with fewer errors and with less
highly trained personnel.

404

May 1952 Journal of the SMPTE Vol. 58

Fig. 4. New Auricon "Super-1200" 16mm professional camera.

Fig. 5. Nord third-dimension converter for any 16mm camera and projector.

Progress Committee Report 405

35mm Picture and Sound Reproduction

Means of removing heat from film
in projection is still receiving con-
siderable attention. Interference-type
heat filters and mirrors have been
written about and tested, heat-absorbing
glass filters, compressed air from jets
and water-cooled aperture plates are
in use on different types of projectors.26"30

There has been a renewed interest in
maskless screens and illuminated borders.
Several theaters have installed maskless
screens. At the Telecinema at the
Festival of Britain the surround changes
intensity with changes in screen
light.2-"-33

A considerable portion of the Sep-
tember 1951 issue of the Journal was used
for papers relating to screen brightness
and viewing conditions. A further
report on screen brightness discussed
screen light distribution.34

16mm Picture and Sound Reproduction

Important in the field of 16mm sound
reproduction is the introduction by a
number of manufacturers of models
designed to play magnetic and photo-
graphic recordings.

The new RCA "400" dual-purpose
projector makes possible the recording
of magnetic sound on processed 16mm
films without studio facilities. It also
projects sound films having optical
tracks. It opens up wide possibilities
for nonprofessional sound-film makers.35

Bell & Howell introduced a 16mm
magnetic recorder-projector, Filmosound
Model 202, which will record sound on
1 6mm films, also play back either optical
or magnetic sound tracks interchange-
ably. The company has also intro-
duced its own sound striping service
(Fig. 6).»

Television

Even though seriously handicapped
by the Federal Communication Com-
mission's freeze on the construction of
new television stations, television as an

industry has enjoyed fantastic growth
during 1951. One of the largest net-
works is reported to be a very close
second to Life Magazine in paid adver-
tising income. Even more startling
is the advance made in the manufacture
of television receivers and accessories.
The industry is now one of the leading
manufacturing activities in the country
and 1951 sales were in excess of Ij
billion dollars.

In the annual report recorded in this
Journal last year, comment was made on
the increase in quality of programming
produced by the television networks.37
This trend has continued and now it is
common for a single television show to
be budgeted as high as $75,000. Such
large budgets are justified on the basis
of increased live coverage now available
to the networks. The microwave relay
and coaxial cable system has been
considerably extended during the year.
On September 4, 1951, the first trans-
continental television program was
broadcast on the occasion of the Japanese
Peace Conference.

There has been a marked increase in
the use of theater television during 1951.
Sporting events and other special events
have been delivered to theater audiences
on an exclusive basis. The increased
use of theater television has been re-
sponsible for reduction of the cost of
equipping theaters for this purpose,
thus giving added impetus to this phase
of the industry. The final growth and
development of theater television is,
like many phases of the business, con-
trolled to some degree by the Federal
Communication Commission. Much
discussion has taken place during the
year on specific allocations to be used
for relaying television pictures from
theater to theater and from city to city.
These discussions are still active and
will undoubtedly continue through a
portion of next year.

Although the backbone of television
so far has been based on black-and-
white pictures, color television has re-

406

May 1952 Journal of the SMPTE Vol. 58

Fig. 6. Bell & Howell Filmosound Model 202,
16mm magnetic recorder-projector.

ceived considerable attention during
1951. The prolonged FCC hearing on
this subject was concluded on October
25, 1951. In spite of this activity in
the color field, the commercialization
of color is still in the future and is beset
with numerous complications, not only
regulatory and economic, but also in
terms of equipment design and manu-
facture.

Film, both direct and in the form of
video recordings, has continued to be a
major source of programming for many
television stations. The networks have
produced thousands of feet of recordings
and, in addition, many independent
film companies have been active in the
production of special features for the
industry. Great progress has been made
in controlling the cost of these special
features and at the same time retaining

acceptable quality for television broad-
casting.

Video recording is still largely done
on 16mm stock. There is some trend,
particularly in larger stations, toward
the use of 35mm for such recording
because even though the television system
is limited, on account of the standards
which have been adopted, 35mm equip-
ment can produce demonstratively better
recording quality.

With the opening of the transconti-
nental microwave relay for television,
the Hollywood television studios were
confronted with a three-hour time differ-
ential between the East Coast and West
Coast. Programs originating in the
East at 8 P.M., for example, would be
available at the West Coast at 5 P.M.,
a time when most people would not be
able to view the program. Accordingly

Progress Committee Report

407

35mm kinescope recording and process-
ing facilities were set up to handle a
half-hour show of 3000 ft of film per
reel.

Television in France is about where it
was in America five years ago. There
are regular daily transmissions from
Paris and Lille by the Radiodiffusion
and Television Francaises which is
government owned and is directed by
the Ministry of Information. The
programs are largely five-year-old films,
live studio sketches and stage plays, a
daily newsreel, and a weekly newsreel
from twenty years ago. This is partly
financed by an annual license fee due
on each receiver equal to about ten
dollars. It is estimated that there
are about 50,000 receivers in operation.

In Belgium, Holland, Germany,
Switzerland and Spain there are experi-
mental transmissions by private interests.
In Italy there are regular broadcasts
from the Vatican.

There is some divergence between
the signal characteristics used. France
uses a provisional low-standard 441 lines,
as well as high-definition 819 lines.
The 441 lines is scheduled to be aban-
doned in 1961. The other countries
are tending toward the Dutch and
German standard of 625 lines. Fifty
cycles per second interlaced vertical
scanning is used by all. Video modula-
tion here is positive, so that white
corresponds to rf modulation peaks, and
sync pulses to rf minimums. The radio-
frequencies in use at present are 40-50
me for London and low-definition
France, and 175-190 me for the high-
definition. London of course uses
405 lines.

There is one microwave radio relay
link between Paris and Lille, and several
others under study in France and Italy.

References

1. Alfred Gilks, "Some highlights in the
filming of An American in Paris, Am.
Cinematographer, 33: 18, Jan. 1952.

2. A. Bowen, J. Moir and H. Turner,
"Projection in Britain's 'Telekinema,' "

Intern. Projectionist, 26: 11, Oct. 1951.
See also: Raymond Spottiswoode,
"Progress in three-dimensional films
at the Festival of Britain," Jour.
SMPTE, 58: 291-303, Apr. 1952.

3. W. F. Kelley and W. V. Wolfe,
"Technical activities of the Motion
Picture Research Council," Jour.
SMPTE, 56: 178-196, Feb. 1951.

4. R. V. Bernier, "Three-dimensional
motion picture applications," Jour.
SMPTE, 56: 599-612, June 1951.

5. J. A. Norling, "Stereoscopic motion
pictures," Intern. Projectionist, 26: 12,
Aug. 1951.

6. J. A. Norling, "Stereoscopic motion
pictures," Am. Cinematographer, 33: 66,
Feb. 1952.

7. Norman McLaren, "Stereographic
animation," Jour. SMPTE, 57: 513-
520, Dec. 1951.

8. J. A. Norling, "Stereoscopic motion
pictures," Intern. Projectionist, 27: 9,
Feb. 1952.

9. "Plan a stereo feature picture via
'Natural Vision' process," news item,
Intern. Projectionist, 27: 22, Feb. 1952.

10. Frederick Foster, "Motion pictures on
tape," Am. Cinematographer, 32: 500,
Dec. 1951.

11. Arthur Rowan "The Wild North intro-
duces MGM's new Ansco Color
process," Am. Cinematographer, 33: 106,
Mar. 1952.

12. "Keeping up with photography."
news item, Am. Cinematographer, 33:
8, Jan. 1952.

13. "Du Font's new 'thin' film related to
'Dacron' fiber," news item, Intern.
Projectionist, 27: 10, Jan. 1952.

14. W. W. Lozier and F. T. Bowditch,
"Carbon arcs for motion picture
studio lighting," Jour. SMPTE, 57:
551-558, Dec. 1951.

15. R. G. Linderman, C. W. Handley
and A. Rodgers, "Illumination in

' motion picture production," Jour.
SMPE, 40: 333-367, June 1943.

16. Arthur Rowan, "Set lighting by re-
mote control," Am. Cinematographer,
32: 444, Nov. 1951.

1 7. "New combat reconnaissance camera,"
Am. Cinematographer, 32: 280, July 1951.

18. C. C. Davis, J. G. Frayne and E. W.
Templin, "Multichannel magnetic film
recording and reproducing unit,"

408

May 1952 Journal of the SMPTE Vol. 58

Jour. SMPTE, 58: 105-118, Feb.
1952.

19. Ralph Lawton, "The Westrex mag-
netic film recording systems," Am.
Cinematographer, 32: 182, May 1951.

20. Loren L. Ryder, "Editing magnetic
sound," Am. Cinematographer, 32: 137,
Apr. 1951.

21. Kurt Singer and H. Gonnell Ward,
"A technical solution of making
recording cost reduction," Jour.
SMPTE, 58: 329-358, Apr. 1952.

22. C. H. Evans and J. F. Finkle, "Sound
track on Eastman Color Print Film,"
Jour. SMPTE, 57: 131-139, Aug. 1951.

23. J. G. Streiffert, "Radial-tooth variable-
pitch sprocket," Jour. SMPTE, 57:
529-550, Dec. 1951.

24. Frederick Foster, "The new Auricon
'Super-1200,' " Am. Cinematographer,
32: 223, June 1951.

25. John Forbes, "Stereoscopic movies
with any 16-mm camera," Am. Cine-
matographer, 33: 72, Feb. 1952.

26. "Interference mirrors for projection,"
news item, Intern. Projectionist, 26: 17,
Apr. 1951.

27. Charles Hahn, "The trail of the
elusive lumen," Intern. Projectionist,
26: 5, Nov. 1951.

28. G. L. Dimmick and M. E. Widdop,
"Heat-transmitting mirror," Jour.
SMPTE, 58: 36-42, Jan. 1952.

29. Hugh McG.'Ross, "The cooling of film
and slides in projectors," Jour. SMPTE,
56: 538-550, May 1951.

30. Robert A. Mitchell, "An object lesson
in film-cooling," Intern. Projectionist,
27: 5, Feb. 1952.

31. "Maskless screen steadily gains favor,"
news item, Intern. Projectionist, 26: 10,
June 1951.

32. "Latest maskless screen wins audience,
technician favor," news item, Intern.
Projectionist, 26: 14, Oct. 1951.

33. "To mask- or unmask," editorial com-
ment, Intern. Projectionist, 27: 16, Jan.
1952.

34. W. W. Lozier, "Further report on
Screen Brightness Committee theater
survey," Jour. SMPTE, 57: 489-493,
Nov. 1951.

35. Ralph Lawton, "Dual-purpose pro-
jector," Am. Cinematographer, 32: 450,
Nov. 1951.

36. E. C. Hajduk, "Bell & Howell intro-
duces 16-mm magnetic recorder-pro-
jector," Am. Cinematographer, 33: 112,
Mar. 1952.

37. C. W. Handley, "Progress Committee
Report," Jour. SMPTE, 56: 568-583,
May 1951.

The Committee

C. W. Handley, Chairman

J. E. Aiken
W. L. Bell
P. G. Caldwell
J. W. Duvall
T. J. Gibbons
G. R. Groves

W. F. Kelley
R. E. Lewis
W. A. Mueller
B. F. Perry
E. H. Reichard
W. L. Tesch
I. D. Wratten

Progress Committee Report

409

Magnetic Print-Through —

Its Measurement and Reduction

By LYMAN J. WIGGIN

A simple dynamic method of measuring the value of print-through and then
a method of reducing it below audibility by application of a supersonic erase
bias during playback are described.

M,

.AGNETIC RECORDING is today's near-
cst approach to perfection in sound re-
cording and reproduction, especially
with the use of quarter-inch tape. How-
ever, there is one flaw in this "near-
perfect" method which, in some in-
stances, makes it unacceptable. This
problem is magnetic print-through,
which causes echoes to be heard preced-
ing or following a signal. This is caused
by the magnetic field, which surrounds
the recorded signal, magnetizing the
coating on adjacent turns of the roll of
tape. In a typical tape application
using 2500-ft rolls at a 15-in./sec speed,
adjacent turns are played at intervals of
from 1 to 2 sec. In certain types of re-
corded sounds, particularly voice with
no background, there are frequent
pauses of sufficient duration where the
print-through can be audible enough to
be annoying. Unfortunately the loud-
est print-through comes just before the
signal, giving a pre-echo which is never

Presented on October 18, 1951, at the
Society's Convention at Hollywood, Calif.,
by Edward Schmidt for Lyman J. Wiggin,
Reeves Sound Studios, Inc., 10 East 52 St.,
New York 22, N.Y.

found in natural sound and so is not
accepted by the ear to the same extent as
a post-echo. It is also very damaging to
certain dramatic effects and to music,
particularly loud chords following rests
where the effect is lost by the warning
given by the pre-echo. These spurious
signals have been termed print-through. •

Reeves Sound Studios encountered
this problem shortly after installing an
elaborate system of making all original
recording on Fairchild "Pic-Sync" quar-
ter-inch tape recorders. Intensive re-
search was initiated to see what could be
done to solve the problem.

The first tests tried involved listening
to recorded 1000-cycle pulses under
various test conditions, changing the
monitor system gain for each test to get
either the same audible level of print-
through or just not any, and then com-
paring the monitor system gains. This
method proved to be unsatisfactory and
inconclusive.

It was realized that some positive
means of measuring the quantity of
print-through had to be found. The
quantity of print-through that could be
tolerated had to be determined also.

410

May 1952 Journal of the SMPTE Vol.58

Permissible Amount of Print-Through

In order to find out how much print-
through could be tolerated, a tape with
male and female voices recorded on it
was reproduced in a studio. The
monitor system gain was adjusted so tiie
acoustic level was as loud as the loudest
level ordinarily used for that studio.
Then the channel system gain was re-
duced by known amounts of attenuation
until the reproduced signal was just in-
audible. This was found to be about
at the point where 55 db of attenuation
was used.

This same experiment was carried out
in a different studio with different per-
sonnel conducting the experiment. The
result was the same.

Thus, the maximum print-through
value which could be tolerated was 55 db
below 100% program level. This figure
was later corroborated very closely by
production department observations.

Method of Recording
a Print-Through Test

The following dynamic method of
measuring the quantity of print-through
was developed as a fast and direct
method. A roll to be tested is com-
pletely erased before this test is made.
A 2500-foot roll of tape is used. This
has an outer diameter of about 10 in.,
and at 15-in./sec tape speed the supply
spool turns about once in 2 sec. Pulses
of 1000-cycle tone at 100% level are re-
corded at intervals. These pulses are
long enough for the supply spool to turn
f revolution, about 1.5 sec. The pulses
are repeated every sixth revolution of the
supply spool, and generally ten such
pulses are sufficient for measurement
purposes.

After the recording, the roll is rewound
and put in storage for a definite period of
time. At Reeves Sound Studios a
period of four hours has been selected as a
standard storage period.

Method of Measuring
a Print-Through Test

In measuring, filters must be used to
eliminate or limit any hum or other noise
from the tape machine output at the
meter on which measurements are to be
made. This meter must have means for
quickly changing its sensitivity over a
wide range and must have a fast-acting
pointer movement.

A bandpass filter effect can be achieved
by using the 500-cycle high-pass filter
and 2000-cycle low-pass filter of a stand-
ard Effects Filter. The V.U. meter of a
General Radio Type 1932-A Noise and
Distortion Meter is used for the indicat-
ing instrument. This meter has inter-
locking push buttons to change sensitivity
and meets all the requirements perfectly.

On reproduction, the print-through
pulses can be observed rising above the
noise level. With recorded pulses every
sixth revolution of the supply spool, con-
sider the first print-through observed as
the one following a recorded signal.
The next will be the second print-
through from this signal. The next
will be the combination of third print-
throughs from the previous signal and
the next one coming up. The next will
be the second print-through from the
signal coming up. The next, always the
highest in value, will be the first print-
through from the signal coming up. An
example of this is shown in the readings
below of an actual print-through test.
These figures are all in db below the
100% recorded 1000-cycle pulses.

55 £ db — 1st print-through after the signal
61 db — 2nd print-through after the signal
61 \ db — 3rd order print-through
58 db — 2nd print-through before the next

signal
51^ db — 1st print-through before the next

signal

Note: Noise level was 63 db below 100%
signal in this test.

We now have a measure of how much
print-through reduction is required, as
well as a positive and simple method of
measuring.

L. J. Wiggin: Magnetic Print-Through

411

Many experiments were made with
little or no results. These experiments
included tape with thicker base, cold
storage of tape, recording at lower levels
and change of recording bias, using a
narrower playback head to eliminate any
edge effect and using special tapes.

One method of print-through reduc-
tion stood out as being practicable with-
out suffering too much from the "in-
evitable compromise." This method
consists of actually erasing the print-
through just prior to the program repro-
duction at the playback head or at least
erasing enough of it to make the remain-
ing amount inaudible. If print-through
is caused by inclusion of very easily
magnetizable fractions in the coating
and is mainly a surface effect, then
threshold erase is a solution.

The following paragraphs and tables
are the final results of what seems to be a
positive method of eliminating trouble
due to audible print-through, at least
with the Fairchild "Pic-Sync" machines.

Print-Through Erasing

Although the application of some sur-
face erasing seems to eliminate audible
print-through, complete tests had to be
made to see if any other trouble was be-
ing introduced in the reproduction of
program material.

In all the tests that follow, the record-
ings were made in the normal manner
described before. In the reproduction
of these tests the following changes were
made to the Fairchild Machine:

1 . The record head plug was removed
from its receptacle.

2. The erase head plug was inserted
into the record head receptacle. This
connects the erase head to the normal
record head circuit.

3. A 3700-ohm resistor was added in
series with the present 2200-ohm V.I.
meter bridging-out resistor for the reading
of bias. This allowed a reading of bias
to the erase head of 0.5 db which corre-
sponded to a Ballantine VTVM reading
of 16.5 v across the erase head and 3.5 v

across the 130-ohm resistor normally in
series with the record head.

4. The relay coil circuit in the power
amplifier chassis was opened so the
14,000-cycle control signal would not be
applied to the record head receptacle to
which the erase head is now connected.
This was necessary to avoid loss of syn-
chronization on playback.

The above changes put what was
normally the 69,000-cycle recording bias
into the erase head instead. The value of
this bias is adjustable with its usual con-
trol.

The first test made was to see what
happened to frequency response with
erasing before playback. A complete
frequency test was recorded 10 db below
100% signal level in a normal manner.
It was reproduced immediately in a
normal manner. Then it was immedi-
ately reproduced three more times with
69,000 cycles to the erase head each time.
The results are shown in Table I. They

Table I. Frequency Response,

in db.

69,000 cycles

to erase

head

On

On

On

Fre-

1st

2d

3d

quency

Off

re prod. re prod.

reprod.

30

0

+0.1

+0.1

0

40

0

+0.1

+0.1

0

50

0

+0.1

+0.1

+0.1

70

0

0

0

0

100

0

0

0

0

200

0

0

0

-0.1

500

0

-0.2

-0.2

-0.2

1,000

0

-0.3

-0.3

-0.3

2,000

0

-0.3

-0.3

-0.3

3,000

0

-0.4

-0.4

-0.4

4,000

0

-0.4

-0.4

-0.4

5,000

0

-0.5

-0.6

-0.5

6,000

0

-0.5

-0.5

-0.5

7,000

0

-0.5

-0.5

-0.5

8,000

0

-0.5

-0.5

-0.5

9,000

0

-0.5

-0.5

-0.5

10,000

0

-0.4

-0.5

-0.5

11,000

0

-0.5

-0.5

-0.5

12,000

0

-0.6

-0.6

-0.6

13,000

0

-0.6

-0.5

-0.6

Note : Correction applied to all readings so
normal playback response shows flat.

412

May 1952 Journal of the SMPTE Vol. 58

show two important things: first, a slight
loss in high frequencies (the only com-
promise found so far); and second, im-
mediate repeated playings have no more
effect. These effects can be explained
by assuming that the erasing being done
is only on the tape surface. The high'er
frequencies, occupying less depth in the
magnetic coating, had a higher propor-
tion of their signal erased, and after the
surface is erased once, there is no more
effect, at least for immediate repeated
playings. Final tests have not been
made to determine what happens if this
same roll is stored a while and then re-
produced again in a like manner.

The above test was repeated using
rolls from three tape manufacturers,
except that the repeated playings with
the 69,000 cycles were considered un-
necessary. The results showed no ap-
preciable differences. However, it is
quite conceivable that recordings made
on another machine, other than a Fair-
child, could show a difference. Whether
or not this could occur would depend on
the record head magnetic field distribu-
tion and the resulting ratio of frequency
vs. depth of magnetization on the tape.

The next test made was for distortion.
An intermodulation signal of 2000 and
1 00 cycles at a 1:4 ratio was recorded,
varying the input level in regular steps.
The results in Table II show that there
is no change at all in intermodulation
distortion with the application of 69,000
cycles to the erase head, either during the
first playback or for three successive
immediate playbacks. This holds true
also for tapes of other manufacturers
recorded on the Fairchild machine.

The next test made was to see if output
level changes would follow input level
changes over a reasonable range. The
results of this test are shown in Table III.
Appropriate high-pass and low-pass fil-
ters were inserted in the playback circuit
to limit the effect of noise as far as possi-
ble, although some radio-television inter-
ference was present. With the applica-
tion of 69,000 cycles to the erase head the

Table II. Intermodulation Distortion,

Input
level

69,000 cycles to! erase head

Off On

On

On

-4 1.0 1.0 1.0 1.0

-2 1.0 1.0 1.0 1.0

0* 1.3 1.3 1.3 1.3

+2 1.8 1.8 1.8 1.8

+4 2.7 2.7 2.7 2.7

+6 4.0 4.0 4.0 4.0

+8 6.0 6.1 6.0 6.0

* Normal 100% input level.

1000-cycle output dropped about 0.25 db
and the 10,000 cycle output dropped
about 0.5 db. This is to be expected
from the frequency-response results
shown in Table I, as the same roll of tape
was used. The second and third suc-
cessive playbacks with the 69,000 cycles
to the erase head again show no signifi-
cant difference from the first playback
under the same conditions.

The above test was repeated with
tapes of other manufacturers with prac-
tically identical results.

A specific test for the effect on noise
level of 69,000 cycles applied to the
erase head was not made. However, in
all print-through tests the noise level
showed no change with the restricted
frequency range used.

In trying to leave no stone unturned,
any possible effect of a subharmonic of
the 69,000 cycles applied to the erase
head causing trouble was investigated.
As any effect of this sort should show up
in a frequency-response test, a recording
was made 10 db below 100% level of a
very slow frequency sweep up and then
down from 6000 to 8000 cycles. This
sweep included 6900 cycles, the 10th
subharmonic of the 69,000 cycles. The
total recording time of this test was 2 min
and 40 sec. In playing back this re-
cording with the 69,000 cycles applied
to the erase head no significant change in
output level was noticed at any fre-
quency.

L. J. Wiggin: Magnetic Print-Through

413

Table III. Output vs. Input Level Change Linearity, in db.

69,000 cycles to erase head

Input
level

Playback at 100
1st 2d

Variations from reading without 69,000 cycles
cycles Playback at 1000 cycles Playback at 10,000 cycles
3d 1st 2d 3d 1st 2d 3d

0*
-10
-20
-30
-40
-42
-44
-46
-48
-50

-0.1
-0.1
-0.1
-0.2
0
-0.2
-0.1
-0.1
-0.4
-0.4

-0.1
-0.1
-0.2
-0.2
0
-0.1
-0.1
0
-0.2
-0.4

0
0
-0.1
-0.1
+0.1
0
-0.2
0
-0.2
-0.3

-0.3
-0.4
-0.3
-0.3
-0.2
-0.1
-0.2
-0.3
-0.2
+0.1

-0.3
-0.3
-0.3
-0.3
-0.2
-0.1
-0.2
-0.3
-0.3
0

-0.3
-0.3
-0.2
-0.3
-0.2
-0.2
-0.2
-0.4
-0.4
+0.1

-0.5
-0.6
-0.4
-0.6

-0.6
-0.4
-0.6
-0.4
-0.1

-0.6
-0.9
-0.6

-0.7

-0.5
-0.4
-0.6
-0.4
-0.2

-0.8
-0.9
-0.7
-0.8

-0.4
-0.5
-0.6
-0.4
-0.1

Normal 100% input level.

The last test was to determine the
print-through reduction effectiveness of
the 69,000 cycles to the erase head on
tapes of three manufacturers. Tape
rolls A, B, G and D were put through the
regular print-through test routine except
that two groups of 1000-cycle pulses
were recorded on each roll. The first
group on each roll was played back
with the machine in its normal condition
and the second group with the 69,000
cycles to the erase head. The results are
tabulated in Table IV. Accepting the
figure of 55 db below 100% signal level as

Table IV.

Level, -db Below

Print-Through and Noise
100% Signal

Tape Roll

Tape Tape Ta
A B

ape

Tape
D

69,000 cycles,

off
69,000 cycles,

on
Noise level

53^ 49$ 57

65 66

63

59 $
63

being just satisfactory, tapes A, B and D
would reproduce audible print-through
under normal reproducing conditions
and tape G would not. With the appli-
cation of 69,000 cycles on the erase head,
none of the four tapes would reproduce
audible print-through.

As a result of these tests, the print-
through problem (as applied to produc-
tion methods) appears to be solved.
The only compromise in making this
possible is a slight reduction in high-
frequency response. This effect is small
enough to be ignored, but it can be com-
pensated for in the playback amplifier
equalization. Reeves Sound Studios
have put this system of print-through
reduction in operation.

In conclusion I wish to acknowledge
the helpful assistance of Raymond E.
Biondi, Homer H. Elder, Charles E.
Campbell and Richard J. Vorisek of
Reeves Sound Studios, Inc.; Ernest W.
Franck and Edward Schmidt of Reeves
Soundcraft Corp.; and Wentworth D.
Fling of Fairchild Recording Equipment
Corp.

414

May 1952 Journal of the SMPTE Vol.58

A Scientific Approach to Informational-
Instructional Film Production and Utilization

By G. R. CARPENTER and L. P. GREENHILL

This is a report on a research program sponsored by the military services for
the past four years, dedicated to production of 16mm informational films.

_L HE EQUIPMENT of the motion picture
and television industries has been de-
veloped to high performance standards
and there is continuing effort to make still
further improvements. These improve-
ments are the result of scientific research
in such fields as physics, chemistry, op-
tics, electronics and engineering. Re-
search and development have been con-
centrated mainly on technical equipment
and its performance.

The primary function of motion pic-
ture and television equipment is to com-
municate from some people to other
people. Therefore, the men who use
this equipment must deal with human
factors which are even more complex
than the electronic and optical proc-

Presented on October 17, 1951, at the
Society's Convention at Hollywood, Calif.,
by C. R. Carpenter and L. P. Greenhill,
The Instructional Film Research Program,
The Pennsylvania State College, State
College, Pa. The research on which this
report is based is sponsored jointly by the
Departments of the Army and Navy
through the Special Devices Center, Office
of Naval Research.

esses used in recording and reproducing
ideas or information on films or tele-
vision.

Motion picture and television engi-
neers, regardless of their specific technical
jobs, are working within a matrix of hu-
man factors, processes and variables.
The study and control of the interaction
between equipment and people, is an
area of applied psychology which has
been termed Human Engineering.

When we look at the field of communi-
cation as a whole it is clear that:

1 . We have made great progress in de-
veloping mass communication equip-
ment.

2. We have made little progress in sci-
entific research on the psychological proc-
esses of communication.

3. Technological developments are far
ahead of the human engineering develop-
ments.

We have developed wonderful ma-
chines and systems for communicating,
but we are uncertain how to use them to
the best advantage for worth-while pur-
poses. We can communicate with mil-
lions of people in this and other countries,

May 1952 Journal of the SMPTE Vol. 58

415

but we are not sure what should be com-
municated. Even when we know what
to communicate, we do not have confi-
dence in our knowledge of methods for
doing this effectively. There are chal-
lenging social problems which the com-
munications industries might aid in solv-
ing with their available tools, but these
tools are not being applied effectively to
these problems. The science of human
engineering as applied to radio, motion
pictures and television is in its very early
infancy.

We believe that this lack of develop-
ment is due in large part to the fact that
the human variables in mass communi-
cations have not been thought of as a
legitimate and important field for scien-
tific research and development. Conse-
quently this kind of research, both basic
and applied, has been grossly neglected.
Resources for this kind of work have not
been made available: men with ade-
quate competencies have not been
trained for the job; suitable research
laboratories have not been built.

It is encouraging, however, to observe
that there is a growing interest in the
human aspects of communications. A
few universities are establishing depart-
ments of communication. Government
and military research organizations are
supporting and encouraging research in
this field. It is being realized also that
"audience research," though important,
is not sufficient for learning all we need
to know in order to apply intelligently the
potential powers of the communication
industries to urgent problems of our so-
ciety: problems of information dissemi-
nation, of group and class tensions and
conflicts, of instruction necessary for
people's adjustment and survival in a
complex world of successive crises.

Should it not be that the primary mis-
sion of the communications industries is
that of training, instructing and informing
people, and that the amusement and
entertainment of people is the secondary
mission of these great industries? If this
emphasis were to be arranged, these in-

dustries would assume a place of para-
mount significance in our world.

Importance for Engineers

The foregoing considerations are of
great importance for motion picture and
television engineers. New nonentertain-
ment uses for the mass media are making
up a large part of current film produc-
tion. The need for extending the appli-
cation of sound motion pictures and tele-
vision is a matter which challenges pro-
motion, financing and business organiza-
tion. It is not mere speculation to visu-
alize the possibilities of nonentertain-
ment communications expanding to the
point where the volume of business for
this purpose will exceed the volume of
business for entertainment purposes.

The engineer has a vested interest in
the effects on people of the programs which
he helps produce. For example, the
strength of impact of a program calcu-
lated to sell products and the ways in
which this impact, over a period of time,
changes the behavior of audiences may
determine whether or not the program
will be continued. The engineer's
understanding of human engineering fac-
tors can have a vital influence on that
impact. Furthermore, the intelligent
communications engineer has the right
and the responsibility to be concerned
with the effects of programs on people.
If both he and the medium are to have
integrity he certainly cannot allow him-
self to be a mere automaton of the com-
munications machinery. The control of
the mass communications industries can
be centralized and authoritative or the
control can be democratic; that is, it can
be dispersed to conform to democratic
management which will permit the indi-
vidual exercise of judgment and knowl-
edge by the engineer in determining the
final influence of a motion picture or
television program.

Thus, both from the viewpoint of busi-
ness and of professional-social responsi-
bility, the engineer has a vested interest
in the effects of programs on people.

416

May 1952 Journal of the SMPTE Vol. 58

THE INSTRUCTIONAL FILM RESEARCH PROGRAM

The Instructional Film Research Pro-
gram of The Pennsylvania State College
is one limited effort to learn more about
the human factors and the ways they
interact with the sound motion picture
as a medium of communication. The
Program is sponsored by the Depart-
ments of the Army and the Navy through
the Special Devices Center, Port Wash-
ington, L.I., N.Y.

The objective of the Program in terms
of the controlling Task Order is to dis-
cover facts and principles which will im-
prove the effectiveness of films for the
rapid training, instruction and educa-
tion of large numbers of people.

It is the central responsibility of the
Instructional Film Research Program to
apply and test the application of learning
principles to instruction by motion pic-
tures, as well as to try to develop new
principles and procedures for increasing
the effectiveness of film-mediated in-
struction.

Since its beginning in 1 947 the research
of the Instructional Film Research Pro-
gram has been limited mainly to the in-
structional and informational film. The
entertainment film has been of interest to
the extent that it may involve processes
which can be used to improve instruc-
tional and informational films.

Research Methods and Procedures

There is nothing obscure or mysterious
about the methods and procedures being
used in this research. Essentially the
methods consist of applying and adapting
the general principles of scientific re-
search to the content and arrangement of
stimulus variables in instructional motion
pictures. Research is done also on the
responses of individuals to these film vari-
ables. In general, the problem is one of
determining the effects on target audi-
ences of varying the streams of stimuli-
events channeled through the sound mo-
tion picture.

Theoretical concepts are used at many
points and in many ways. Existing and
generally accepted theories of learning
may be tested in a film context. Theo-
retical hypotheses may be stated as set-
tings for research projects. Theoretical
outlines may be used to guide the direc-
tion and to help maintain the balance of
effort of the continuing program. Fi-
nally, theoretical statements may be em-
ployed to aid in summarizing and inter-
preting results. For the practical man there
is nothing so practical and useful as sound,
tested theory.

Usually we proceed by defining the
variable which we wish to study in a film.
For example, we may define rate of de-
velopment (the screen time devoted to each
phase of the action) as the variable to be
studied. The next step may be to pro-
duce several film versions, preferably
three or more, which differ from each
other only with respect to rate of develop-
ment. These different versions of films
are then shown to matched groups of sub-
jects, the amounts of learning are meas-
ured, and the results are compared.
Thus, we learn which rate of development
is most effective for the experimental film
when used with a particular audience.

Several variables may be combined in
a single version. For example, rate of
development may be combined with con-
trolled repetition. By producing system-
atically a number of experimental film
versions, some with and some without
specified variables, it is possible to deter-
mine the relative contribution of single
variables to learning, and also to deter-
mine how the variables interact with each
other and with the audience. Some
variables may be additive in their effects
while others may interfere with each
other.

An experimental film variable is a
single, definable, controlled character-
istic of a film, such as rate of develop-
ment, repetition, "subjective" camera

Carpenter and Greenhill: Instructional Film Production

417

Fig. 1. Class learning tumbling skills from film loops and daylight projector.

angle or level of difficulty of materials
presented.

An experimental film version is a film
which contains the controlled stimulus
material (one or more variables) which is
presented to a test group of subjects.
A number of such film versions varied in
systematic fashion may be required for
solving a defined problem.

The design of experiments for testing
the relative effectiveness of several vari-
ables is somewhat complicated. How-
ever, there is one clear, simple and impor-
tant concept, namely the effects of a vari-
able are measured in terms of the re-
sponses of the individuals in the test popu-
lation. For example, if the film's pur-
pose is to teach facts, the central question
is: How does the experimental variable
contribute to or interfere with the learn-
ing of the presented body of facts? The
measurement is made by informational
tests, usually of the multiple-choice type.

The objective is to measure the amount
of learning and retention which is the
result of the film, and furthermore to
measure the increment of learning which
results from the presence in the film of
the one or more experimental variables.
With experimental films which have
as their objective the teaching of perform-
ance skills, essentially the same pro-
cedures are used, except that actual per-
formance tests are given. The basic
questions are: Does the film actually
teach the performance of a skill, and how
much of this teaching is the result of the
experimental variable? Does a film
which is supposed to teach first aid to the
injured actually teach trainees something
in addition to what they already know?
If one of the versions contains the experi-
mental variable showing errors to be avoided
in giving first aid, is this film version su-
perior to one which does not include the
variable showing errors to be avoided? With

418

May 1952 Journal of the SMPTE Vol. 58

performance-type films the desirable
thing is to measure the actual performance
of the trainees before and after filmic in-
struction, i.e., to determine whether the
performance of a group trained by a film
under specified conditions is superior to
that of a control group, i.e., an equivalent
group not subjected to the filmic in-
struction.

In brief, the emphasis is on measuring
actual changes in the behavior which re-
sults directly from the stimulus value of
the film and its controlled character-
istics.

Research Projects and Results

The Instructional Film Research Pro-
gram has formulated more than sixty
research projects. These projects may
be classed into four groups: (1) the de-
velopment of new research equipment,
and testing of new equipment and
methods of using it; (2) research on
films produced for teaching performance
skills of various kinds; (3) research on
films for teaching factual information;
and (4) research on films for changing
attitudes, opinions and personal orien-
tation.

These four general groupings of proj-
ects which have been undertaken by the
Instructional Film Research Program
include many phases of work related to
both the producing and the using of
sound motion pictures. Many of the re-
search projects are also closely related to
basic problems of radio and television as
well as to the sound motion picture.
Regardless of the research problem or the
type of materials used, the basic require-
ment is to discover variables of communication
which produce desired changes in the behavior
(perception, learning, motivation, actions, etc.}
of defined target audiences.

Development and Testing of Equipment

A. The Classroom Communicator and
Film Analyzer systems1-2 have been
planned and developed primarily as re-
search tools appropriate to the problems
being attacked. However, this equipment

may be adapted for a wide range of uses
in practical situations where reactions
of people to film, radio and television pro-
grams are required. Figures 1 through
4 show the general features of these sys-
tems.

The Classroom Communicator and
Film Analyzer when used together are
capable of doing the following:

1. Immediately recording discrete re-
sponses of individuals in audiences of up
to 40 people in size.

2. Continuously recording reactions,
decisions and judgments of individuals
while programs are being presented.

3. Rapidly informing either individuals
or the group as a whole of the results of
their responses, i.e., correctness or incor-
rectness of choices.

4. Providing summated numerical rec-
ords of scores for individuals.

B. The Tele-kit. Research was done
last year (1951) on field-testing and de-
veloping methods of use for the Tele-kit
daylight 16mm projector along with the
T.A. Repetitive Impact continuous film
magazine.3'4 The equipment for testing
was made available to us by Capt.
William C. Eddy of Television Associates,
Inc.

The purposes of one series of experi-
ments were to determine the functional
characteristics and limitations of the
equipment, to develop methods and pro-
cedures for using the equipment, and to
test the amounts of learning of skills in-
duced by methods of instruction using
this equipment in comparison with other
methods of instruction. Using the film-
loop projector alongside the training
area, and providing for interspersed prac-
tice was found to be a very effective
method of teaching physical performance
skills.

Further research5 was conducted at
The Great Lakes Naval Training Station
to study the following problems relating
to the use of the Tele-Kit and the film
loop:

Carpenter and Greenhill: Instructional Film Production

419

Fig. 2. Film Analyzer Recorder for continuous or intermittent recording
of audience responses

1. The effects on learning of various
screen-viewing distances and angles.
The distances studied ranged from 4 to
24 screen widths, and the angles of view-
ing the screen ranged from 0° to 60°.
The screen size was 12 in. X 18 in.

2. The effects of ambient illumination
on learning from a film projected on the
"daylight" rear-projection screen. Tests
were made under normal room-lighting
conditions and in a darkened room.

3. The effects of repeated viewing of a
film loop.

4. The relative effectiveness of films
having "slow" and "fast" rates of de-
velopment.

5. The effects of having the trainees
practice an assembly task while viewing
the film loop.

The task taught by the film was the
assembly of the breech block of the 40mm

antiaircraft gun. The effects of the film
variable or the methods of presentation
variables were measured by testing the
actual performance of the subjects in as-
sembling the breech block immediately
after the film showings. About 2000
Navy trainees were used as test subjects.

Preliminary analyses of data show the
following results:

7. Viewing conditions. The optimum
area for viewing the daylight projection
screen lies within a total angle of 60°
(30° on each side of the projector axis).
This area extends out to a distance of
16 screen widths from the screen (24 ft
for the 1 8-in. screen) . Of the men seeing
the film from within this area, approxi-
mately 75% succeeded in assembling the
breech block. Thirty to forty indi-
viduals could be seated in this area.

An area of reduced learning effective-
ness in which approximately 50% of the

420

May 1952 Journal of the SMPTE Vol.58

Fig. 3. Instructor's Console of classroom communicator
(Veeder-Root Bank of Individuals' Score Summators to right).

men succeeded in learning the task ex-
tends around the optimum area de-
scribed above to a maximum total angle
of 100° for daylight viewing and 120°
for viewing in the dark. This area of
reduced learning extends out to a maxi-
mum viewing distance of 24 screen widths
along the central projection axis.

2. Repetition. Preliminary results of
these tests for the repetition variable are
as follows: repetition of the film by the
film-loop method aided greatly in teach-
ing trainees to assemble the breech
block. Two showings were better than
one showing, and three showings were
significantly better than two, i.e., with
repetition more and more men were able
to assemble the breech block. It may

be assumed that more repetitions would:
(1) teach more of the trainees to reach
the minimum performance requirements,
and (2) improve the assembly skills of
trainees beyond the minimum perform-
ance requirements. It may be assumed
also that a level of diminishing returns of
repetition relative to learning would be
reached. Training requirements should
determine the number of repetitions used.
These tests provide evidence for the
soundness of the "repetitive impact train-
ing" method and the usefulness of equip-
ment which makes this method relatively
easy to apply. By repeated showings of
well-prepared short units of training films,
predetermined training standards can
be achieved.

Carpenter and Greenhill: Instructional Film Production

421

3. Rate of development. A film which
was produced with a slow, deliberate rate
of development, i.e., the showing of an
action or sequence of actions with more
screen time than is normally used, was
much superior to a film with a fast rate of
development. Results show that repe-
tition amplifies this advantage and that
the disadvantages of rapid, heavily
packed or concentrated films cannot be
entirely compensated for by repeated
showings.

4. Participation. Practice by the trainees
of the performance at the same time
as it is being shown on the screen
aids learning of the skill when used with
the slow-development film, and reduces
learning when used with the fast-develop-
ment film. It can be generalized from
these and other findings6*7 that in order
for concurrent practice of a skill being
taught by a film to be effective the film
must have a sufficiently slow rate of de-
velopment or allow time between se-
quences in order to provide favorable
conditions for practice or participation.

These field tests of the actual effec-
tiveness of the Tele-Kit daylight rear
projection equipment and the T.A.
Repetitive Loop magazine demonstrate
the kinds of functional testing that can
be done with many other kinds of film,
radio and television equipment. Such
field tests can be used to establish the
"operational characteristics" of equip-
ment in terms of what this equipment can
actually be expected to do to people, and
not in terms of opinions about what the
equipment might do when people are
involved. For example, there is a gen-
eral standard that the maximum screen-
viewing distance should be limited to six
times the screen width. Field tests in
terms of effects on learning a skill show
that for the film and projector used the
effective viewing distance extends to 16
screen widths from the screen but that
beyond this distance there is a substantial
decline in the subjects' learning. These
findings have important implications for
television viewing.

Film Variables Positively Related to Effi-
cacy of Instruction in Performance Skills

Four major projects of the Instruc-
tional Film Research Program have iso-
lated eight variables that aid learning.
In other words, films with the following
eight characteristics are likely to be
more effective in teaching performance
skills than films which lack these char-
acteristics:

7. Medium verbalization. Tests have
shown that the optimum number of
words in the commentary should range
between 100 and 130 words per minute of
film.8 Fewer rather than more words
are probably desirable. When complex
pictorial material is shown, or when it is
necessary to use new terms and words,
repetition of these new concepts should
be employed.

2. Audience participation. Practice of
the skills at the same time as the film is
being shown aids learning greatly if
favorable conditions for practice are pro-
vided.7 These can be achieved by pro-
ducing films with slow rates of develop-
ment so that the action on the screen can
be followed and the practice carried out
without loss of attention to either. Or,
short sequences of films may be shown,
followed by opportunities for practice,
repetition of the same sequences, or the
presentation of a new element of the skill
to be learned. Portable projectors, day-
light screens and television tubes make it
possible to present audio-visual instruc-
tion to trainees in actual work situations,
e.g., on assembly lines. Thus skills can
be taught to trainees in situations where
they can learn with expert guidance from
the film as they practice. When these
methods are applied great economies
may be made by reducing the amount of
trial and error in training, by savings in
man hours of both trainees and instruc-
tors, and by increasing the amount of trans-
fer of learning from the training situation

to the actual work situation.

3. Slow rate of development. It has been
found to be advantageous to use what
might be called a slow rate of develop-

422

May 1952 Journal of the SMPTE Vol. 58

Fig. 4. Communications Test Room showing: Instructor's Console; Film Analyzer;

Veeder-Root Bank for summing scores; correct-answer, signal-light panel (on wall);

and response stations (on desks).

ment of the subject being presented.8
This requirement is especially important
in elementary introductory training.
Clearly, optimum rates of development or
rates of pacing a film will vary depending
on the complexity of the skills being
taught, the abilities and previous experi-
ence of the trainees, and other factors
such as the conditions under which train-
ing is being done. It is impossible, there-
fore, to state a simple rule of thumb.
Pacing or development rate interacts
with repetition. Repetition may, in
part, compensate for the inadequacies of
films which "move" too fast. It would
seem on the basis of inspection and testing

of large numbers of training films that
those currently being produced are more
often too fast in development rather than
too slow. The abilities of trainees to per-
ceive and learn from a film are generally
overestimated by those of us who produce
films.

4. Repetition. Repetition has been ac-
cepted both in theory and practice as a
necessary condition for learning. Sud-
den learning ("insight") without trials,
errors and successes is an exceptional
occurrence. The factor of repetition has
been accepted and used not only in edu-
cational procedures, but is recognized
and applied in journalism, advertising
and the arts.

Carpenter and Greenhill: Instructional Film Production

423

The results, therefore, of gains in learn-
ing or acquisition of a skill as a conse-
quence of the repetition of film presen-
tations were to be expected.8 The em-
ployment of repetition in films for teaching
skills can now be recommended with
confidence.

There are, to be sure, many unan-
swered questions: How can the variable
of repetition be employed most effec-
tively? How many repetitions are desir-
able for a particular training job, with a
specific group of trainees, in order to
achieve a given level of training? How
is repetition to be used and yet monotony
and lack of interest avoided? How can
introductory and summarizing sequences
be used to repeat and reinforce instruc-
tion? How can repetition be varied so
that more generalized training rather than
very specific training will be the result?
Some of these and other questions must
be answered for specific kinds of training;
general rules have limited use. This
does not negate the proposition that repe-
tition increases the effectiveness of a
training film, whether simple repetition
of the film as a whole or internal repetition
with variation is used. It is surprising in
this connection that more use has not
been made of this simple principle of
repetition by producers and users of
training films.

5. Showing errors to be avoided. Is it de-
sirable to show errors of performance in
training films? The results of Jaspen's
experiments8 indicate that errors when
shown aid learning of the required per-
formance provided the error or wrong
way is clearly described as an error to be
avoided and differentiated from the cor-
rect method. The trainees must be
taught — shown and told — what the
errors are and how to avoid making
them. It would appear that "negative
training," i.e., trainees performing incor-
rectly rather than correctly as a result of
seeing the errors, is a consequence of in-
complete training in the discrimination
of errors. It might be mentioned here
that good coaching usually involves di-

recting the learner's attention to his mis-
takes. The simple ride is to show clearly
errors to be avoided in a performance.

6. Camera angle. Roshal6 did an ex-
periment on a pair of variables — the 0 °
versus the 180° camera angle. In other
words the training task was photo-
graphed from the viewpoint of the per-
former (0 ° angle) and from the viewpoint
of the observer (180° angle). The film
which was photographed from the view-
point of die performer proved in actual
tests to be superior to the film photo-
graphed from the viewpoint of the ob-
server. From this and related studies a
simple rule can be formulated: In skills-
training films wherever possible show the
job exactly as it will be seen by the trainee when
he performs it.

7. Personalized commentary. Zucker-
man9 has studied some of the character-
istics of commentaries in training films.
He found that the direct personalized
form of address proved to be better than
several other forms of address. In other
words, addressing trainees as "you" and
otherwise personalizing the instructional
commentary had advantages. In addi-
tion, the timing of the commentary to the
visuals may have an important influence
on learning. In teaching a complex
skill, alerting the learner to the action to
appear on the screen by having the com-
mentary slightly lead the picture was
found to be helpful.

8. Motion. Results of experiments tend
to confirm the widespread belief that in
teaching skills involving action, motion
pictures are superior as instructional ma-
terials to successive stills of the action.6
Where the crucial cues to be learned are
action or movement cues, then a motion
picture representation is superior to those
methods which do not represent the com-
plete action. This raises a basic ques-
tion: What are appropriate training
tasks for motion pictures or television?
The suggested answer is that those
training tasks which involve crucial and
complex action are appropriate to those

424

May 1952 Journal of the SMPTE Vol. 58

media which have capabilities of showing
such action.

Variables Related to the Efficacy
of Informational Films

7. Idea density. We have approached
the problem of the amount, rate and diffi-
culty of materials presented by means of
films.10 This problem arises out of con-
siderations of the following questions:
How much information can be presented
effectively in a given period of time?
How does the level of difficulty affect the
learning of this material by given audi-
ences? What treatment of a given body
of information most effectively carries the
informational content to be communi-
cated? For example, is the story form of
organization more or less effective than a
straight expository presentation, or is an
organization with a prominent outline
which is repeated and made very clear by
titles more or less effective than a smooth
uninterrupted development?

The research on these problems is not
yet complete or definitive. However,
we have developed a strong conviction
that for a single film there is an optimal
amount of information for a given audi-
ence or trainee population. We are also
convinced that many instructional sound
films are overloaded with information,
i.e., more material is presented than can
be effectively learned by the audience.

2. Introductions and summaries. We have
made a rational analysis of the functions
served by film introduction and summary
sequences.11 Tests have shown that
introductions and summaries may either
add or detract from the instructional value
of films depending on their adequacy.
Poor, sketchy and ambiguous introduc-
tions and summaries may reduce the
effectiveness of the film. On the other
hand, cogent, clear and well-integrated
introductions and summaries increase
the effectiveness. These parts of a film
may be used to provide much needed
repetition, review and emphasis. They can
also be employed to set learning goals or

purposes for trainees, to clarify organiza-
tion and to emphasize the importance
and practicability of the film's contents.

3. Pretests and knowledge of results. We
have found that pretesting trainees on
the information to be learned increases
the amount of the learning.12 Also,
when practical, it is worth while to in-
form trainees of the results (both errors
and correct responses) of their attempts
to master the learning tasks presented by
means of sound films.13

4. Color vs. black-and-white films. We
have preliminary findings which suggest
that most learning tasks may be presented
equally effectively by monochrome or
color films. 14 Color films have an advan-
tage when crucial cues for learning de-
pend on color. If the thing to be
learned (such as the identification of
flowers, kinds of wood or geological
specimens) depends heavily on color,
then the discriminative learning may
best be done from color films. How-
ever, it would seem there are subtle and,
as yet, unmeasured distractive effects in-
volved in instructional color films.

5. Special effects. Studies suggest that
devices used in films solely to gain atten-
tion, such as "stop-motion" shots of still
objects, pictures of pretty girls, arresting
sounds or noises and unusual angles used
purely for their striking effect, have little
or no influence on improving the learn-
ing scores of trainees. Such devices do
not seem to justify the effort and expense
of including them.15

An experiment was recently completed
on the effect of opticals on learning from
instructional films.16 Three versions of
each of two different informational films
were prepared. One version of each
film had no opticals (straight cuts all the
way through); a second version had
fades only between major divisions of the
film; and the third version, which was
the film as originally produced, had a
liberal number of fades, wipes and dis-
solves used in accordance with the gener-
ally accepted "rules" for using these
effects. The learning results showed no

Carpenter and Greenhill: Instructional Film Production

425

significant differences between the ver-
sions.

6. Use of films exclusively. It has been
shown repeatedly 4-6-7'8'17 that good sound
films can do an instructional job without
the aid of highly trained instructors.
Thus, adequate films can release instruc-
tors from much instructional routine and
give them time for instruction of the kind
which cannot be done by films, e.g., per-
sonal attention to individuals, or apply-
ing instruction to immediate specific

situations. Furthermore, film-to-trainee
instruction puts responsibility for learn-
ing directly on the trainee where it ulti-
mately must rest and reduces the re-
sponsibility which the instructor must
carry.

7. Practice in learning from films. Finally,
some of our results strongly suggest that
students learn to learn from films.18 In
other words, practice in learning from
films increases the facility with which stu-
dents acquire information from subse-
quent films.

PRACTICAL IMPLICATIONS

1. In the production and use of instruc-
tional and informational films full cog-
nizance must be taken of the character-
istics, abilities and limitations of the
people in the audiences who are to be in-
structed and informed. In order to
check whether a film is suitable for the
characteristics of the target audience and
whether it achieves its objectives, it is
believed to be necessary to conduct
"proving-ground tests" on the effective-
ness of the film with samples of the in-
tended audience.

2. Tests of instructional efficacy or
learning need not be delayed until films
are completely produced, but they may
be conducted at several stages during pro-
duction and prior to release and distri-
bution.

3. No single film can be entirely suit-
able for an audience with a wide range of

backgrounds and abilities; therefore,
multiple versions of films which permit
great flexibility of use are desirable to
meet the needs of different audience
levels.

4. Existing methods of film production
and utilization can be greatly improved
by applying psychological research meth-
ods and results. By using suitable
films as the main medium of training,
high levels of effective instruction can be
achieved.

5. Research on the functional char-
acteristics of sound motion pictures and
television, as these interact with audi-
ences, is equal in importance to research
on equipment and technical processes.
The human engineering approach would
seem to be essential for further important
advances in the communications indus-
tries.

REFERENCES

1. C. R. Carpenter, R. C. Eggleton, F.
T.John and J. B. Cannon, "The Class-
room Communicator," Technical Re-
port SDC 269-7-14, The Instructional
Film Research Program, The Penn-
sylvania State College, 1950.

2. C. R. Carpenter, R. C. Eggleton, F.
T. John and J. B. Cannon, "The Film
Analyzer," Technical Report SDC
269-7-15, The Instructional Film Re-
search Program, The Pennsylvania
State College, 1950.

3. S. F. Harby, "The development of a
procedure for using daylight pro-
jection of film loops in teaching skills,"
report in preparation, The Instruc-
tional Film Research Program, The
Pennsylvania State College.

4. J. Murnin, W. Hayes and S. F.
Harby, "The use of film loops as the
exclusive means of teaching motor-
skills," report in preparation, The
Instructional Film Research Program,
The Pennsylvania State College.

426

May 1952 Journal of the SMPTE Vol. 58

10.

11

P. Ash and N. Jaspen, "Optimum
utilization characteristics of the rear-
projection daylight screen," report in
preparation, The Instructional Film
Research Program, The Pennsylvania
State College.

S. Roshal, "Effects of learner repre-
sentation in film-mediated perceptual-
motor learning," Technical Report
SDG 269-7-5, The Instructional Film
Research Program, The Pennsylvania
State College, 1949.
N. Jaspen, "The effects on training of
experimental film variables,' Study
II." Technical Report SDC 269-7-1 1 ,
The Instructional Film Research Pro-
gram, The Pennsylvania State College,
1950.

N. Jaspen, "The effects on training of
experimental film variables, Study I,"
Technical Report SDC 269-7-17, The
Instructional Film Research Program,
The Pennsylvania State College, 1950.
J. V. Zuckerman, "Commentary varia-
tions: level of verbalization, personal
reference, phase relations in instruc-
tional films on perceptual-motor tasks,"
Technical Report 269-7-4, The In-
structional Film Research Program,
The Pennsylvania State College, 1949.
W. S. Vincent, P. Ash and L. P.
Greenhill, "Relationship of length and
fact frequency to effectiveness of
instructional motion pictures," Tech-
nical Report SDC 269-7-7, The
Instructional Film Research Program,
The Pennsylvania State College, 1949.
C. W. Lathrop, Jr., and C. A. Norford,
"Contributions of film introductions
and film summaries to learning from
instructional films," Technical Report
SDC 269-7-8, The Instructional Film

Research Program, The Pennsylvania
State College, 1949.

12. J. Stein, "Motivating effects of pre-
tests on learning," report in prepara-
tion, The Instructional Film Research
Program, The Pennsylvania State
College.

13. R. Hirsch, "Effect of knowledge of
results on learning from instructional
sound motion pictures," report in
preparation, The Instructional Film
Research Program, The Pennsylvania
State College.

14. A. W. VanderMeer, "Relative effec-
tiveness of color and black and white
in instructional films," report in
preparation, The Instructional Film
Research Program, The Pennsylvania
State College.

15. D. M. Neu, "The effect of attention
gaining devices on film-mediated learn-
ing," Technical Report SDC 269-7-9,
The Instructional Film Research Pro-
gram, The Pennsylvania State College,
1950.

16. J. Mercer, "The influence of optical
effects and film literacy on learning
from films," report in preparation,
The Instructional Film Research Pro-
gram, The Pennsylvania State College.

17. A. W. VanderMeer, "Relative effec-
tiveness of instruction by: films ex-
clusively, films plus study guides, and
standard lecture methods," Technical
Report SDC 269-7-13, The Instruc-
tional Film Research Program, The
Pennsylvania State College, 1950.

18. A. W. VanderMeer, "Effects of film-
viewing practice on learning from
instructional films," report in publica-
tion, The Instructional Film Research
Program, The Pennsylvania State
College.

Carpenter and Greenhill: Instructional Film Production

427

Film Production Principles-
The Subject of Research

By KEN KENDALL

The considerable number of technical and progress reports issued by the
Instructional Film Research Program at The Pennsylvania State College for
1948-50 are reviewed. The results of the research are reported and assessed
particularly as to their possible meaning for other production as well as for
instructional films.

THOUGH by an overnight snowfall,
the face of the art of communication
has been changed. By comparison with
what is coming, television is reactionary,
a mere extra convenience in a new
science. The scope of this revolution
has even changed the meaning of the
word "engineering." This is the story.

Originally engineering meant the art
of managing an engine. Yesterday it
came to mean the science of making
matter and power useful to man. Today
it has come to mean a new science, the
science of making man useful to man
and machines fitted to man. This is
called Human Engineering.

To do anything with man, to use him
in any way at all we require a tool.
That tool is communication. Human
Engineering can only be as effective
as the completeness of the mastery of
the mechanics of that tool. Thus the
engineer's striving toward better com-

A contribution, made at the request of
the Society, by Ken Kendall, The National
Film Board of Canada, John St., Ottawa,
Canada.

munications has now shifted its emphasis.
The old objective of higher fidelity is
now a by-product of the art. The old
purpose was to carry a message — an
audio-visual message right up to the
point of impact on men's senses. The
new purpose today is to have com-
munications go behind the eyes and
ears and include the mind, the whole
man and his behavior. It is now re-
quired of good communication engineer-
ing that the message be recollected and
used by the recipient.

It is worth while thinking of this in
terms of money — of communication
costs. The revenue-producing motion
picture now has a lusty young com-
petitor. It is the sponsored picture.
A documentary, an advertising film, a
television program, a teaching or train-
ing film — any and all of these may be
sponsored and shown without a box-
office purpose in mind. There are other
purposes.

Yet these two masters, the box office
and the sponsor, require very different
production psychologies. This is be-

428

May 1952 Journal of the SMPTE Vol. 58

cause each has a different way of getting
back its money.

Box-office returns depend on fleeting
entertainment values. Memory of the
picture is not too important. In fact,
should viewers forget their first ex-
perience and see the show again the
returns would increase.

Sponsors, however, seek to purchase
pictures whose details will be specifically
recollected. With a sponsor, the finan-
cial justification of a picture depends on
this, and on how well the message is
recalled, the information used or the
suggested course of action followed.

Inevitably, therefore, the engineering
evaluation of the system and the planning
of film or television production cannot
be divorced or isolated steps in a chain
of events. The .costs of transmission
fidelity now share in the balance with
the evaluation of the fidelity of human
reactions. Like it or not, picture pro-
ducers, wherever engaged by sponsors,
are now working in a branch of human
engineering and must follow the engi-
neering approach of understanding and
putting to use the findings of scientific
research.

Picture producers and scientific re-
searchers are apt to view any work
wedlock between themselves with mutual
horror. Artists, writers, musicians and
producers have long told the world that
all science is a shackle, while research
groups test everything until it yells
for mercy. Again, most researchers try
hard to withhold their findings until
proof is completely unchallengeable, a
point at which it has usually become
totally incomprehensible to the layman.

The researchers say, "Research must
be complete before it can be divulged."
Others argue, "But think! Research is
never complete. The whole of human
progress stands on imperfect knowledge,
or incomplete but continuing research."

Of extreme news value to the motion
picture and television field is the fact that
one of the human engineering research
groups has decided to allow its interim

findings to be reported. The sponsor-
ship of this research is a joint one:
the United States Departments of the
Army and the Navy have a human
engineering project conducted under
contract by The Pennsylvania State
College.

The following summary is digested
from reports covering over sixty re-
search projects in various stages of de-
velopment. These concern the elements
of effectiveness in films which alter
audience behavior. A defense project,
this research sought to measure the
amount of learning produced by special
films used in the training of thousands of
men. As in all valid science, opinions
and judgments not based on test results
have been meticulously excluded from
the findings or, in some cases, stated only
tentatively.

Concerning the Research Method

Skip this section if you wish. It con-
cerns the question of the validity of the
new findings. Experienced producers
have and will take exception to them.
They feel that research teams are not
"showmen." Are such teams, they ask,
qualified to judge? Similar questions
may rightly be expected from all who
are connected with motion picture and
television production.

The heart of this challenge rests
squarely on the matter of the opposed
purposes of box-office and sponsored
motion pictures. The greater part of
the producer's "expert opinion" as to
the desirable factors in the production
of motion pictures has been based either
on box-office returns, or on audience
entertainment reactions in theaters, or
both. All such evaluation tends to
measure only the entertainment value
at the time of seeing the picture.

But all sponsored pictures are funda-
mentally informational; therefore the
sponsor (unconsciously or otherwise) is
buying a package of "learning" which
he wishes the audience to accept and
retain. Thus the picture industry's

Ken Kendall: Production Principles Research

429

evaluation of entertainment elements is
not relevant to efficiency of informa-
tional films.

For such reasons, scientific research
into the dynamics of learning from films
has had to start from scratch. The
measured actual behavior changes in
individuals exposed to informational
films have formed the basis of all the
findings. As with all valid science,
the expert opinions of those in the art
are held to be unreliable until inde-
pendently confirmed by actual tests.

This does not mean that the experience
and judgment of experts have been
excluded from this research program.
The number and caliber of learned
people connected with the project are
formidable indeed. A later section of
this paper gives a partial list of the
advisers, consultants and researchers
who are or have been involved in this
project.

The project, called "The Instructional
Film Research Program," is under the
direction of Dr. C. R. Carpenter, Pro-
fessor of Psychology, The Pennsylvania
State College; and Mr. Leslie P.
Greenhill is the Program Coordinator.
The scope of this work is shown by the
following approximate outline of the
personnel and physical facilities used
to date:

Personnel

Advisers and consultants 19

Researchers and associates .... 33

Engineering and development staff . 9

Film production staff 5

Joint military advisory body ... 7

Cooperating university divisions . . 7

Unit audience (instant reaction) . . 40

Other unit audiences, average . . 70
People used in experiments, approx.

50,000

Tools Used

Special daylight projectors.
Combination workrooms with concurrent
projection facilities which enable in-

stant putting into practice of film-
taught skills.

Special small projection theater.

Complete instructor or researcher public
address facilities enabling dual film
and personal instruction to take place.

Special repetitive impact machines.

Special "classroom communicator" or
computating multichannel audience
reaction machine.

Special "film analyzer" polygraph or
synchronous and continuous recorder
of audience reactions, decisions and
judgments.

Commercial "Pressey Teach-Test" de-
vice.

Testing facilities enabling films in
production to be tried for efficacy on
sample audiences.

Separated audio and visual channels,
permitting audio teaching impact
versus visual teaching impact to be
measured separately and together.

Three-dimensional film facilities; use
of color film.

Complete production of experimental
films in which the various elements
influencing learning can be segregated
and tested.

78 special experimental films produced
by IFRP for this research ranging
from 2 to 30 minutes in length to fit
the needs of twelve of the projects.

A number of existing films modified to
meet the requirements of other proj-
ects.

Clerical system to handle individual
viewer recollection fidelity over speci-
fied time lapses.

Computing machines for statistical
evaluation of the research measure-
ments.

Research Trends

Starting with 45 original research
projects (see list at the close of this
report) the program to date has formu-
lated and undertaken over 60 research
projects. They can be classified into
four main categories:

430

May 1952 Journal of the SMPTE Vol. 58

1. Research on the governing factors
of films which are expected to change
the personal attitudes of the viewer.
This attitude change includes
opinions, values, beliefs and re-
sulting actions which may be found
to last in a given person.

2. Research on the governing factors
of films which are expected to impart
concepts, facts and principles to
viewers.

3. Research on the governing factors of
films which are expected to impart
skills and arts to viewers.

4. Research on the development of re-

search tools for this new branch_of
human engineering.

This outline of the Instructional Film
Research Program should be appraised
against its purpose. Sponsored by the
U.S. Department of Defense through
the Human Engineering Branch of the
USN Special Devices Center, this project
has been a specific inquiry into the fac-
tors controlling "rapid mass learning."
The results have disclosed exact infor-
mation about the processes (production
concepts) which effectively influence
human behavior through film or tele-
vision. Here are the findings:

Film and Video Production Principles Revised by Research Results

Like radio and the press, film and
television are transmission devices only.
The media of communication are the
symbols used, be they words or pictures.
Loose thinking has confused this matter
so much that so-called research has been
directed to establishing which trans-
mission device, radio, television, film,
etc., possesses the greatest audience
influence. Thus the question of fidelity
of transmission has been confused with
fidelity of learning to produce dense clouds
of meaningless statistics and much
frustration in production craftsmen.
Certain definitions in regard to com-
munication symbols have been estab-
lished and they can be briefly reviewed
to advantage here. (For brevity, the word
film is extended to include video.)

Music is a symbolic form. It articu-
lates concepts frequently difficult to
express in language or in photographs.
It symbolizes moods and feelings, emo-
tions and tensions. The fact that we
cannot always name the mood, emotion
or tension conveyed by music is in itself
evidence of the symbolic character of
music, and of its ability to communicate
meanings which are not verbal.

Language on the other hand is a
symbolic vehicle of thought and of
reason. It is an instrument of naming

and conceiving objects and of combining
and manipulating concepts and proposi-
tions. Language is also the vehicle
by which higher intellectual processes
are carried on.

Photographs are direct symbols of an
elementary or literal type acting as a
record of a visual experience.

Motion pictures combine all three of
these symbolic forms. It is from the
use and integration of these symbolic
forms and from their richness in cues to
concepts already formulated by the viewer
that motion pictures derive their enor-
mous potential power to influence human
behavior. Thus, while all communica-
tions are made by means of symbols,
motion pictures are, perhaps, most
complicated in this regard.

This complexity of symbolic medium,
when considered in film production,
gives rise to the following rule. The
photographer or producer must actively
exclude the mere "literal record" aspect
of his science. Each picture must
concentrate on communicating its
meaning-evoking content. This means that
every effective shot or picture must have
a specific symbolic meaning. This inner
meaning which is "seen" by the script
writer or photographer must also overlap
the specific experience history of the

Ken Kendall: Production Principles Research

431

intended viewer. No significant or
lasting recollection is created unless
these two considerations have governed
the film's production. And of these
two, the question of the overlap of the

real experiences of the viewer is most
frequently apt to be neglected. The
point most often forgotten is that what is
obvious to the expert or film producer
may be anything but clear to the viewer.

Learning Accelerators Found Effective for Films

Factors Related to the Effectiveness

of Films for Teaching Performance Skills

So far, 13 factors involved in film
production concepts for teaching skills
have been the subject of research. They
were tested to find which of them posi-
tively improved the learning which
results from seeing a film that is intended
to teach the audience how to do some-
thing. The factors which have been
tested are as follows:

1. Level of verbalization.

2. Explanation of "how it works."

3. The use of technical or specialist
terms.

4. Audience participation.

5. Condensed or succinct treatment of
the subject

6. Rate of development of subject
matter.

7. The showing of errors to be avoided.

8. The effect of several showings to a
viewer.

9. The effect of different camera angles.

10. Effect on learning of motion versus
still pictures.

11. Showing the hands of the operator
in a film demonstration of a skill
being performed.

12. Effect of personal reference in the
commentary.

13. Effect of various time relations
between verbal instructions and
picture.

Of these thirteen factors which are
called "variables" by the researchers,
those which follow proved to have a
positive and most significant influence '
on the effectiveness of films designed to
teach new skills.

1. Verbal level. The amount of com-
mentary used to describe the action

affects the performance of the viewer.
The amount, as measured, rises in
efficiency slowly up to 100 words per
minute of film and falls as gradually
beyond that level. Effectiveness is
down by 25% at 40 and at 140 words
per minute of film.

4. Participation. Audience participa-
tion or practice proved to be most
effective as a utilization device under
suitable conditions. The rate at which
the commentary and picture presenta-
tion is developed controls this element.
Rapid development plus participation
acted against learning and retention.
Slow development which allowed time
for the viewers to watch the screen and
practice the skill helped learning con-
siderably. An alternative to the slow
rate of development is the use of a
medium rate of development with oppor-
tunities to practice between film showings.
It was noted that most instructional
films used verbal and visual develop-
ment rates common to theatrical films.
These rapid rates were found to be the
least effective in promoting viewer
recall and performance.

6. Rate of development. A slow rate
of development is a most important
factor in making a film effective. New
information in films should be covered
pictorially at & speed which is appro-
priate to the abilities of the viewer.
The customary practice of present
production is far too fast a rate of
development.

7. The showing of errors. The showing
of common errors or faulty methods to
be avoided increases the instructional
value of a film. The "right and wrong
way" is a most potent film device pro-

432

May 1952 Journal of the SMPTE Vol. 58

vided errors are made explicit and
clearly shown as errors.

8. Effect of multiple showings. Repeti-
tion of the film demonstration ma-
terially aids its impact on the viewer.
It is recommended that the film itself
incorporate a repetition of basic se-
quences, perhaps with slight variations.
(A suitable film, using slow develop-
ment, inbuilt repetition, and right versus
wrong methods was found to be effective
in teaching 98% of Navy trainees how
to do a gun assembly task. A single
showing and no other instruction was
used.)

9. Camera angles. On the basis of the
tests, it has been shown that learning
of a new performance improves as the
film visualization approaches the repre-
sentation of the viewer himself per-
forming the act. Thus a training film
is most effective when a zero degree
camera angle is used. The usual "posi-
tion of an observer" or 180° camera
angle was shown to be less satisfactory.

10. Motion versus still views. The com-
munication and teaching of action be-
comes effectively transmitted when the
film shows all the movements involved in
the doing of the new task. A series of
static shots which show various steps in
the action has been found relatively
less effective in the teaching process.

72. Personal references in commentary.
The use of strong directive statements in
teaching-type films is, in general, likely
to promote greater learning than the
commonly used impersonal type of com-
mentary. Example, "do this" (impera-
tive) or "you do this," etc. This finding
is slanted toward military training
objectives.

13. Phase or time relationship of audio
to visual. The use of commentary which
is in advance of the picture and alerts
the viewer to forthcoming visual ele-
ments of importance may be desirable.
The order of "lead" found most con-

tributory to learning in a specific experi-
ment amounted to a few seconds.

Of the measurements on the other fac-
tors in the list of thirteen, the following
were found to be inconsistent or negative:

2. "How it works" film explanations.
This variable showed results which were
very inconsistent. Research is con-
tinuing.

3. Use of technical terms. The intro-
duction of new names or technical
terms was found to impede the learning-
and-recall aspects of skills or perform-
ance-type instruction films where per-
formance was the measure of learning.

5. Succinct treatment of subject. Con-
densed or succinct treatment of the
subject was found to give exceedingly
ineffective results. While educators
have been aware of this danger, many
film producers and technical advisers
seem to consider compact productions
to be satisfactory. Perhaps cost controls
the matter. The term "succinct treat-
ment" is used to refer to a production
practice using a fast rate of development,
minimal use of repetitions, and generally
presenting only the bare essentials of the
subject to be remembered. While this
produces a complete minimum film
presentation it was found to be 600%
less effective than the better experimental
teaching films.

77. Showing hands of the operator per-
forming task. In the one case where a
comparison was made between showing
the hands of the operator and not show-
ing them (moving objects by stop mo-
tion) results were inconclusive.

Factors Found Directly Related to the
Effectiveness of Informational Films

The evidence indicates that films
designed to impai^ information effectively
parallel, to a large extent, the factors
needed for skills instruction.

Slow rate of development and the use
of built-in repetition are important
contributors to effectiveness.

Ken Kendall: Production Principles Research

433

it has been found that film introduc-
tions and summaries may be designed
to boost considerably the information
retained by the viewer. A poor intro-
ductory sequence may mislead the viewers
and impair learning.

People will retain more of the informa-
tion shown when they are intentionally
made aware of the amount that they have
learned from the film. Such a "knowl-
edge of results" may be made a part of
the film. Commentary and flashback
in support of a "how much do you
remember" recapitulation may imply or
lead to self-evaluation results. For
teaching functions students learn far
more when they are specifically kept
informed of their progress.

Color films were found of help only
when the color was of crucial importance
in the imparting of a specific concept or
"crucial cues" in the film.

It has been well demonstrated that
information can be taught exclusively
by means of films. Groups of viewers
having no previous knowledge of the
subject, who were supervised by un-
skilled instructors, learned as much from
a series of films as equivalent groups who
were trained by expert instructors.
(Industry and military sponsors please
note.)

Instructional films, if they are to be
effective, must match the viewer's back-
ground experience and abilities in both
an auditory and a visual sense. This
indicates:

1 . Several versions or treatments adapted
to different audience levels;

2. the adoption of "production-stage"
showings using sample audiences
who are later subjected to recollection
tests and other reactions to the film
in preliminary form.

Factors Found of Interest
in Attitude-Changing Films

So far, the research evidence suggests
that film concepts influencing human
behavior are most likely to be perceived

and adopted if they do not conflict
with prior opinions and belief systems
of the viewers.

The established attitudes of viewers
toward a film's main character and
theme are matters of importance to the
film's effectiveness in modifying attitudes.
From a production point of view, the
effectiveness of a film designed to change
audience attitudes will depend on the
selection of a main character who is
suitable for "hero-worship" or "identi-
fication mechanisms" set up in the
intended audience. Films which exclude
the identification mechanism as a per-
suasive factor have not been proven
capable of effectively altering human
attitudes or behavior.

Functional Factors of Importance to
Effective Learning From Films

A preliminary evaluation of the
influence of functional factors on learning
was measured by scoring Navy personnel
after film instruction. About 2,000 men
were used for these tests.

7. Viewing angle and screen distance. The
optimum viewing area for a small
"daylight" rear projection screen was
established as about the same for both
daylight and dark viewing conditions
within an angle 60° wide. A negligible
decrease in learning was found at
viewing distances up to 16 screen widths
from the screen (24 ft for an 18-in.
screen). Of the men in the tests, 75%
learned successfully in these positions
and at these distances.

An area of reduced learning effective-
ness extends around the optimum area
to a maximum total angle of 100° for
daylight viewing and 120° for viewing
in the dark. The maximum viewing
distance was 24 screen widths. Within
this area 50% of the men learned success-
fully.

These results have important implica-
tions for film and television. The
present standard of six times the screen
width as the maximum satisfactory view-

434

May 1952 Journal of the SMPTE Vol. 58

ing distance might be revised for in-
formational film theaters, etc.

2. Repetitive showings. The repetition
of film showings added greatly to learn-
ing. Three showings resulted in signi-
ficantly more men mastering an assembly
task than did two showings.

3. Consecutive versus spaced film showings.
The measurements indicated that consec-
utive presentation of several films was
as effective in ensuring two-week reten-
tion of the film information as showings
of the same films spaced over several
days. The establishment of the efficiency
of hour-long showings is of importance
to military training applications.

Music in Informational Films

So far experimental evidence in this
connection is not conclusive. The
following discussion is suggested by the
traditional psychological principles of
learning, without the supporting evi-
dence of experimental tests.

The researchers point out that the
kind of music normally attached to
informational films will certainly not
operate to accentuate learning. Re-
search has indicated that the mere
presence of music may operate in many
films to distract and divide attention,
or to give the entertainment "set" for
viewers. The following relationships
suggest useful preliminary trends of
thought to be confirmed by future experi-
mentation.

Attitude and opinion molders. Music
which is regarded highly by the audience
might be used to set up favorable
attitudes toward the audio-visual ma-

terial in the film. The same device
could operate in reverse and help form
unfavorable attitudes.

Memory reinforcing. The strengthening
of new learning by association of the
familiar with the unfamiliar may be
assisted by the use of familiar music
as a framework to aid recall. The
repetition of music with a given visual
and its variations might be a desirable
memory link.

Concept-forming aids. Music might be
used as a clue suggesting association by
inference with a new experience not
previously related to familiar ones.
In the same way, the function of music,
used as a clue, might aid in pointing
toward a problem's solution.

Emotional drives for learning. Music
might provide an emotional tone or
excitement to the learning experience.
Correctly conceived informational films
are designed to leave unanswered questions
in the mind of the viewer toward the
solution of which he must actively
participate. Music might be used to
provide a kind of reward, in that the
viewer would feel pride in recognizing
correctly the association intended by
the music.

Music as a "pointer." Music might
be used to direct attention to a par-
ticular occurrence in the visual stream
or in the sound track. It might provide
a source of direction for attention by
overcoming previous distractions. Con-
trasting tone color might be used to sus-
tain attention for long periods and
prevent day dreaming.

Interpretation of Research Findings

Some Values of Film in Education,
Instruction or Informational Roles

subjects. A longer retention of new informa-
tion has been found in some cases than was
achieved through other methods of instruction.
Discussion. The origin of this research
was a postulation of a shortage of compe-
tent instructors in the event of a national
1. Films have been found to be equivalent emergency. The study was undertaken
to a good instructor in teaching specific to determine to what extent instructional

"Do films tend to be of significant
value if used for education, training or
information?"

Ken Kendall: Production Principles Research

435

films could carry the entire teaching
burden using supervisors only.

As an example of the procedure by
which this type of research was con-
ducted the case is cited of three com-
parable groups of 9th grade high school
students who were taught a four-unit
course in general science over a period
of a full semester. The first group was
taught exclusively through a series of
44 films. The second group was taught
through the medium of the same films.
They also studied before and after each
film from specially prepared short
study guides. The third group was taught
by competent teachers using a standard
textbook and the customary classroom
teaching techniques. No films were
used.

The students were given objective
tests immediately at the end of each unit
of training. Three months after the end
of the course a retention test covering
all four units was given. They were
also tested both before and after the
experiment with a standardized test of
general science knowledge.

Test Results. Analysis of the test
results from the three groups revealed
that the three methods were of almost
equal effectiveness. The immediate
recall tests gave slightly better scores in
favor of the films plus study guides
group over the classroom-taught group;
the films only group was slightly inferior
to both of these. The delayed recall
tests indicated slightly better perform-
ance for the two film groups over the
classroom-taught group. Of the two
film methods the one using films plus
study guides was slightly more effective.

Conclusions. The results of this study
suggest that subject matter such as high
school general science may be taught by
films alone as effectively as by a good

teacher using the usual repertoire of
classroom techniques and demonstration
materials. Films introduced and supple-
mented by brief study guides are better
still. It is worth noting that the films
selected for these teaching tests were pro-
duced without the benefit of the current
research into the dynamics of learning
through the medium of films. Nor did
they make up a systematic series which
thoroughly covered the subject matter.
Correctly produced films should be able
to provide a more than adequate solution
for a shortage of competent instructors.

2. People can learn more in less time and re-
tain longer the information derived from films
made on these principles.

This has been demonstrated repeatedly
when films have been tested against com-
parable reading materials or lecture
presentations. The films required less
instructional time. They imparted far
more factual learning in the same time.
Films in combination with other instruc-
tional materials are perhaps better than
either alone. This holds for both the
immediate and delayed measurements of
the learning effects. Audiences have
been found to remember more from a
film after serveral weeks than after a few
days or a few hours. These findings held
true even when the films used in the
tests did not conform completely to the
principles disclosed in the current re-
search.

3. Informational films have been designed to
stimulate other learning activities.

When films incorporate established
principles of learning they stimulate in
viewer groups adult activities such as
group discussions, teamwork and the like.
Individual viewers have been induced to
engage in voluntary reading. Motiva-
tion or desire to undertake the develop-
ment of new skills was induced and fol-
lowed by effective action.

436

May 1952 Journal of the SMPTE Vol. 58

4. Films have been found to facilitate thinking
and to aid in problem solving.

Evaluation of the evidence of film re-
search clearly indicates that the contri-
bution of films as a communication
medium is to give greater comprehensipn
and understanding rather than to de-
velop specific detailed recollection. Re-
search studies have demonstrated that
people taught by film were better able
to apply their learning than people who
had other forms of instruction.

5. In short, films can be specifically produced
to influence viewer behavior in the long-term
sense.

Thus through carefully produced films
human beings can effectively acquire fac-
tual knowledge, the development of new
skills, the formation of new attitudes,
motivations and opinions. Films may be
expected to effect other educational ob-
jectives such as appreciation, orientation,
etc.

Evaluation and Summary of Experimental Research Have Suggested Principles
Which Govern the Dynamics of Film Influence on Behavior

For convenience, each principle is
stated and discussed and some practical
implications for film production or utili-
zation are suggested.

The overriding concept in the knowl-
edge of the dynamics of film influence is
that the meaning of a motion picture
always differs among the people who see
the film. What is perceived in a given
film will differ with each viewer and will
condition the meaning of the film to him.

In order to conform to the dynamics of
learning, a film must contain familiar ele-
ments or backgrounds, and not too much
that is completely new or unfamiliar such
as interpretations that are not easily rec-
ognized. Yet there must be new infor-
mation or there will be little learning.
At the same time, understanding will be
blocked should the information be too
new or too difficult. To complicate
matters, adult audiences demand or ex-
pect material which has the appearance
of novelty. Otherwise they feel that
their learning ability has been under-
estimated. Therefore a good training
film must appear to challenge the
viewer's learning ability. Thus it is
essential to understand that while a mo-
tion picture does not vary objectively
from one showing to another or from one
group to another, there will be a varia-
tion in its meaning for different indi-
viduals. This will depend on the inter-
action between the psychological char-
acteristics of the viewer, the social cir-

cumstances surrounding the audience
and the content and treatment of the
film. Any effective informational film
will owe its validity to the matching of
these variables.

First Principle — Films possess their greatest in-
fluence when their content has been designed to
reinforce and extend the previous knowledge,
attitudes or motivations of the viewer.

Discussion. A film will not substantially
influence the behavior of a person unless
that person can respond to the film in
terms of what he already knows — or
what he can do — or how he feels — or
what he wants. The film can be de-
signed to help change his attitudes and
opinions, his knowledge and his skills,
provided that it extends or reinforces
those elements which he already
possesses.

The effects of any motion picture de-
pend on the reinforcing of the viewer's
experiences which preceded, follow or are
coincident with the actual film showing.
Tests have shown that the influence of
any one film is limited while the influence
of several films is cumulative in the dy-
namics of learning.

Application. The sponsor's money will
be wasted if the film is not carefully
adapted to the viewer's knowledge-level,
or if the film content is allowed to run
counter to existing attitudes or motiva-
tions of the viewer.

Ken Kendall: Production Principles Research

437

From the sponsor's point of view,
whether an influencing film is expected
to extend and reinforce, or to reorganize
and redirect the present behavior of the
intended audience, a given film is in-
effective unless it is planned, produced, dis-
tributed and used as one of a series of re-
lated and cumulative experiences operat-
ing in a common direction and all de-
signed for the same specific viewers in the
audience.

When it is the purpose of the sponsor to
redirect behavior patterns and to reorient
the motivations of an audience such as a
group of Navy trainees, it may be neces-
sary to reinforce the film with comple-
mentary impacts through other nonfilm
avenues of instruction.

A second principle is that the behavior-influenc-
ing impact of film is usually specific and not
general.

Discussion. The principle that films
have a specific effect holds for all infor-
mational objectives. The cumulative
effect of related films shown over a period
of time and/or reinforced by other means
of instruction may be general. Even
here, however, this general influence is
limited to the area of the instructional
content of the films.

Application. From the production point
of view, the sponsor has to be brought to
the realization that instructional or infor-
mational films must be designed to
achieve very specific objectives. A state-
ment of film objectives in general terms
is of little value to either a sponsoring or a
producing agency.

Failure to define the film objectives
specifically at the planning stage of pro-
duction is a handicap which makes it
highly improbable that the film will be
effective in influencing behavior or other-
wise creating conditions for viewer recol-
lection. (Throughout this digest that
aspect of viewer recollection which re-
lates to the entertainment value of the
film is excluded as being irrelevant to
educational objectives.)

The third principle is that required film in-
fluence increases directly as the content of the
film matches the specific audience response re-
quired by the sponsors.

Discussion. The subject of the film and
the way that subject matter is treated is
instrumental always and only to a spe-
cific end product of audience response.
This means that the behavior pattern
that the film is intended to produce must
be directly related to the content and
treatment of the film.

Application. It is necessary for the film
sponsor to spell out the instructional or
informational objectives in terms of the
specific behavior the film is intended to in-
fluence. This means sponsors must indi-
cate what or how the viewers are ex-
pected to know, think, feel or do as a
result of seeing the films they buy.

When the film purpose is established in
this specific way, production time, fa-
cilities and expense can be materially
saved by the omission of content and
treatment irrelevant to the specific be-
havior the film is intended to produce.

The effectiveness of a given film may
be increased by audience participation
relevant to the informational objectives.

The fourth principle is that variations in the
prejudices or predispositions of the audience in-
fluence the reactions to a specific motion picture.

Discussion. Some elements of these
variations depend upon audience liter-
acy, abstract intelligence, formal edu-
cation, age, sex, or previous experience in
the subject. Differences in heredity and
social experience mean equivalent differ-
ences in reaction to the film, and these
differences seem to increase with ma-
turity.

It has been found that intelligence and
formal education are directly related.
Viewers of above-average intelligence
and education learn more from films
than those with average or below-average
education.

Below-average education viewers learn

438

May 1952 Journal of the SMPTE Vol. 58

very much better from films than from ver-
bal instruction.

The retention of film content has been
found to decline with age after a certain
point.

Sex differences in response occur when
the values or occupations shown in the
film are sex-typed.

A film has bias but the bias of the
audience also counts. The recollection
tendency of the viewer depends on his ac-
ceptance, rejection or indifference to the
bias of the film.

Tests show that the more an audience
knows about a given subject the more it
will learn from a film on that subject.

One interesting point which the re-
search has brought to light concerns the
influence of many films on the same
viewer. The first principle showed that
a series of related film experiences all
operating in the same direction is cumu-
lative. However, the fourth principle
exemplifies the fact that the more films
of any type which are seen the more the
viewer tends to learn from any single
film. People learn to learn from films.

Application. The research has disclosed
that while the behavior-influencing im-
pact of a film may be in the direction of
the bias of the film the force of this im-
pact will vary among the viewers de-
pending upon their respective histories.
To a surprising extent there will be in-
stances of behavior influences the reverse
of those intended by the film. An
effective film will not have this result be-
cause its production is planned and it is pro-
duced and used according to an integrated
psychology using the dynamics of learn-
ing-
Effective informational film planning,
production and use depend on infor-
mation as to the age, attitude, intelli-
gence, education and social outlook of the
specific audience for which the film is
designed. These must be spelled out by
the sponsoring agencies.

If informational films are designed for
a general audience they should be sighted

slightly below the average of intelligence
and education rather than above it.
This practice has been found to be the
most effective treatment. Viewer learn-
ing was measured and it dropped rapidly
when the "sighting" of the film was
slightly above the audience educational
level.

If a sponsor intends to influence audi-
ences of widely different mental levels it
has been found almost essential to have
several versions of the film made for
several IQ's.

The fifth principle is in two parts:
7. Both audio and visual elements of films are
effective channels of communication. Neither
channel is consistently better than the other.
Each channel is uniquely capable of convey-
ing certain types of information and the two
should be properly integrated.
2. The overall influence of the motion picture
is thought to be primarily in the picture and sec-
ondarily in the accompanying language. It is
relatively unaffected by the slickness of pro-
duction.

Discussion. The measurements indi-
cated that the presentation of a film as a
whole or the presentation of either the
audio or the visual channel alone resulted
in significant learning.

Both channels together were consist-
ently better than either one alone. This
"both" factor has been identified. It is
established that some items are learned
jointly from the audio and visual ele-
ments working together. Evidence also
exists to show that items are often taught
via both audio and visual channels in an
overlapping sense, in which case the
cumulative value of the "both" factor is
reduced.

Color film has not been demonstrated
as generally superior in information and
instruction to black-and-white film.

Attention-gaining devices, either visual
or auditory, have not been found to add
significantly to learning in an otherwise
correctly made informational film.

Optical effects and other film tricks
have not been found to contribute sig-

Ken Kendall: Production Principles Research

439

nificantly to learning from informational
films.

Too much or too little talking in words
per minute of film has been found to de-
tract from the teaching effectiveness of a
film. The optimum word rate is about
100 words for each minute of film.

Application. No film should be planned
that does not lend itself to fluent picture
conception and specification.

With sound films equal care should be
given to the verbal conception and speci-
fications.

Since both channels together are more
effective than either alone the objective
is to achieve the best possible integration
of the visual and audio elements of film.
The "both" factor of this integration
should be controlled. That is, single
concepts should be imparted through the
audio and visual channels working to-
gether.

The various attention-getting devices
and other luxuries of entertainment films
are found to be not significant in the dy-
namics of instruction by films and are
seldom noticed by the audience.

The findings of these studies appear to
be relevant to television.

The sixth principle is that the recollection of a
film depends on the viewer's feeling that the
action is significant and is in a familiar back-

Discussion. Not everything shown or
said in a motion picture is seen or heard
by the viewer. His response to film is
selective not photographic. Scenes and
sequences are best recalled when the pic-
torial background is familiar to the
viewer and when the action has specific
meaning to him. What counts is not
the action but the importance of the
action, not the close-up but the signifi-
cance of the objects in the close-up, not
the manner of performing the task but
the meaning of the task to the viewer.

Application. The special forms of visual
and audio film treatment such as cartoon,

live photography, dramatic or straight
treatments are not instructionally or in-
formationally important in themselves.

Whatever form of treatment is used in
the film, it is essential that important
scenes and interpretations be made to
appear important to the viewer.

Light humor helps, but slapstick hurts
instruction and information.

The seventh principle is that an intense, effi-
cient and predictable response occurs when the
picture content has a personal relationship to
the viewer.

Discussion. It has been found that the
influence of the film on the attitude and
factual learning of the viewer is related to
the prestige attached by the viewer to
the role of the principal character.

The position of the viewer, or zero
camera angle, should be used instead of
the 1 80 ° angle which is so frequently used
in informational films. It has been
found that the subjective approach is
important to long-term recollection.

Showing the errors likely to be made
when carrying out the task improves the
instructional value of a training film.

Direct instructions or direct address to
the viewer should be used. The third-
person, passive voice has been found to
retard learning.

Application. The content of an infor-
mational film is always better if it is
treated in the way members of the audi-
ence would see the subject if they were
dealing with it in real life.

The production treatment should be
designed so that the viewer can see him-
self in the picture and identify himself
with the principal characters.

The appropriateness of a film to an in-
tended audience should determine its dis-
tribution. Its photographic excellence or
its appeal to an expert should be second-
ary considerations.

The eighth principle is that the rate of develop-
ment of a film's message must be slow rather
than fast.

440

May 1952 Journal of the SMPTE Vol. 58

Discussion. Where recollection, learn-
ing or information rather than entertain-
ment is involved, a slow, rather than a
rapid rate of development is important.
Rapid development of the presentation
of a film subject reduces the amount ^of
learning very materially.

Application. For maximum recollection
it is necessary to gear the rate of develop-
ment of the subject and the information
of the film to the rate of learning of the
audience. This presupposes that the
audience level of intelligence is known.

The rate of learning of the audience is
generally slower than the film producer
thinks. It is a waste of the sponsor's
money to try to cover too much too
quickly in any one film.

The ninth principle is that instructional tech-
niques built into the film or applied by an in-
structor substantially increase learning.

Discussion. The research conclusively
shows that the following techniques add
to the effectiveness of instructional films:

1. An orienting introduction and a rele-
vant summary of the content of the
film are of significant value.

2. An opening announcement of a
check-up or quiz on the learning
from the film measurably improves
the recollection value.

3. Repeated film showings and/or repe-
tition within the film itself, materially
improve its recollection value.

4. Audience participation or practice
during or following the film snowing,
"locks in" the teaching.

5. Presenting the viewer with a knowl-
edge of the results of his learning is
of great significance.

Application. For rapid mass teaching it
is desirable to make films which under-
take the total training. They should
have instructional techniques and meth-
ods built right into them.

Auxiliary instructors can provide moti-
vation, interest and leadership. These
are necessary because motivation and
morale carry over to the film learning
even if the instructor is not present during
the film showing.

These leaders or instructors should be
trained in the dynamics of learning and
its application to rapid teaching by film.

Conclusion

The unusual and commendable action
of the Instructional Film Research Pro-
gram in publishing interim reports on
research in progress is planned to con-
tinue. These researchers point out that
their main responsibility is to do research
and, therefore, limits must be expected
on the amount of report writing.

The most important implication of this
work is that existing methods of film pro-
duction and film utilization can be
greatly improved by applying research
methods. The conventional research
into factors improving communication
tools such as radio, motion pictures and

television must be extended to become an
exact knowledge of audience-influencing
factors.

If the scope of communication is to
meet the sponsor's requirements, if com-
munication is to be extended behind the
eyes and the ears to include effective in-
fluencing of viewer behavior, then the
human engineering approach is essential
for further advances in motion picture
production and other branches of com-
munication engineering. The com-
munication industries have high stakes
in this emerging field of research and
application.

Ken Kendall: Production Principles Research

441

Present and Former Participants in the Instructional Film Research Program

Advisers and Consultants

Dean M. R. Trabue, School of Education
(Chairman)

Dr. C. R. Carpenter, Director of Program
(Secretary)

Dr. R. Adams Dutcher, Chairman, Re-
search Council

Dean Ben Euwema, School of Liberal Arts

Dean George L. Haller, School of Chemis-
try and Physics

Dr. George F. Johnson, Professor of
Agriculture Extension

E. L. Keller, Executive Assistant, Central
Extension

Dr. Bruce V. Moore, Head, Dept. of
Psychology

Dr. Eric A. Walker, Director, Navy Ord-
nance Laboratory

Dr. P. C. Weaver, Assistant Dean, Dept.
of Education

I. C. Boerlin, Central Extension — Super-
visor, Audio-Visual Aids

S. L. Land, Head, Dept. of Industrial
Education

Hugh C. Pyle, Central Extension — Super-
visor, Informal Instruction

Dr. Stephen Corey, Teachers College,
Columbia University

Dr. Edgar Dale, Ohio State University

Dr. Charles Hoban, Jr., West Chester
State Teachers College

Harold Kopel, Encyclopedia Britannica
Films

Arthur A. Lumsdaine, Director, Yale
Film Research Project

Dr. Mark A. May, Director, Institute of
Human Relations (Yale)

Research Staff and Associates

Dr. C. R. Carpenter, Professor of Psy-
chology, Director

Leslie P. Greenhill, Research Associate,
Program Coordinator

Dr. Philip Ash, Associate Professor of Film
Research

Dr. Nathan Jaspen, Associate Professor of
Film Research

Edward McCoy, Research Assistant

Dr. Hugh M. Davison, Professor of Educa-
tional Research

Charles Mclntyre, Research Assistant

Dr. Harold E. Nelson, Assistant Professor
of Speech

Dr. Albert K. Kurtz, Professor of Psy-
chology

Dr. Kendon R. Smith, Associate Professor
of Psychology

Dr. Kinsley R. Smith, Professor of Psy-
chology

Dr. Charles Hoban, Associate Professor of
Education, The Catholic University of
America

Dr. E. B. Van Ormer, Professor of Psy-
chology

Dr. A. W. VanderMeer, Associate Pro-
fessor of Education

Edward Abramson, Assistant Professor of
Sociology

Dr. James Gemmell, Associate Professor of
Economics

William Hittinger, Research Assistant

John Tyo, Research Assistant

Chester L. McTavish, Doctoral Candidate

Mrs. Marjorie Straube Mertons, Doctoral
Candidate.

Joseph Murnin, Research Assistant

Miss Fanna E. Brown, Writer and As-
sistant in Drama

Sol M. Roshal, Research Assistant

John V. Zuckerman, Research Assistant

Joseph N. Grosslight, Assistant Professor
of Psychology

Dr. William S. Vincent, Professor of
Education

John Stein, Doctoral Candidate

Mrs. John Stein, Doctoral Candidate

John P. Kishler, Doctoral Candidate

D. Morgan New (Psychology), Doctoral
Candidate

Dean S. Northrop (Education), Doctoral
Candidate

Loran S. Twyford (Psychology), Doctoral
Candidate

Miss Mary C. Welch (Education), Re-
search Assistant

Engineering Research and Development

F. T. John, Engineer (Director)
John B. Cannon, Jr., Project Engineer
Reginald Eggleton, Project Engineer
Ray A. Bland, Draftsman
Charles Brouse, Construction Technician
Melhart D. Chelosky, Construction Tech-
nician

442

May 1952 Journal of the SMPTE Vol. 58

William E. Shaw, Construction Technician
Milton C. Stone, Construction Technician
Harris Zeitzew, Construction Technician

Motion Picture and Recording Staff

Frank S. Neusbaum, Administrative Head,

Motion Picture Production
Delmer P. Duvall, Assistant Specialist,

Motion Picture Production
Henry Miller, Associate Specialist, Motion

Picture Production
Paul H. Seitzinger, Assistant Specialist,

Motion Picture Production
Mrs. Marjorie Bloomfield, Secretary

Joint Military Services and
Advisory Committee

Joseph Gaberman, Scientific Officer,
Special Devices Center, O.N.R., Chair-
man

Fred E. Kelly, Signal Corps Photographic
Center, Secretary

Dr. C. R. Carpenter, Director, Instruc-
tional Film Research Program

Dr. A. A. Lumsdaine, Human Resources
Research Laboratories Air Force

Paul Murdock, Army Pictorial Service

L. J. Tate, Bureau of Personnel, Navy

Dr. William Timmons, Navy Photo-
graphic Center

List of Research Reports

Technical Reports

SDC 269-7-2, Music in Motion Pictures:
Review of Literature With Implications
for Instructional Films (Rapid Mass
Learning) May 15, 1949.

SDC 269-7-3, The Relative Effectiveness
of Massed Versus Spaced Film Pre-
sentation (Rapid Mass Learning) June
30, 1949.

SDC 269-7-4, Commentary Variations:
Level of Verbalization, Personal Ref-
erence, and Phase Relations in In-
structional Films on Perceptual-Motor
Tasks (Rapid Mass Learning) Dec. 15,
1949.

SDC 269-7-5, Effects of Learner Repre-
sentation in Film-Mediated Perceptual-
Motor Learning (Rapid Mass Learning)
Dec. 15, 1949.

SDC 269-7-6, Learning Theories and
Instructional Film Research (Rapid
Mass Learning) June 1949.

SDC 269-7-7, Relationship of Length and
Fact Frequency to Effectiveness of
Instructional Motion Pictures (Rapid
Mass Learning) Nov. 1949.

SDC 269-7-8, Contributions of Film Intro-
ductions and Film Summaries to Learn-
ing From Instructional Films (Rapid
Mass Learning) November 1949.

SDC 269-7-9, The Effect of Attention
Gaining Devices on Film-Mediated
Learning (Rapid Mass Learning) Mar.
1950.

SDC 269-7-10, The Effects of Prestige and
Identification Factors on Attitude Re-

structuring and Learning From Sound
Films (Rapid Mass Learning) Mar.
1950.

SDC 269-7-11, Effects on Training of
Experimental Film Variables Study II:
Verbalization, "How-It-Works," No-
menclature, Audience Participation, and
Succinct Treatment (Rapid Mass Learn-
ing) Mar. 1950.

SDC 269-7-12, Effect of Repetitive Film
Showings on Learning (Rapid Mass
Learning) Nov. 1949.

SDC 269-7-13, Relative Effectiveness of
Instruction by: Films Exclusively, Films
Plus Study Guides, and Standard Lec-
ture Methods (Rapid Mass Learning)
July 1950.

SDC 269-7-14, The Classroom Communi-
cator (Rapid Mass Learning) Oct. 1950.

SDC 269-7-15, The Film Analyzer (Rapid
Mass Learning) Oct. 1950.

SDC 269-7-16, The Effects of Inserted
Questions and Statements on Film
Learning (Rapid Mass Learning) Sept.
1950.

SDC 269-7-17, Effects on Training of
Experimental Film Variables, Study I:
Verbalization, Rate of Development,
Nomenclature, Errors, "How-It-Works,"
Repetition (Rapid Mass Learning) Oct.
1950.

SDC 269-7-18, Comparison of the Audio
and Video Elements of Instructional
Films (Rapid Mass Learning) Nov. 1950.

SDC 269-7-19, Instructional Film Re-
search 1918-1950 (Rapid Mass Learn-
ing) Dec. 1950

Ken Kendall: Production Principles Research

443

Progress Reports

Special Report No. 1, Practical Principles
Governing The Production and Utiliza-
tion of Sound Motion Pictures, SDC
Human Engineering Project 20-E-4.

Incidental Report No. 2, Some Aspects of
learning From Films, SDC Human
Engineering Project 20-E-4.

Progress Report No. 9, Instructional Film
Research Program, Nov. — Dec. 1948,
SDC Human Engineering Project 20-
E-4.

Progress Report No. 10, Instructional Film
Research Program, Jan. — Feb. 28,
1949, SDC Human Engineering Project
20-E-4.

Progress Report No. 11-12, Instructional
Film Research Program, Mar. — June
30, 1949, Number, Title, Status and
Progress of Research Projects, General
Summary of Trends of Results: The
Instructional Film Research Program
1 947-1 949, Summary Report on Project
No. 5, SDC Human Engineering Project
20-E-4.

Progress Report No. 13, Instructional Film
Research Program, July — Nov. 30,
1949, SDC Human Engineering Project
20-E-4.

Progress Reports No. 14-15-16, Instruc-
tional Film Research Program, Dec. 1,
1949 — Mar. 31, 1950, SDC Human
Engineering Project 20-E-4.

444

May 1952 Journal of the SMPTE Vol. 58

Audio Visual Instruction Conference

Report by D. F. LYMAN

A,

.N INVITATION to attend the Winter
Conference of the Department of
Audio-Visual Instruction (DAVI) of the
National Education Association was ex-
tended to Boyce Nemec, Executive
Secretary of the Society of Motion Pic-
ture and Television Engineers, by Don
White, Executive Vice-President of the
National Audio- Visual Association, Inc.
(NAVA), and J. J. McPherson, Execu-
tive Secretary of DAVI.

NAVA has been cooperating with
DAVI and the American Association of
School Administrators in preparing speci-
fications for the design of classrooms and
school buildings that will insure effective
use of projected audio-visual material.
In his invitation, Mr. White stated that
two important points require immediate
attention: methods of darkening class-
rooms; and provision in each classroom
of the necessary facilities for projection.
These include essentials such as conduits,
switches and electrical outlets. This in-
formation is needed soon for the archi-
tects now designing the many school
buildings that will be built in the next
few years. Later, guidance will be
needed on questions such as the optimum
size and luminance of the screen, and the
amount of illumination in the room dur-
ing projection. Mr. White invited the
SMPTE to cooperate in locating existing

A report submitted February 15, 1952, by
D. F. Lyman, Eastman Kodak Company,
Camera Works, 333 State St., Rochester
4, N. Y.

technical information and in helping
with any research that may be needed.
The Winter Conference was held by
DAVI at the Hotel Kenmore in Boston
on February 6-9, 1952, during which
there were several meetings on this
point.

Through arrangements made by F. T.
Bowditch, Engineering Vice-President of
SMPTE, the writer of this review went to
the conference and attended the sessions
of Action Planning Section 3, Buildings
and Equipment.

DAVI Organization

As brought out in the business meeting
held on February 8, DAVI is an affiliate
of the National Education Association,
with Headquarters at 1201 16 St., N.W.,
Washington 6, D.C. A staff of six people
handles the correspondence and other
work of the organization. There are
now over 1400 members, and the annual
operating budget is $32,000 of which
$6,000 is from membership dues. DAVI
has 15 Action Planning Committees, of
which 13 are now national in scope.
Many are concerned with general edu-
cational problems related to the audio-
visual field, but the following are of par-
ticular interest to the SMPTE: Section
3, Buildings and Equipment; Section 7,
Instructional Materials; Section 8, Pro-
duction of Audio-Visual Materials by
Colleges and Universities; Section 10,
Radio and Recordings in Education,

May 1952 Journal of the SMPTE Vol. 58

445

Section 11, Relationships Between Edu-
cation and the Audio-Visual Industry;
Section 12, Research; and Section 15,
Television in Education.

As literary outlets, there arc available
to the organization DAVI Proceedings
and the magazine Educational Screen.
One committee is working on a DAVI
Yearbook on the Administration of
Audio-Visual Programs of which the first
issue is due in the next year or two. As
soon as the tentative specifications on
Buildings and Equipment are finished by
Section 3, they will be published in tem-
porary form. Then in 1954, it is ex-
pected that the first Yearbook on Build-
ings and Equipment for Audio-Visual
Education will be ready, with the re-
sults of the research now being insti-
gated.

Operation of Conference

At this Winter Conference, there were
over 350 registrants from 25 states and
Puerto Rico. Their method of operating
was different from that of the SMPTE in
that prepared papers or lectures were re-
stricted to two evening sessions and the
final morning of the conference. There
was a get-together luncheon on the first
day, much like ours, but here a keynote
address was given, after which the main
group divided into Action Planning Sec-
tions which met in separate rooms for
discussions and drafting of recommenda-
tions. In this conference most of the 1 5
groups, which varied in size from 8 or 9
people up to 50 to 60, met three times.
There was also one business meeting in
which the entire conference participated.
One interesting variation took place on
the final morning when a reporter sum-
marized the work performed by the
various committees at this conference.
This was followed by a symposium on
"Television's Challenge to Education,"
in which a chairman and seven speakers
took part, after which the conference ad-
journed.

Buildings and Equipment Section

Section 3, Buildings and Equipment, is
of particular interest to the SMPTE
Committee on 16mm and 8mm Motion
Pictures because of the work done in 1 941
along the lines of projection in class-
rooms,1 and the efforts, after the war, in
which an uncompleted attempt was made
to revise and expand the original report.
Some of the research problems now con-
fronting DAVI are the same problems
that are listed on the SMPTE's agenda
for research. The optimum screen
luminances for various types of film and
various levels of ambient illumination are
included in this category.

A previous meeting of Section 3 had
been held on November 17, 1951, in
New York City.2 Certain provisional
drafts were available from that meeting,
and several members had prepared
recommendations on certain points for
consideration by the group meeting in
Boston.

General Architectural Planning

The notes that follow are not intended
to give the exact content of the speci-
fications, which are not yet in final
form, but merely to indicate the general
trend of the discussions. The first dis-
cussion dealt with the influence that the
architects have on the design of school
buildings and how they obtain their
final plans. Some make a real effort to
survey the needs of the community,
while others work from drawings outlin-
ing certain requested features. Some
school districts have definite ideas of what
they want, and it is essential that in
such cases the utmost cooperation be
obtained from the architects. They
stress, however, that they much prefer to
work from stated requirements rather
than merely to assemble architectural
features that have been prescribed by
someone else. In this way there is less
restriction from a creative point of view.

Ideally, architects should have some
training in education. One source of
trouble in obtaining audio-visual facili-

446

May 1952 Journal of the SMPTE Vol. 58

ties has been that when costs of proposed
buildings are high, features provided for
audio-visual instruction are often the
first to be scrapped.

In one city, 1 8 million dollars have been
allotted for the rehabilitation of old
buildings. There are many local com-
mittees working on their separate needs,
from the standpoint of the uses to which
the buildings are to be put. But fast ac-
tion must be taken by this Section 3 if its
recommendations are to be available in
time for this large remodeling project.
It was decided, therefore, that screens
and other purchased accessories would be
kept in the background for the present,
and that the committee would concen-
trate on architectural recommendations
for the buildings themselves.

Classrooms

Since there is general agreement that
the classroom is the first and most im-
portant place to equip for audio-visual
activities, certain details were laid down
for such rooms. The items covered in-
clude the number of electrical outlets,
their location, their current-carrying
capacity, wall switches for the room
lights, conduits for connections to the
loudspeaker, acoustical properties of the
room, for which standards outlined by
the Acoustical Materials Association3
were specified, ventilation, control of the
illumination in the room so that 0.1 ft-c
can be obtained during projection, stor-
age of equipment, the size and type of
screen, its location with the bottom edge
at the eye level of the pupils in each par-
ticular room, the method of mounting
the screen and dimensions of projection
stands. A screen width of one-fifth the
distance to the farthest row of seats was
preferred to the usual one-sixth. In
many ways this work parallels Adrian L.
TerLouw's recommendations in the
September 1945 Architectural Record.

Subjects for Research

The next considerations were of the
questions that require research: the size

and luminance of the screen with respect
to the illumination in the room; the ex-
tent of the equipment needed; methods
of raising the priority of audio-visual
equipment and obtaining community
support for it; methods of darkening the
room; functions of building coordina-
tors; and ways to persuade those who
publish illustrations of "beautiful" class-
rooms to include audio-visual equipment
and materials in their pictures, and to
show the method of darkening the
room.

A member from California showed a
motion picture demonstrating an inter-
locking type of Venetian blind that is en-
gaged in channels. This screen, which
has been used in some of their schools,
lowers the illumination to about 0.5 ft-c
under the outdoor lighting conditions
encountered in California.

In this connection, there was some dis-
cussion of the possible advantages of
windowless classrooms. Some members
thought that if windows were omitted
entirely, the benefits would be as fol-
lows: the pupils could neither see nor
hear outside disturbing influences, such
as street traffic; darkening of the room
would not be a problem; the regular
lighting for the classroom could be con-
trolled more precisely; it would be easier
to treat the room acoustically; and the
acoustics would be constant, whether or
not the room was darkened. Not much
was said about the arguments against
this idea.

In keeping with the precepts of those
working in the audio-visual field, the
committee made plans for the prepara-
tion and circulation of a film strip illus-
trating good practice in the design of
classrooms, and possibly some of the
things that should be avoided.

Auditoriums

Specifications for auditoriums were
discussed next. Again the acoustical
treatment was left to the specifications
prepared by the Acoustical Materials
Association. One member believed that

D. F. Lyman: Audio- Visual Instruction ^Conference

447

architectural conditions for auditoriums
are so varied that the committee could
not write effective specifications. For
example, some architects are now dis-
couraging the construction of projection
booths because of the complications and
expense involved. It was decided that
there should be no reference to a booth,
but that the switches, outlets, conduits
and other facilities should be specified for
a "projection station." Thus the pro-
jector could be located in a booth or in
some other spot where it would interfere
as little as possible with the functioning of
the auditorium. Here the screen size
was made one-fifth the distance to the
rear row of seats, but not greater than
9 ft X 12 ft. Plans were made for con-
necting the audio-visual equipment to the
large public-address system in the audi-
torium, whenever possible. It was
agreed that the auditorium should have
no windows, and some believed that the
illumination should be even less than the
0.1 ft-c specified for the classroom.

Preview Rooms

It was decided that a statement should
be included in the recommendations that
audio-visual activities belong in the
classroom or in the auditorium, and that
no attempt should be made to substitute
a central room reserved for audio-visual
instruction only. Some members, how-
ever, stressed the need for a room where
the teachers can preview the material
they are going to use in the classroom.
Then there must be space for storing
equipment and editing films. In some
schools there should be office space for
the director of audio-visual aids. In
many cases these facilities can be pro-
vided in one room, and this room should
be centrally located and near the library.

Special Reports

Submitted to the committee for guid-
ance at this meeting, a Preliminary Re-
port of a Study of the Problems of Light
Control in the Classrooms of Southern

Calilifornia, was of considerable interest.
This was prepared by the Research Com-
mittee of the Audio- Visual Education
Association of California, Southern Sec-
tion. It is not complete, but it does cover
567 schools and 8591 classroom units.
Only about one-third of the classroom
units have satisfactory control of the
light for projection. The same ratio ap-
plies to 886 classrooms under construc-
tion, but of 589 classrooms in the plan-
ning stage, only 12% are being equipped
for darkening. Of 191 classrooms being
remodeled, or in the planning stage for
remodeling, only 20% are being so
equipped.

This is a serious situation because it
looks like retrogression instead of prog-
ress. The report points out that large ex-
penditures have been made for audio-
visual materials and equipment, yet
"Audio-visual directors have been faced
with the dilemma of spending money and
energy developing audio-visual resources
only to have difficulty at the point of ap-
plication, the classroom, where, in too
many cases, no light control exists and
children cannot see the fine films and
other materials that are available to help
them learn."

, Some architects are advocating the use
of small, bright screens in fairly well-
lighted rooms. Illuminations of 10, 15,
50, or even more, foot-candles are men-
tioned. This information applies par-
ticularly to slides. Whether or not the
picture is still effective with the low con-
trast obtained in this way is a factor that
should be considered.

Need for Liason

The Society of Motion Picture and
Television Engineers can be of assistance
to the DAVI committee because of the
fund of information in its carefully pre-
pared Journal. Some of this is now being
furnished directly to Action Planning
Section 3 of DAVI. Certain SMPTE
committees may find it desirable to co-
operate directly with DAVI committees.

448

May

Journal of the SMPTE Vol. 58

The personnel of the DAVI organization
have much more direct knowledge of
what is transpiring in the educational
field than most of our members have, and
they know much better what the real
needs are. When the DAVI committee
starts to write specifications or recom-
mendations for equipment, it will need a
great deal of help because of the com-
plexity of the problems involved.

References

1. "Recommended procedure and equip-
ment specifications for educational
16-mm projection," Jour. SMPE, 37:
22-75, July 1941.

2. (Report of Section 3 Meeting) Educa-
tional Screen, 31: 18, Jan. 1952.

3. Theory and Use of Architectural Acoustic
Materials, Acoustical Materials Asso-
ciation, 59 E. 55 St., New York 22,
N.Y.

D. F. Lyman: Audio-Visual Instruction Conference

449

Television Studio Lighting

Committee Report by RICHARD BLOUNT

THE TELEVISION STUDIO LIGHTING COM-
MITTEE of the SMPTE met in New York
City on April 16, 1952, to discuss means
of measuring studio lighting. As origi-
nally planned, the meeting was attended
by numerous lighting directors from
the various networks who provided the
practical approach to this problem.
From the discussion, it became apparent
that incident light is measured because
simple meters are available, but that
brightness would be measured if a meter
of similar simplicity and size were avail-
able. Since such a device will probably
be somewhat more difficult to obtain, the
Committee decided to establish desirable
characteristics of both types of meters
with the thought that an incident meter
could be made without any great delay.
This would give the studio personnel a
standard measuring device which could
be used until a convenient brightness
meter can be produced.

The following listing of specifications
was tentatively agreed to at the meeting,
and these are to be given wide publicity
among meter manufacturers and users for
further suggestions. These proposed
specifications are being published here
to encourage comment from any inter-
ested reader of this report.

Specification for an Incident Light Meter
1. The spectral sensitivity shall con-

A report submitted on April 23, 1952,
by the Committee's Chairman, Richard
Blount, General Electric Co., Nela Park,
Cleveland 12, Ohio.

form closely to that of the 5820 image
orthicon camera tube.

2. The coverage angle shall closely ap-
proach a cosine distribution, i.e., the re-
sponse shall be maximum perpendicular
to the plane of the photocell and shall be
70% of the maximum at ±45 ° from the
perpendicular.

3. The meter shall have a single scale
and shall respond from 0 to 300 ft-c.
The scale shall be logarithmic with 0 to
30 ft-c covering approximately 25% of
full scale.

4. The size, and weight and shape
shall be such as to be conveniently car-
ried in a suit coat pocket. Rounded
edges are desirable since the meter will be
hand held in use. A maximum thickness
of one inch is desirable.

5. An adjustable-length neck cord is
required. The maximum length should
allow the meter to rest in a trousers
pocket.

6. The meter should withstand reason-
ably rough handling similar to that ex-
perienced by photoelectric exposure
meters. A storage case may be pro-
vided at the manufacturer's option.

Preliminary Specifications for a Portable
Brightness Meter

1. The instrument shall incorporate a
photoelectric device for light measure-
ments and shall be dependent upon the
human eye for aiming only.

2. The spectral response shall be
similar to that of the 5820 image orthicon
camera tube.

450

May 1952 Journal of the SMPTE Vol. 58

3. The angle of coverage shall be

i°±J°-

4. The instrument shall measure
brightness from 70 ft-L max down to 1
ft-L, with 0.1 ft-L as a very desirable
minimum. This range may be covered
by a number of steps — perhaps in mul-
tiples of 10. Switching between steps
should be accomplished internally. No
separate, external filters or other devices
shall be used.

5. A sight or viewfinder which pro-
vides an upright image shall be incor-
porated. It shall enable the operator to
quickly and positively identify the area
to be measured. Parallax shall be kept
to a minimum.

6. The meter shall be designed so that
the operator can make a single measure-
ment in less than five seconds, but will
allow him to read the actual value at his
leisure.

7. The meter shall not require ex-
ternal power and shall be built as a
single unit.

8. If battery-operated, it shall be able
to operate continuously for at least 10
hours on a single set of batteries.

9. The calibration shall be stable
throughout the battery life. A simple
external calibration device shall be made
available.

10. The meter should not exceed 400
cu in. and the weight shall be no greater
than 5 Ib.

The Committee appreciates that the
brightness meter specifications may be
difficult to meet, but in the apparent ab-
sence of any meter that meets the needs
of television stations, it was decided to
state the precise requirements and com-
promise later if necessary.

Richard Blount: TV Studio Lighting Report

451

Revision of Screen Brightness Standard

THE PROPOSAL of the Screen Brightness
Committee to revise Z22. 39-1 944, Screen
Brightness for 35Mm Motion Pictures, is
limited solely to the addition of a phrase
excluding outdoor theaters.

A careful survey did not indicate a need
for changing the brightness level recom-
mended for indoor theaters and the pro-
posed change simply recognizes the prob-
lem of outdoor theaters. These theaters,
with their large screens, were not originally
considered in arriving at the standard for
screen brightness, and a large portion of
these are unable to meet the standard with
presently available equipment; hence the
screen brightness standard has been
revised to pertain solely to indoor theaters.

Due to the fact that both indoor and

outdoor theaters use the same release
prints, the Screen Brightness Committee
voted to add a note urging the outdoor
theaters to use this standard as a goal.
The Standards Committee supported this
position and authorized preliminary pub-
lication for a 90-day period of trial and
criticism.

Please send comments to Henry Kogel,
Staff Engineer, at Society Headquarters,
before August 15, 1952. If no comments
are received during the three-month trial
period, this revised standard will then be
submitted to ASA Sectional Committee
PH22 without further vote within the
SMPTE and with the recommendation
that it be processed as an American
Standard.

Proposed American Standard

Screen Brightness
for 35mm Motion Pictures

PH22.39

Revision of
Z22.39-1944

Screen Brightness

The brightness at the center of a screen for viewing 35mm motion pictures
in indoor theaters shall be 10+' foot-Lambert when the projector is running
with no film in the gate.

Note: Outdoor theaters have been excluded from the above standard because of their
inability to meet it. It is recommended that outdoor theaters approach the indoor stand-
ards as closely as possible in view of the fact that the same release prints are generally
used for both types of theaters.

NOT APPROVED

452 * May 1952 Journal of the SMPTE Vol. 58

Engineering Activities

7 1st Convention The Engineering Com-
mittee meetings at the
71st Convention were in the main well
attended and demonstrated a high degree
of member participation and interest.
The highlights of the meetings are:

Film Dimensions The further decrease
in the shrinkage char-
acteristics of 16mm film creates a twofold
problem: (1) the width must be decreased
to preclude film stickage in the gate; and
(2) a decrease in the stated width dimen-
sion in the standard might erroneously
lead equipment manufacturers to decrease
the gate dimension and thereby start a
disastrous merry-go-round.

The proposed solution was that the
existing standard be revised by placing an
asterisk beside the width dimension and
indicating below that for low-shrink film
the standard is 0.628 ± 0.001. Low-
shrink film would then be defined in the
Appendix. A letter ballot on this pro-
posal will be sent shortly to the full com-
mittee.

Film-Projection Practice The commit-
tee first out-
lined plans for a major increase in activity
and then reviewed a previous recom-
mendation that American Standard
Z22.58, Projector Aperture Dimensions,
was unsuitable for international standardi-
zation. It is now felt that although
certain features of the standard do require
revision, the basic dimensions are valid
and valuable both here and abroad and the
previous recommendation was therefore
reversed.

Laboratory Practice The three active
projects were re-
viewed and a fourth one proposed:

1. The Proposed American Standard
Enlargement Ratio for 16mm to 35mm
Optical Printing, published in the January

Journal, is now well on its way toward
standardization.

2. Screen Brightness for Laboratory
Review Rooms, presently out to letter
ballot, is meeting with difficulty primarily
because of the conflicting needs of three
classes of 1 6mm print consumers : amateur,
television and Navy. This is now to be
considered by the Screen Brightness Com-
mittee, after which this committee will
attempt to draft a compromise proposal.

3. Printer Light Change Cueing of
16mm Negatives, a proposal to eliminate
negative notching, has just been sent to
the committee. Preliminary comments
at the meeting indicated that redrafting
may be required.

4. A new project to produce a glossary
of chemical terms used in motion picture
laboratories was initiated.

High-Speed Photography In addition

to the pri-
mary discussion on preparations and re-
sponsibilities for the symposium at the
Fall Convention in Washington, the
committee discussed the need for a glossary
and agreed upon the initial steps to achieve

Screen Brightness The various sub-
committees reported

on their progress to date and mention was
made of the formation of a new subcom-
mittee on illumination practices to make
recommendations on uniformity of illumi-
nation across the screen. The ICI recom-
mendations on screen brightness were
discussed and it was noted that the Ameri-
can Standard falls within the recommended
range :

ICI = 7.3-14.6ft-L
American Standard = 9-14ft-L

Plans were made to continue the screen
brightness survey of outdoor theaters as
soon as weather permits.

453

16mm and 8mm An extensive agenda
was considered and a
continued high level of activity projected.
As a result, the full committee will soon
l>c voting on revisions of six standards in
an effort to make them consistent as to
emulsion position, edge guiding and titles
(Z22.9, .10, .15, .16, .21, .22). A letter
ballot is also to be taken on a compromise
proposal on PH22.75, "A" and "B"
Windings of 16mm Raw Stock which may
resolve the existing deadlock on this
thorny issue. Reels for a 15-min show
(600-ft size) as well as reels over 2000 ft
were discussed and assignments given for
gathering additional required information.

in nature, and officially submitted them
to the Standards Committee for further
processing as American Standards.

Consideration was briefly given to other
magnetic sound track proposals and also
to magnetic test films. The nature of
the discussion revealed the need for a
meeting of the Magnetic Recording Sub-
committee which was scheduled for and
held the following day. The Subcom-
mittee reached no decisions but proceeded
efficiently to outline a major program of
work on test film specifications (azimuth
and multifrequency) and additional sound
track standards.

Sound Of the several items considered,
the magnetic sound track pro-
posals were, of course, paramount. These
were published in the July 1951 Journal
for trial and criticism and it was up to the
committee to pass on the comments
received. After a full and rounded
discussion, the committee approved the
proposals as published, with minor modi-
fications which in the main are editorial

ISO Delegation A very workmanlike
job was done to prepare
the group for its role at the forthcoming
meeting of ISO TC/36 in New York on
June 9 and 10. The delegation discussed
point by point the position it would take
on the various Agenda items. A copy of
the Agenda for the international meeting
and the U.S. position is available upon
request. — Henry Kogel, Staff Engineer.

University Film Producers Association

The University Film Producers Association
held its first meeting in the summer of
1947 at the University of Iowa, Iowa City,
Iowa. The organization came into being
as a result of the evidenced need for uni-
versity film makers to meet and discuss
their mutual problems. The initial session
was a one week's conference and workshop.
This pattern was so successful that the
group still follows it.

At the time of the first meeting, many
problems confronted the university film
producers, such as personnel, music rights,
exchange, distribution, technical practice
and laboratory service. Many universi-
ties had the same problems. Some had
solutions, some did not. It was evident
that a clearinghouse was needed for the
exchange of information and ideas.

The first meeting included representa-
tives from nine educational institutions

and two commercial organizations. Today
there are approximately 60 members from
educational institutions, as well as repre-
sentatives from commercial film com-
panies, laboratories, film distributors, and
equipment manufacturers.

The purpose of the University Film
Producers Association has been very clear
since the inception of the organization.
"It shall be the purpose of the UFPA to
further and develop the potentialities of
the photographic and recording arts in
improving instruction and communica-
tion." Individuals, organizations and
educational institutions qualified for mem-
bership are encouraged and invited to
join the University Film Producers
Association. There are at present four
classifications of membership: active,
associate, institutional and sustaining.

454

In its half-dozen years of existence the
UFPA has grown in accordance with its
original plans. It is today an incorpo-
rated, nonprofit organization whose mem-
bers include teachers, professors, film
makers, film technicians, film companies,
film laboratories, film distributors, equip-
ment manufacturers and dealers, and
university motion picture students.

Annually, during the month of August,
the Association meets in a city chosen by
the members. August of 1952 will find
the organization guests of Syracuse Uni-
versity, Syracuse, N.Y. The week-long
conference includes a variety of program
subjects, such as technical problems of
film production, research, film curriculums,
distribution, film screenings and films for
television. Equipment manufacturers and
distributors display and demonstrate new
equipment at the meetings. The lectures,
demonstrations, round tables, panel dis-

cussions and original papers offer the
membership introduction to new areas
and developments in the film field. A
highlight of the annual session is the pres-
entation of awards for outstanding films
produced by the university film makers.
Officers of the organization are:

John R. Winnie, University of Iowa,

President

Wilbur Blume, University of Southern

California, Vice-President

Roland J. Faust, Indiana University,

Secretary- Treasurer

Communication regarding membership
should be addressed to the Secretary.
Information concerning Journal subscrip-
tion or manuscript submission should be
sent to the Editor, Lu Snyder, Audio-
Visual Center, Syracuse University, Syra-
cuse, N.Y. — John R. Winnie.

Obituary

Louis Gerard Pacent, an early researcher
for sound motion pictures, died in April
at the age of 58.

Mr. Pacent began experimenting with
wireless transmission when he was a youth
and he had his own amateur station when
he was 16. In 1913 he was a communi-
cator in the Naval Militia and in 1917 he
served aboard the U.S.S. Gloucester. Also
during World War I he worked on the
development of military communications
equipment.

After that war he organized the Pacent
Electric Company, Inc., which was active
in the design and production of electric
and radio facilities for General Electric,
Westinghouse, Western Electric, RCA and
the Government. Before World War I
he was influential in encouraging and
instructing amateurs and in 1921 the first
short-wave transatlantic message was trans-
mitted to Scotland from Greenwich,
Conn., on 200 meters which had been
Mr. Pacent's suggestion.

Mr. Pacent was active in research for
sound motion pictures in the 1920's. He
is reported to have designed the first all
power-operated motion picture sound

equipment while a consultant for Warner
Brothers Pictures, the equipment having
been installed in 1928.

At the time of his death he was president
of the Pacent Engineering Corp., a firm
which he had founded 20 years ago. The
corporation developed portable sound
reproducers, inter-office communicating
equipment, high-fidelity radio and public
address systems.

In 1946 Mr. Pacent was given a Certifi-
cate of Appreciation by the War Depart-
ment in recognition of his firm's valuable
assistance to the Signal Corps during
World War II.

He was a native of New York and was
graduated in 1916 from Pratt Institute of
Technology with the degree of Industrial
Electrical Engineer. He was the author
of many papers and books on communica-
tions engineering. He was a Fellow of this
Society and also of the Institute of Radio
Engineers, and a member of the American
Institute of Electrical Engineers. In 1951,
he received the Marconi Memorial Medal
of Achievement from the Veteran Wireless
Operators Association.

455

New Members

The following members have been added to the Society's rolls since those last published.
The designations of grades are the same as those used in the 1950 MEMBERSHIP DIRECTORY.

Honorary (H) Fellow (F) Active (M) Associate (A) Student (S)

Aicher, John W., SRT-TV Studios.
Mail: 400 W. 56 St., New York 19,
N.Y. (S)

Bailey, G. C., Export Sales, Eastman
Kodak Co., Rochester, N.Y. (A)

Bailey, James H., Projectionist, Warner
Brothers Studio. Mail: 813 N. Rose
St., Burbank, Calif. (A)

Behrmann, Louis L., Photographic Tech-
nologist, U.S. Naval Research Labora-
tory. Mail: 12240 Viers Mill Rd.,
Silver Spring, Md. (A)

Belinkoff, Irving R., Junior Engineer,
Federal Manufacturing & Engineering
Corp. Mail: 449 Beach 67 St., Arverne,
L.I., N.Y. (A)

Bernier, Maj. Robert V., U.S. Air Forces.
Mail: 4505 Arcadia Blvd., Dayton 10,
Ohio. (M)

Bessor, John O., Jr., Motion Picture
Cameraman, Byron, Inc. Mail: 1537
Roosevelt Ave., Falls Church, Va. (A)

Bordelay, Jack F., Director, Cameraman,
Syracuse University. Mail: 108 Corn-
stock Ave., Syracuse 10, N.Y. (A)

Braunstein, Simeon, Director, Photo Sec-
tion, Engineering Research Div., New
York University. Mail: 2666 Valen-
tine Ave., New York 58, N.Y. (M)

Buckner, W. C., Vice-President and Chief
Engineer, NECO, Inc. Mail: 12132
Herbert St., Culver City, Calif. (M)

Carrington, H. K., Motion Picture Pro-
ducer, Nationwide Pictures, Melba
Theatre Bldg., Dallas, Tex. (A)

Cass, Lewis S., Recording Technician,
Paramount Pictures. Mail: 240 W.
98 St., New York, N.Y. (A)

Cooper, James E., Motion Picture Film
Editor and Projectionist, The Calvin
Co. Mail: 3208 Highland Ave., Kansas
City, Mo. (A)

Creamer, C. C., Partner, Theatre Equip-
ment Business. Mail: 75 Glenwood
Ave., Minneapolis, Minn. (A)

I)c Hi us, Bert A., Motion Picture and
Sound Salesman, Fotoshop, Inc. Mail:
112-41 —72 Rd., Forest Hills 75, L.I.,
N.Y. (A)

Diament, Clifton L., Motion Picture
Laboratory Technician, Byron, Inc.
Mail: 1017 Dashill Rd., Falls Church,
Va. (A)

Dickerson, Malon, Chemical Physicist,
Southwest Research Institute. Mail:
8500 Culebra Rd., San Antonio 6, Tex.
(A)

Dratch, Nicholas, Quality Control Engi-
neer, Bolsey Corp. of America. Mail:
1569 Metropolitan Ave., Bronx 62, N.Y.
(A)

Faber, John, Technical Representative,
Eastman Kodak Co. Mail: 5 Edge-
water Dr., Denville, NJ. (M)

Faust, A. Donovan, Assistant General
Manager, WDTV (Allen B. DuMont
Laboratories, Inc.). Mail: 121 Abbey-
ville Rd., Pittsburgh 28, Pa. (A)

Fielding, Raymond, University of Cali-
fornia at Los Angeles. Mail: 1333^
Federal Ave., Los Angeles 25, Calif.
(S)

Freedman, Myron L., General Manager,
Crescent Film Laboratories, Inc. Mail:
7510 N. Ashland Ave., Chicago 26, 111.
(M)

Garcia, Orlando M., Custom House
Broker. Mail: Habana #412 (Altos),
Entre Obispo y Obrapia, La Habana,
Cuba. (A)

Geis, Donald C., Sound Engineer, Hearst
Metrotone News, Inc. Mail: 5939
Forest Glen Ave., Chicago 30, 111. (A)

Geller, Gilbert, Cameraman, Films for
Industry. Mail: 160-37 Highland
Ave., Jamaica, L.I., N.Y. (A)

Gimborn, Charles J., Jr., Chief Motion
Picture Photographer, WCAU-TV.
Mail: 412 W. Delphine St., Phila-
delphia 20, Pa. (A)

Giovanelli, Frank, Cameraman (Anima-
tion), Consolidated Film Industries,
Inc. Mail: 2716 Marion Ave., Bronx
58, N.Y. (M)

Giovanelli, Frank Jr., Consolidated Film
Industries, Inc. Mail: 2716 Marion
Ave., Bronx 58, N.Y. (A)

Glandbard, Max, Producer, Filmwright
Productions, Inc. Mail: 195 Mohawk
Dr., River Edge, NJ. (A)

Goldsmith, Charles B., Sound Technician,
Radio Corp. of America. Mail: 307
S. Renter Ave., West Los Angeles,
Calif. (A)

Goldstein, Raphael L., Projectionist, Tele-
vision Film Editor, South City Drive
In, WFIL-TV. Mail: 1812 S. Fifth
St., Philadelphia, Pa. (A)

456

Griffiths, Peter H., Electronic Engineer.
Mail: 28 Eccleston Rd., South Shore,
Blackpool, England. (A)

Grove-Palmer, Clifford O. J., Research
Engineer, British Admiralty, Royal
Naval Scientific Service. Mail: 59
Kings Rd., Rosyth, Dunfermline, Fife,
Scotland. (M)

Harvard, Emile A., Motion Picture Pro-
ducer, Cameraman, Harvard Produc-
tions. Mail: Box 1597, Tel Aviv,
Israel. (A)

Henion, William C., Project Engineer,
USAF Projection Equipment, Box 8015,
Area B, Photo Reconnaissance Labora-
tory, WCE, WADG, WPAFB, Dayton,
Ohio. (A)

Henning, Clarence G., Photographic
Technician, David White Co., 315 W.
Court St., Milwaukee 12, Wis. (M)

Hollo way, F. P., Carbon Development
Engineer, National Carbon Co., Div. of
U.C. & C. Corp., Fostoria, Ohio.
(M)

Hooper, Joseph K., Laboratory Manager,
Byron, Inc. Mail: 1608 Tyler Ave.,
Falls Church, Va. (A)

House, Robert A., Film Recording Group,
Radio Corp. of America, RCA Victor
Div. Mail: 3309 Gwinett Walk, Cres-
cent Park Apts., Camden, NJ. (A)

Johnson, Virgil L., Order Dept., Motion
Picture Laboratory, Byron, Inc. Mail:
2816 Buena Vista Ter., S.E., Wash-
ington 20, D.C. (A)

Kalian, Peter, Chief Photographer, Atomic

Energy Project, University of California,

• ' K W«

(A)

Box 4164, West Los Angeles 24, Calif.

Keating, Clifford M., Technologist (Photo-
graphic Chemistry), U.S. Naval Photo-
graphic Center. Mail: 147 Ivanhoe
St., S.W., Washington 20, D.C. (A)

Kern, Albert W., Assistant Cameraman,
Free-lance. Mail: 49 County Center
Rd., White Plains, N.Y. (A)

Kimura, T., Chief, Liaison Dept., Daiei
Motion Picture Co. Mail: #2, 3-
Chome, Kyobashi, Chuo-ku, Tokyo,
Japan. (M)

Klages, William M., Television Engineer,
National Broadcasting Co. Mail: 451
W. Fulton St., Long Beach, N.Y. (A)

Koster, William D., Laboratory Techni-
cian, Byron, Inc. Mail: 10422 Hay-
wood Dr., Silver Spring, Md. (A)

Layne, Joseph L., Mechanical Engineer,
Signal Corps Engineering Laboratories.
Mail: 116 Fifth Ave., Neptune, NJ.
(A)

Leighton, Thomas C., Optical Engineer,
John H. Ransom Physics Laboratories.

Mail: 31081 W. 67 St., Los Angeles 43,
Calif. (A)

Lewis, Keith B., Manager, Washington
Office, Eastman Kodak Co., 444 Shore-
ham Bldg., Washington 5, D.C. (M)
Lewis, Robin R., Motion Picture Labora-
tory Supervisor, U.S. Army Signal
Corps. Mail: 260-44 Langston Ave.,
Glen Oaks, L.I., N.Y. (A)
Mabrey, Layton, Motion Picture Director,
Extension Div., University of Oklahoma.
Mail: Box 615, N.C., University of
Oklahoma, Norman, Okla. (A)
Mamas, Harry, Cameraman, Telerraft
Film Productions. Mail: 17 Adams
St., Medfield, Mass. (A)
Marchiel, Stanley, Sensitometric Sound
Control, Paramount Pictures, Inc.
Mail: 81 Oakland St., Brooklyn 22,
N.Y. (M)

Marks, Jesse, Contract Negotiator and
Administrator, U.S. Navy Motion Pic-
ture Exchange, Bldg. 311, N.Y. Naval
Shipyard, Brooklyn 1, N. Y. (M)
Martens-Hughes, Margot, Visual Aids
Specialist, IIA, U.S. Dept. of State,
1778 Pennsylvania Ave., Washington,
D.C. (A)

Merriken, George T., Production Man-
ager, Byron, Inc. Mail: 112 Lynmoor
Dr., Silver Spring, Md. (A)
Miller, Robert W., Composer and Musical
Director. Mail: 4097 Valley Meadow
Rd., Encino, Calif. (A)
Moe, Sigurd M., Sound Recording Tech-
nician, U.S. Navy. Mail: Combat
Camera Groups Pacific, Com Nav FE,
FPO San Francisco, Calif. (A)
Mu ji dell, J., Newsreel Cameraman,
WBAP-TV. Mail: 2013 Ruea, Grand
Prairie, Tex. (M)

Nordman, 2d Lt. Robert G., Photo Eng.
Section, U.S. Air Force, Edwards Air
Force Base, Edwards, Calif. (A)
Oiler, Arthur H., Jr., Motion Picture
Laboratory Technician (Timer), George
W. Colburn Laboratory. Mail: 4416
Florence Ave., Downer's Grove, 111.
(A)

Pabst, Glenn C., Photographer, Bell
Aircraft Corp. Mail: 17 Raymond PL,
Hamburg, N.Y. (A)

Parker, B. E., Engineer, Gates Radio Co.
Mail: 720 Kentucky St., Quincy, 111.
(M)

Peltz, Leo G., Radar Technician, North
American Aviation, Inc. Mail: 500
Seventh St., Manhattan Beach, Calif.
(A)

Pierry, Michael J., Jr., Quality Control,
Precision Film Laboratory. Mail: 285
James St., Teaneck, NJ. (A)

457

I M. i k u n. Bernard D., Technical Writer,

General Precision Laboratory, Inc.

Mail: 15 May wood Ave., Port Chester,

N.Y. (A)
Portalupi, Piero, Director of Photography,

Lux Film. Mail: Viale Bruno Buozzi,

83 int F, Rome, Italy. (M)
Pratt, Harry E., Sales Representative,

W. J. German, Inc., 6700 Santa Monica

Blvd., Hollywood 38, Calif. (M)
Rayhack, Michael, Free-lance Motion

Picture Cameraman, Local 644, IATSE.

Mail: 10 Overlook Ave., Box 284,

Great Notch, NJ. (A)
Rosso, Lewis T., Production Manager,

Republic Productions, Inc. Mail: 4703

Mary Ellen Ave.. Sherman Oaks, Calif.

(M)
Ruellan, Gilbert, Director, Establisse-

ments Andre Debrie. Mail: 16 Rue

Piccini, Paris 16°, France. (M)
Russey, Elwood M., Vice- President,

Byron, Inc. Mail: Box 45, R.F.D. #1,

Vienna, Va. (M)
Scherer, Marc, SRT-TV Studios. Mail:

121 Parkside Ave., Brooklyn 26, N.Y.

(S)
Schnebly, John C., Motion Picture

Laboratory Technician, Byron, Inc.

Mail: 219 Winchester Way, Falls

Church, Va. (A)
Silverman, Bert, Film Editor, CBS-TV.

Mail: 999 Aldus St., New York 59,

N.Y. (A)
Skinner, John M., Laboratory Technician

(Printing), Byron, Inc. Mail: 127

Rolph Dr., Forest Heights, Md. (A)
Small, Elliot H., Photographer, Shell Oil

Co. Mail: 55 Auburn St., West Med-

ford 55, Mass. (M)
Smith, Newland F., Television Engineer,

WOR-TV, General Teleradio, Inc.

Mail: Spruce St., Riverside, Conn.

(M)
Smith, Robert E., Television Engineer,

National Broadcasting Co. Mail: 1001

Riverside Dr., Burbank, Calif. (A)
Smith, Thomas W. B., Regional Geologist,

Gulf Oil Corp. Mail: Box 1166,

Pittsburgh 30, Pa. (A)
Spafford, Ronald W., Assistant to General

Sales Manager, Industrial Products,

National Carbon, Ltd. Mail: 345

Kingsdale Blvd., Willowdale, Ont.,

Canada. (A)
Speed, William C., President, Audio

Devices, Inc., 444 Madison Ave.,

New York 22, N.Y. (M)
Stchncy, Michael C., Photographer, Di-
rector, Sarra, Inc. Mail: 18411 Dun-
dee, Homewood, 111. (M)
Stone, Tames, Engineer, Consolidated

Film Industries, Inc. Mail: 426 Ivy

La., Englewood, NJ. (M)

Taris, Charles M., Member, Technical
Staff, Bell Telephone Laboratories, Inc.
Mail: 11 Adams Ave., Cranford, N.T.
(A)

Taylor, Barney L., Chief Dental Tech-
nician, U.S. Navy. Mail: 534 Eleventh
St., NHA2, Honolulu 18, Hawaii. (A)

Teasley, Ernest, Secretary-Treasurer.
EDL Co. Mail: Miller Station, Garv
5, Ind. (A)

Thorn, Thomas C., Laboratory Manager,
Pathescope, Ltd. Mail: 29 Florida
Rd., Thornton Heath, Croydon, Surrey,
London, England. (M)

Todaro, Fred G., Design and Engineering,
Negative and Positive Processing Equip-
ment, Color Service Co. Mail: 320
Albemarle Rd., Brooklyn, N.Y. (M)

Tydings, Kenneth S., Photographic
Author, Podiatrist. Mail: 110 E.
Chester St., Long Beach, N.Y. (A)

Vance, Robert G., Cameraman, Byron,
Inc. Mail: Route 4, Box 325, Alex-
andria, Va. (A)

Vanoni, Vito A., Professor, California
Institute of Technology, Hydro Labora-
tory, Pasadena 4, Calif. (A)

Vary, Willard E., Head, Physical Tests
Div., U.S. Naval Photographic Center.
Mail: 2107 Ft. Davis St., S.E., Wash-
ington 20, D.C. (M)

Webb, Harry, University of Minnesota.
Mail: 813 University Ave., S.E.,
Minneapolis, Minn. (S)

Weiss, J. Paul, Research Physicist, Photo
Products Dept., E. I. du Pont de
Nemours & Co., Inc. Mail: 908 New
England Dr., Westfield, NJ. (M)

Young, H. A., Electrical Design Engineer,
RCA Victor Div. Mail: 702 N.
Naomi, Burbank, Calif. (A)

Zambuto, Mauro, Technical Director,
Motion Picture Studio, Sealer a Films.
Mail: Via Cavour 114, Rome, Italy.
(M)

Zan, Aung Phaw, University of Southern
California. Mail: 1072£ W. 31 St.,
Los Angeles 7, Calif. (S)

CHANGES IN GRADE
Cruse, Andrew W., (A) to (M)
DuVall, John W., (A) to (M)
King, Roy D., (A) to (M)
Paul, Morrison B., (A) to (M)
Petrasek, A. G., (A) to (M)
Potter, Johnson, (A) to (M)
Valentine, Fred, (S) to (A)
Wadlow, Huston E., (A) to (M)
Wendt, Paul R., (A) to (M)
Willey, Lyle E., (A) to (M)

458

Book Reviews

IES Lighting Handbook
(Second Edition)

Published (1952) by The Illuminating
Engineering Society, 1860 Broadway, New
York 23. i-xiii + 740 pp. + 37 pp.
appendix + 24 pp. index + 172 pp. advt.
482 illus. + numerous tables. 6 X 9 in.
Price $8.00.

This volume fills a long-standing need
for a compendium which presents essential
lighting theory and data in condensed and
readable form. The new edition repre-
sents an extensive revision of the original
1947 publication, more than 75% of the
material, according to the editors, being
new or completely rewritten. Some 200
pp. of text have been added, and data
have been revised in line with the best
current values. Bibliographies at the
ends of the sections have been extended
to include material published as recently
as September 1951.

The first part of the Handbook includes
sections dealing with physics of light,
light and vision, standards and nomen-
clature, measurement of light, color, light
control, daylighting, light sources and
lighting calculations. The second part is
devoted to applications, with discussions
of interior and exterior lighting, sports
lighting, street and highway illumination,
aviation and transportation lighting, minia-
ture lamps, and photographic, reproduc-
tion, projection, television and radar
screen lighting. A section on miscel-
laneous applications covers uses of ultra-
violet and infrared energy. Illumination
requirements of the various lighting fields
and methods for fulfilling them are well
covered.

The Handbook is, of course, designed
primarily to serve the needs of the illumi-
nating engineer. SMPTE members will
find it a handy compilation of "time-tested"
methods and techniques, although, as is
natural in a volume covering the entire
lighting field, material of direct interest

to the SMPTE member receives rather
brief treatment. The section on light
sources includes two pages on the carbon
arc, as well as tables giving performance
characteristics of d-c and flame-type car-
bons. The discussion on lighting for
motion picture photography includes de-
scriptions and illustrations of incandescent
and carbon-arc lamps commonly used for
set lighting, tables showing beam charac-
teristics, and figures giving spectral energy
distributions. Some eight pages are de-
voted to picture projection lighting with
paragraphs on brightness levels, screen
surfaces, viewing conditions, projection
booth design, and light output of typical
carbon-arc systems. There are short dis-
cussions of television studio lighting and
lighting of drive-in and motion picture
theaters.

The concise summaries contained in the
Handbook are supplemented through
bibliographies appended to each section.
An appendix contains conversion factors
and equations, I.G.I, tristimulus computa-
tion data, and tables of selected ordinates.
A detailed subject index, index tabs, and
a complete table of contents for each sec-
tion facilitate access to specific information.
The paper used is of high quality, the
print is larger and more legible than that
usually found in handbooks, and illustra-
tions and photographs have been provided
with unusual generosity.

As an authoritative and convenient
summary well provided with guides for
further reading, the IES Lighting Handbook
should find a place on the desk of every
engineer required to deal with broad
problems of lighting. The sections on
light sources, measurements and calcula-
tions will be especially useful to the
SMPTE member, though for detailed
information on problems of motion picture
studio, theater and projection lighting,
it will still be necessary to consult the
publications of the SMPTE.— M. S.
Wright, National Carbon Research Labora-
tories, Cleveland, Ohio.

459

Basic Electron Tubes

By Donovan V. Geppert. Published
(1951) by McGraw-Hill, 330 W. 42 St.,
New York 36. 332 pp. 257 illus. 6 X
9 in. Price $5.00.

As the title implies, this book deals with
the principles of operation of well-known
vacuum tubes.

Written as a text for undergraduates, it
fulfills the requirement for lucidity by an
understandable style and a novel arrange-
ment of subject matter. Undoubtedly
founded on the axiom that an interested
student learns rapidly and well, the author
has reversed the conventional order of
presenting his subjects.

Instead of starting with a few abstract
statements and a welter of mathematics,
he presents first the physical nature of the
easily recognized practical tube, giving
qualitative theory in explanation. From
this he goes to the electrical nature of the
tube, presenting the characteristic curves
and an illustrative circuit. The theory
is often explained with the assistance of the
rubber-membrane rolling-marble model,
which is shown in a well-executed three-
dimensional illustration.

By this time even the interested tyro
has a good conception of the device and
then the author launches into the mathe-
matical analysis that must be a part of any
mature treatment. This is given meaning
by sample calculations of problems likely
to be faced in practice. Both cgs and
mks units are used and the relation be-
tween the two systems explained.

The tubes treated are indicated by the
chapter headings: 1. High-vacuum and
Gas Phototubes; 2. High-vacuum Ther-
mionic Diodes; 3. High-vacuum Triodes;
4. Tetrodes and Pentodes; 5. Beam-
power Tetrodes; 6. Cathode-ray Tubes;
7. Glow-discharge Tubes; 8. Thermionic
Gas Diodes; 9. Thyratrons; 10. Mercury-
pool Arc Rectifiers; and 11. Ignitrons.

A useful feature of the book is a summary
of the practical consequences of altering
tube parameters. For instance, in the
case of the triode, nine numbered sentences
give the alteration of electrical characteris-
tics for stated alterations of the tube
structure. This tends to establish funda-
mentals in the mind of the student or

engineer and serves as a focal point for
reference when memory fades.

Additional references and problems are
to be found at the end of each chapter.

The book is the fourth in a series of
fourteen on electrical and electronic
engineering, for which the well-known
and esteemed Dr. Frederick Emmons
Terman is Consulting Editor. This fact
accounts for the title and the absence of
discussion of the microwave tube. This
device is more often a part of the circuit
than a separate entity and so is treated in
another book of the series.

The book would appear to be useful
to nearly all motion picture and television
engineers as a convenient reference. —
Harry R. Lubcke, Consulting Engineer,
2443 Creston Way, Hollywood 28, Calif.

Bases Techniques
de la Television

By H. Delaby. In French. Published
(1951) by Editions Eyrolles, 61, Blvd.
Saint-Germain, Paris (Ve). 340 pp. 273
illus. 6| X 9 f in. Paper bound. Price
2,200 fr. (approx. $6.30).

The principal interest of a reader of
this Journal in a technical book in a
foreign language lies in whether, having
surmounted the difficulties of translation,
the reader will obtain information on
other methods and devices that cannot be
obtained in his native language. This
reviewer's interest was largely in the
practical details of the controversial 819-
line French television system and its
operation in the same studio plant with a
625- or 455-line system. Unfortunately,
this book, although excellent in its way, is
not informative about such details.

With its companion volume by the same
author Principes Fondamentaux de Television,
to which frequent reference is made, this
book serves as an adequate text for a
course in television at what would be in
the United States, college senior level.
The general principles of video amplifiers,
synchronizing generators, studio cameras,
film cameras, transmitters, receivers and
antennas are covered in varying degrees
of detail. Studio control equipment rates
8 pp. ; the receiver, 24 pp. ; and the trans-
mitting antenna, 54 pp.; to cite a few

460

examples of the somewhat strange balance
between topics. This is perhaps an
unfair criticism, since much material of
importance, such as the whole subject of
scanning generators, is apparently covered
in the companion volume.

There are several general texts on tel^e-
vision published in the United States
which cover in more detail essentially every-
thing included in this book. Actually, over
half the references are to technical maga-
zines and books in English.

The one section which contains in-
formation not available in American (as
distinguished from English-language) texts
is that on the television transmission of
film. The use of flying-spot scanners
with continuously-moving film, and the
peculiar problems of 50-cycle power
supplies are discussed in reasonable detail.
From the references given, however, this
reviewer had the impression that, if he
had obtained a copy of the proceedings of
the "Congres de Television" which was
held at Paris in 1948, he would be in a
better position to learn about French
television methods than by a study of
H. Delaby's book.— S. W. Athey, General
Precision Laboratory, Inc., Pleasantville,
N.Y.

Television Principles

By Robert B. Dome. Published (1951) by
McGraw-Hill, 330 W. 42 St., New York
36. i-xii + 281 pp. + 9 pp. index. 170
illus. 6 X 9 in. Price $5.50.

The material for this book was taken
from a series of lectures which formed one
of the radio training courses for engineers
of the General Electric Co.

The book covers all stages of television
transmission and reception. There are
chapters on scanning and reproduction,
transmitting apparatus, antennas for trans-
mission and reception, propagation and
relays, RF input circuits and noise factors,
IF amplifiers, picture second detector and
the scanning system. The author has
followed the television signal from the
camera through the receiver.

Mathematical development of many
principles is shown and practical problems
using these principles are given. The

design problems make use of many of the
latest types of tubes.

Schematic circuits and diagrams are
used and there are no pictures or diagrams
of commercial installations or equipment.
The book is about engineering rather than
operations and is concise and to the point.

On the transmitting side Mr. Dome has
lightly covered the operation of the
different pickup tubes and antennas. He
has paid particular attention to video
frequency amplifiers and picture trans-
mitters.

On the receiving end he has emphasized
radio frequency input circuits, intermediate
frequency amplifiers and scanning circuits.
The chapter on RF input circuits includes
the cascode amplifier and has material on
noise factors for each circuit.

A miscellany chapter covers such things
as d-c restoration, automatic gain control,
overall fidelity and the author's own inter-
carrier sound system.

The book is an excellent work and is a
welcome addition to the McGraw-Hill
Television Series. — Otis S. Freeman, Asst.
Chief Engineer, WPIX, 220 E. 42 St.,
New York 17.

Application of the Electronic Valve in
Radio Receivers and Amplifiers (Vol. II)

By B. G. Dammers, J. Haantjes, J. Otte,
and H. Van Suchtelen. Published (1951)
by N. V. Philips' Gloeilampenfabrieken,
Eindhoven, Netherlands. Distributed in
U.S.A. by Elsevier Press, Inc., 402 Lovett
Blvd., Houston 6, Texas, i-xviii + 425
pp. + 6 pp. index. 343 illus. 6 X 9 in.
Price $7.75, English ed.

This is the second volume of a trilogy
being put out by the famous Philips of
Eindhoven on the uses of tubes in receivers
and amplifiers, principally the former.
Volume I covered rf and if amplification,
frequency changing, interference and dis-
tortion, and detection. The present vol-
ume is devoted to af voltage and power
amplification and power supplies. That
being the case, it is of considerable value
to motion picture workers, even though it
is primarily concerned with receivers.

The treatment is by no means super-
ficial and is entirely applicable to amplifiers
and power supplies for any purpose.
Tube performance is analyzed mathe-

461

inatically and graphically and design
criteria given for standard circuits. An
interesting section, not strictly within the
purview of tubes is on design of af trans-
formers.— Richard H. Dorf, Audio and
TV Consultant, 255 W. 84 St., New York
24, N.Y.

Agfa color Process, a Short Bibliography

Compiled by Alexis N. Vorontozoff (1951),
25 mimeographed pages, 8£ X 11 in.
Available from the Author, 10 rue Made-
moiselle, Paris. Price, $0.50 plus postage.

Mr. Vorontozoff has done a noteworthy
job in compiling 236 references on the
Agfacolor Process which he has published
alphabetically according to author, with
a cross-reference list according to subject.
He has indicated also whether or not the
reference has been consulted directly, the
language of the original paper, references
to abstracts of each paper published in
other periodicals, availability of reprints,
translations, etc. The bibliography covers
all aspects of the Agfacolor Process,
including numerous references applicable
to the motion picture field. — Lloyd E.
Varden, Pavelle Color, Inc., 533 W. 57 St.,
New York 19, N.Y.

Transmitting Valves

By J. P. Heyboer and P. Zijlstra. Pub-
lished (1951) by N. V. Philips' Gloeilam-
penfabrieken, Eindhoven, Netherlands.
Distributed in U.S.A. by Elsevier Press
Inc., 402 Lovett Blvd., Houston 6, Texas,
i-xii + 281 pp. + 2 pp. index. 256 illus.
6 X 9 in. Price $6.25, English ed.

This volume is Book VII of the fast-
growing Philips library. It is concerned
with the characteristics of transmitting
tubes — pentodes, tetrodes, and triodes in
which transit-time effects are negligible —
and the circuits in which they are used.
Chapters give thorough mathematical
design treatments of tube construction,
rf power amplifiers, oscillators and fre-
quency multipliers, as well as some data
on special uses such as vhf feedback
circuits. One of the appendixes contains
a table of technical data on Philips trans-
mitting tubes. As with the receiver book
from Philips (reviewed above), the trans-
lation is excellent, the language clear
and concise. — Richard H. Dorf, Audio and
TV Consultant, 255 W. 84 St., New York
24, N.Y.

Positions Wanted

Photographic Chemist: 3 yr. experience black-and-white and color film laboratory
practice and quality control. Familiar with all commercial color processes and sensi-
tometry. Have conducted research in new processing methods. Position desired in
research or development on new products and processes. Will relocate. Write M-52,
c/o Lichtig, 3758 Tenth Ave., New York 34, N.Y.

Production, TV or Motion Picture: NYU BA in motion picture and TV production;
participated in productions as director and unit mgr; experience as motion picture
sensitometrist ; at present motion picture negative assembler and cutter; worked swing
shift while attending college; licensed 35mm projectionist; single, 29, veteran, resume
on request; go anywhere. Harold Bernard, 560 Eastern Pkwy, Brooklyn 25, N.Y.

Sound mixer and transmission engineer: 5 yr experience 35mm magnetic and optical
16mm optical and disc recording systems. As mixer has experience stage recording and
re-recording; in transmission has installed a recording channel complete from design to
operation, also maintenance. Will accept position any geographic location. Write
L-30, c/o Fifer, 143 Church St., Phoenixville, Pa.

Motion pictures in color depend on the engineers' knowledge of the "Principles of Color
Scnsitomctry." A 72-page article bearing that title and prepared by the Color Sensitom-
ctry Committee appeared in the Journal for June 1950. Attractive reprint copies may
be purchased for $1.00.

462

Meetings

72nd Semiannual Convention of the SMPTE, Oct. 6-10, Hotel Statler,

Washington, D. C.

Other Societies

Society of Photographic Engineers, Photographic Instrumentation Symposium, June 4-5,

Naval Ordnance Laboratory, White Oak, Md.

American Institute of Electrical Engineers, Summer General Meeting, June 23-27,

Hotel Nicollet, Minneapolis, Minn.

American Physical Society, June 30-July 3, Denver, Colo.

National Audio-Visual Association, Convention and Trade Show, Aug. 2-5, Hotel

Sherman, Chicago, 111.

University Film Producers Association, Annual Meeting, Aug. 11-15, Syracuse Univer-
sity, Syracuse, N. Y.

Photographic Society of America, Annual Convention, Aug. 12-16, Hotel New Yorker,

New York

American Institute of Electrical Engineers, Pacific General Meeting, Aug. 19-22, Hotel

Westward Ho, Phoenix, Ariz.

International Society of Photogrammetry, Conference, Sept. 4-13, Hotel Shoreham,

Washington, D.C.

American Standards Association, Third National Standardization Conference, Sept.

8-10, Museum of Science and Industry, Chicago, 111.

Illuminating Engineering Society, National Technical Conference, Sept. 8-12, Edge-
water Beach Hotel, Chicago, 111.

Biological Photographic Association, Annual Meeting, Sept. 10-12, Hotel New Yorker,

New York

National Electronics Conference, Annual Meeting, Sept. 29-Oct. 1, Sherman Hotel,

Chicago, 111.

Optical Society of America, Oct. 9-11, Hotel Statler, Boston, Mass.

American Institute of Electrical Engineers, Fall General Meeting, Oct. 13-17, New

Orleans, La.
American Standards Association, Annual Meeting, Nov. 19, Waldorf-Astoria, New York

Six American Standards have been added to the Motion Picture Set of 60 which the
Society has had available for sale. To holders of the present set the Society has made
available the seven new standards: PH22.11-1952, PH22.24-1952, PH22.73-1951,
PH22.74-1951, PH22.76-1951, PH22.77-1952 and PH22.82-1951. The price is $1 plus
3% sales tax on deliveries in New York City.

The new set of 67 standards in a heavy three-post binder with an index is available at
$14.50 plus 3% sales tax on deliveries in New York City; foreign postage is $.50 extra.

All standards in sets only are available from Society Headquarters. Single copies of
any particular standard must be ordered from the American Standards Association,
70 E. 45th St., New York 17, N.Y.

Back issues of the Journal available: A set of Journals from January 1945 through 1951
is available at $15.00 plus packing and carrying costs from Richard W. Maedler, 32-52 —
46 St., Long Island City 3, N.Y.

Back issues of the Journal available: Don Canady, 5125 Myerdalc Drive, R.R. 15,

Cincinnati 36, Ohio, desires to dispose of a complete set, in excellent condition, from
January 1930 to date, plus one issue of September 1928. Anyone interested in acquiring
the complete set should communicate directly with Mr. Canady.

463

New Products

Further information about these items can be obtained direct from the addresses given.
As in the case of technical papers, the Society is not responsible for manufacturers' state-
ments, and publication of these items does not constitute endorsement of the products.

The Tener has been announced as a 25th Anni-
versary Topper by Mole-Richardson Co., 937 N.
Sycamore Ave., Hollywood 38, Calif. This 10,000-
w lamp, Type 416, is described with these speci-
fications: constructed of sheet metal with inter-
locking channels for ventilation; condenser, special
20-in. Fresnel, 15° to 40° divergence; mirror,
Alzak aluminum and completely adjustable; socket,
Mogul Bipost; globe, 10,000-w G96 Mogul Bipost;
focusing by handle on front or back; cable, 25-ft
loom-covered with stage plug and furnished separate
from lamp for attachment with pin plugs; switch,
100-amp; weight, 117-lb head, 16-lb cable and
37-lb pedestal. Accessories, sold separately but
also described in a Mole-Richardson brochure,
are: barn door, diffuser frame and shutter.

Fundamentals of Magnetic Recording

is a 50-page handbook covering the subject
chiefly under these headings: magnetic
recording method, magnetic relations, bias,
erasing, output, uniformity of output,
frequency response, distortion and noise,
modulation noise, tape construction, head
wear, printing, splicing, selecting a tape
recorder, and maintenance. Written by
C. J. LeBel, Vice-President of Audio
Devices, it is available at no charge upon
a written request to Audio Devices, Inc.,
444 Madison Ave., New York 22, N.Y.

Correction and Amplification: In the

description of a lightweight sound-proof
blimp, p. 274 of the March 1952 Journal,
the manufacturer of the Arriflex camera
was erroneously noted. The Arriflex is
made in Germany by Arnold & Richter
K.G., 89 Tuerkenstrasse, Munich, Western
Zone; and the sole agents in North
America are Kling Photo Supply Corp.,
235 Fourth Ave., New York 3, N.Y.

SMPTE Officers and Committees: The roster of Society Officers and the
Committee Chairmen and Members were published in the April Journal.

464

The Ansco Color
Negative-Positive Process

By HERMAN H. DUERR

The basic principles of the Ansco Color Negative-Positive Process are outlined.
The paper deals with the essential characteristics of the color film materials
used for the process and outlines the printing and processing steps required.
Methods used to comply with the requirements of the motion picture industry
in regard to color dupes for optical effects, protection masters, color negative
master dupes, and color release printing are described. Requirements of
sound and procedures to produce silver sound tracks are discussed.

I

N 1945,1 the Ansco Color Process
for professional motion pictures was
proposed. This process was based on
the principle of reversible development
of monopack materials. Several motion
pictures have been produced using this
process. It was realized, however, that
a color process using the negative-
positive approach would be preferable
for a number of reasons. In the first
place, such a process would follow more
closely the long established black-and-
white practices of the motion picture
industry. More important, however, a
color film process using the negative-
positive cycle is superior due to the
higher speed attainable and the con-
siderably greater latitude in exposure,
processing and printing.

The Color Negative-Positive process,

Presented on October 18, 1951, at the
Society's Convention at Hollywood, by
Herman H. Duerr, Ansco Division of
General Aniline & Film Corp., Bingham-
ton, N.Y.

however, presented many problems, par-
ticularly in regard to methods of pro-
viding dupes for optical effects, protec-
tion masters and other essential require-
ments for the production of feature
motion pictures. These problems have
now been satisfactorily solved and the
Ansco Color Negative-Positive process
will now replace the older process using
film types 735 and 732, requiring re-
versible development.

The Ansco Color Negative-Positive
Process, like the earlier reversal process,
is a subtractive color process, using the
principle of color-forming development.

There are three different 35mm color
film materials involved in the process:

Color Negative Film Type 843,
Color Dupe Negative Film Type 846, and
Color Positive Release Printing Film
Type 848.

In addition to these color film materials,
a panchromatic fine grain dupe film,
such as the Eastman Panchromatic Sepa-

June 1952 Journal of the SMPTE Vol. 58

465

Blue-Sensitive Emulsion Layer

Green-Sensitive Emulsion

Gelatin Surface Layer

Yellow Filter Layer

Gelatin Separation Layer

Red-Sensitive Emulsion Layer

Safety Film Base

_i Antihalation Layer

Not to scale

Fig. 1. Scheme of layer
arrangement of Ansco
Color Negative, Type
843, before processing.

ration Film, Type 5216, can be used for
making separations.

The three multilayer color film types
used in the negative-positive cycle are
similar in structure, although quite
different in other characteristics.

Figure 1 shows the layer arrangement
typical for the three color film types used
in the process. The conventional layer
arrangement, with a red-sensitive bottom
layer, a green-sensitive middle layer
and a blue-sensitive top layer, is being
used. This figure illustrates the Color
Negative Film Type 843 before proc-
essing.

As has been described before,2 the
three colors, cyan, magenta and yellow,
in the Ansco Color Process are formed in
their respective layers during one color
developing step. It is ore of the im-
portant characteristics of the Arsco
Color Process that the none! iffus ing,
colorless color couplers are dissolved
and uniformly distributed in the gelatin
of the photographic emulsion layers and
completely surround the individual light-
sensitive silver halide grains. After
exposure and during the color develop-
ment, the developing agent becomes
partially oxidized by reducing exposed
silver halide grains to metallic silver.
The partially oxidized developing agent
reacts with the color couplers to form

dye deposits. The oxidized color de-
veloping agent is soluble and can move
freely in the layer. In order to produce
a dye image in closest proximity with
the originally exposed grain, it is, there-
fore, important that the color coupler
surround the silver halide grain so that
the coupling reaction can take place
truly in situ with the silver halide grain.
This characteristic of the Ansco Color
Process is important and is responsible

EXPOSED (unprocessed) FILM

AFTER COLOR
DEVELOPMENT

AFTER
SILVER BLEACHING
Dye Image Only

Fig. 2. Scheme of three stages in
the processing of Ansco Color Film:

A. Silver halide grain before color
development;

B. Silver grain after color develop-
ment; and

C. Grain after silver bleach

466

June 1952 Journal of the SMPTE Vol. 58

Gelatin Surface Layer

Clear Gelatin Layer

Gelatin Separation Layer

Not to scale

for the good image sharpness and defi-
nition.

A schematic illustration of the mech-
anism of dye image formation taking
place in closest proximity to the exposed
silver halide grains is shown in Fig. 2.
The Ansco color couplers are immobilized
in the emulsion layers by means of the
specific chemical configuration of the
color coupler molecules. To provide the
characteristics of complete solubility
and at the same time immobility and
nondiffusing properties, the molecular
structure has been arranged in such a
way that a fatty acid molecule of large
molecular size, through a short linkage,
is chemically combined with the dye
coupler molecule.3 The fatty acid mole-
cule acts somewhat like an anchor, pre-
venting the diffusion of the dye coupler
and of the formed dye image within
the layer, as well as from one layer to
another.

A typical cyan colorformer of this
configuration, with substitution re-
ferred to as "fat-tail," is:

OH

I

O— NH— G17H37

S03H

Fig. 3. Scheme of color
negative layers after
color processing.

Ansco Color Negative Film, Type 843

Camera Requirements: The Ansco Color
Negative Film, Type 843, can be exposed
in conventional motion picture cameras
as they are used for black-and-white
photography. The only additional re-
quirement is that, for maximum image
definition, the lenses used should have
good color correction.

Film Characteristics: The layer arrange-
ment of the Color Negative Film, Type
843, before processing has been shown
in Fig. 1. The layers after processing
are shown in Fig. 3.

More recently the Type 843 film has
been supplied on gray base instead of
on clear base with the soluble antihalo
back layer, and results regarding hala-
tion in motion picture practice have
been quite satisfactory. The gray base
does not interfere with the subsequent
printing operations and the absence of
a soluble back layer on the negative
film has certain advantages in processing.

Sensitometry: The sensitometric curves
of the three emulsion layers of the Color
Negative Film, Type 843, are illustrated
in Fig. 4. When developed to the
proper contrast for direct printing on
Color Positive, Type 848, or for the
preparation of black-and-white tricolor

Herman H. Duerr: Color Negative-Positive

467

Log Relotive CxpotuVt

,

20

separations, the gamma values for the
three layers, measured as integral densi-
ties on the Macbeth-Ansco Color Densi-
tometer Model 12, should be approxi-
mately as follows:

Blue-sensitive layer (yellow) . . 1.25
Green-sensitive layer (magenta) . 1 . 00
Red-sensitive layer (cyan) ... 1 . 00

Spectral Sensitivity: The spectral sensi-
tivity of the color negative film is shown
in Fig. 5. The sensitivity peaks are at
450, 555 and 655 m/u, respectively. The
film is balanced for light of daylight

Fig. 4. Integral density curves
of color negative film, Type 843.

Fig. 5. Spectral sensitivity of
color negative film, Type 843.

4<X> 440 410 520 S60 600 64°

ABSORPTION CURVE
OF DYES

MO 560 600 640

WAVfllNGTH IN MILLIMICRONS

Fig. 6. Absorption characteristics of color negative film, Type 843.
468 June 1952 Journal of the SMPTE Vol. 58

II

Log Relative Exposure

Log Relative Exposur

20 23

Fig. 7. Integral density curves of color Fig. 8. Integral density curves of color
dupe negative film, Type 846. positive release printing film, Type 848

SPEC

THAI S

NSITIVI

rv

/Y

s^.

— s

-^

V^

\

^

K

/

f

><

V

* — *

>

/

/

\

S

A

\

410 S20 560 600 640

WAVELENGTH IN MILLIMICRONS

Fig. 9. Spectral sensitivity of color positive printing film, Type 848.

quality. For interior illumination the
overall color balance should be approxi-
mately 5400 K. While the overall color
balance is not critical, it is important
that the different light sources on a set
be balanced to the same spectral quality.
The use of ultraviolet filters, such as
Ansco UV-16 or Wratten Filter No. 1,
for outdoor exposures and indoor ex-
posures with arc lights is recommended.

Absorption Characteristics: The dye
images recorded in the color negative
film are in colors complementary to the
colors of the original. The absorption
characteristics of the dyes in the color
negative film are illustrated in Fig. 6.

The absorption maxima are: yellow,
440 m/i, magenta, 540 rmi and Cyan,
675

Sensitivity and Resolving Power: The
color negative film has an exposure
index of 16. The resolving power, as
measured by the method proposed by
Sayce,4 is 44-48 lines/mm.

Ansco Color Duplicating Negative
Film, Type 846

Film Characteristics: The Color Dupe
Negative Film is similar to the Color
Negative Film Type 843 in layer arrange-
ment and color absorption characteris-
tics. However, the color sensitivity is

Herman H, Duerr; Color Negative-Positive

469

different from that of Type 843, but it is
similar to that of the Color Positive Film
Type 848, as shown in Fig. 9. In order
to produce dupe negatives with fine
grain and good resolution, the emulsions
used for this film type are much slower.
The exposure index is approximately
0.6 to 1.0. This film type is also on
gray base of the density of the Negative
Type 843 so that it can be readily inter-
spliced with it. The resolving power is
approximately 66 lines/mm. The sensi-
tometric characteristics of the color
dupe negative film are shown in Fig. 7.
The color dupe negative film is used
to make color negative dupes from tri-
color separation positives made from
the color negative originals. Optical
effects, fades, lap dissolves and other
special effects can be introduced via
these color dupe negatives. The various
methods which can be used to obtain a
color positive release print will be de-
scribed later, below.

An sco Color Positive Release Printing
Film, Type 848

Film Characteristics: In emulsion layer
arrangement the color positive release
printing film is similar to the color
negative film shown in Fig. 1. The
color positive film can be exposed either
directly from color negative originals,
from color negative dupes or from black-
and-white tricolor separation negatives.
The sensitometric curves of the indi-
vidual layers of the color positive film,
plotted as integral densities, are shown
in Fig. 8. In Fig. 9 the spectral sensi-
tivity of the color positive printing film
is illustrated. Good separation of the
spectral sensitivity ranges with a mini-
mum of overlaps is desirable for good
color reproduction in the printing film.

Dye Absorption: The absorption charac-
teristics of the dyes produced in the color
positive film are different from those in
the color negative and dupe negative
film. The absorption maxima are: 440
rm* for the yellow layer, 540 m/u for the

magenta layer, and 660 in/z for the cyan
layer.

Film Speed and Resolving Power: The
sensitivity of the color positive film is
similar to black-and-white positive, ap-
proximately exposure index 1.5. The
resolving power is 64—66 lines/mm.

All three color film types used in the
Ansco Color Negative-Positive Process
are on low-shrink, safety base.

Color Processing Procedures
and Solutions

The three color film types used in the
process require very similar processing
steps and processing solutions. Only the
Color Negative Film, Type 843, requires
a different color developing solution.
The color dupe negative film and the
color positive release printing film can
be developed in the same solutions
throughout. The color developing time
of these two types is different, as shown
in Table I.

For uniform processing of all types of
color film materials good control of
the processing solutions at all times is
very important. Basic control pro-
cedures which apply also to the handling
of color negative-positive have been
described by Bates and Runyan,5 while
analytical procedures to control and
maintain solution strength and uni-
formity have been presented by Brunner,
Means and Zappert.6 General informa-
tion in regard to color sensitometry may
be found in the report of the SMPTE
Color Sensitometry Subcommittee.7

Methods of Release Printing
From Ansco Color Negatives

Methods of release printing from
Ansco color negatives for the printing
of Ansco color negative originals from
different methods and certain variations
thereof can be used to produce color
positive release prints. These methods
are summarized in Fig. 10.

Not all of these methods are equal in
regard to color quality and cost. A
more detailed discussion of the ad-

470

June 1952 Journal of the SMPTE Vol. 58

Table I. Processing Steps and Materials for the Three Color-Film Types
Used in the Ansco Color Negative-Positive Process.

Alkaline pre-bath *
Jet rinse
Color developer
(608)
Color developer
(609)
Rinse (859B) . . .
Hardening fixer
(804)
Wash
Bleach (71 5A) . .
Wash
Fixer (800) . . .
Wash

Neg.
Neg. Dupe Pos.
Type Type Typ£
843, 846, 848,
min min min
2

Potassium iodide (KI) .... 3 mg
Water to 11

(Developing time for 843, 10-12 min at
68 F)

30 sec
10-12 6-7

11-14
30 sec 30 sec 30 sec

888
444
666
444
8 8
66 —
25 25 —
'-bath prepares the
lo layer for removal

#122 A Replenisher for Ansco Color Developer
#608
Water (60-70 F) 750 ml

Hexametaphosphate (Calgon or
Nullapon BFC) ...... 1 g

Sodium sulfite (Na2SO3) . . . 3.75g
Ansco S-5 Color Developer Salt 11 g
Sodium carbonate
(Na2CCvH2O) 75 g

Sodium sulfate (anhydrous) . . 30 g
Potassium bromide (KBr) ... 1 g
Ansco DA-1 Accelerator Solu-
tion 6 ml
Sodium hydroxide (NaOH) ap-
prox. * 2 g
Water to 11

Drv

* The alkaline prc
alkali-soluble antiha
by the jet rinse.

* The hydroxide content is adjusted to
obtain a pH 0.45 higher than that of a
fresh mix of 608 Developer.
(A representative replenishment rate is
11 1/1000 ft of 35mm Color Negative
Type 843.)

Air squeegeeing and edge treat-
ment of sound track (silver
track) 30 sec
Wash 2 min
Final fix (800) 4 min
Wash 8 min

Stabilizing bath . .
Rinse

2 min
1 sec

Ansco Color Developer 609
Water (60-70 F) 900 ml

Drv .

. . 25 min

Processing Formulas

Ansco Color Negative Developer 608

Water (60-70 F) 900 ml

Hexametaphosphate (Calgon or

Nullapon BFC) 1 g

Sodium sulfite (Na2SO3) ... 3 g
Ansco S-5 Color Developing

Agent 7 g

Sodium carbonate

(Na2CCvH20) 75 g

Sodium sulfate (anhydrous) . . 30 g
Potassium bromide (KBr) ... 2 g
Ansco DA-1 Accelerator Solu-'

tion (5%) 5 ml

Hexametaphosphate or Nulla-
pon BFC 1 g

Sodium sulfite (Na2SO3) ... 2 g
Ansco S-5 Color Developing

Agent 5 g

Sodium carbonate

(Na2CCvH2O) 60 g

Potassium bromide (KBr) ... 1 g

Potassium iodide 1 mg

Water to 11

Normal pH approx. 10.5, as determined
with Beckman Model G pH meter with
long-range electrode type 42 or equivalent.
Developing time for 846, 6-7 min at 68 F.
Developing time for 848, 11-14 min at 68 F.

Continued on the following page.

Herman H. Duerr: Color Negative-Positive

471

Table I — Concluded.

#706C Replenisher for Ansco Color Developer

#609

Water (60-70 F) 750 ml

Hexametaphosphate 1 g

Sodium sulfite (Na2SO3) ... 3 g
Ansco S-5 Color Developer Salt 7 . 5 g
Sodium carbonate

(Na2CCvH2O) 60 g

Sodium hydroxide (NaOH)

approx. * 2 g

Water to 11

* The hydroxide content is adjusted to

obtain a pH 0.4 higher than fresh mix of

#609 developer.

A representative replenishing rate is 11

1/1000 ft of 35mm Color Negative Type

843.

Rinse (859B)

Acetic acid (glacial) 3 ml

Sodium acetate (anhydrous) . . 30 g

Water to 11

pH fresh about 5.4

Replenisher (858}

Acetic acid (glacial) 10 ml

Sodium acetate (anhydrous) . . 20 g

Water to 11

Replenish continuous to maintain pH 5.4
to 5.8.

Hardening Fixer (804}

Water 750 ml

Chrome alum

[KCr(S04)2-12H20] .... 30 g

White alum [KA1(SO4)2 • 1 2H2O] 20 g

Sodium acetate (anhydrous) . . 10 g

Sodium sulfite (anhydrous) . . 10 g

Hypo (Na2S2O3-5H2O)

Water to

pH = 4.0

200 g
1 1

Bleach (715A}

Water 750ml

Hexametaphosphate */2 g

Potassium ferricyanide .... 100 g

Sodium acetate (anhydrous) . . 40 g

Acetic acid (glacial) 2.25ml*

Water to 11

* Vary to adjust to pH 4.5-4.7.

Fixer (800}

Water 750 ml

Sodium thiosulfate

(Na2S2Cv5H20) 200 g

Water to 11

Stabilizer

2 % solution of formaldehyde

Sound Track Developer

Solution A

DA-5 lg

Metol 20 g

Sodium sulfite (anhydrous) . . 40 g

Hydroquinone 20 g

Sodium thiosulfate 3 g

Formalin (37%) 20 ml

Water to . 11

Solution B

Thickener (Cellosize) Stock Solution
Hydroxyethyl cellulose (WP40) 45 g

Water to make 11

For use: Add 100 ml of Solution B to
900 ml Solution A and add 200 cc of Thick-
ener Stock Solution.

vantages and disadvantages of each of
these methods is, therefore, in order.

Method A. Printing From Original Color
Negatives Interspaced With Opticals on
Color Dupe Negatives

This method, shown schematically in
Fig. 11, comes closest to present black-
and-white practices, at least as far as
domestic releases are concerned. The

color negative originals, which may
represent 60-75% of the total footage,
are interspliced with optical effects
such as fades, lap dissolves, etc., made on
Color Dupe Negative Film Type 846
Tricolor separations on Fine Grain
Duplicating Pan Film are made from
the full-length negative. They are used
as protection masters. The optical
effects are produced from sections of

472

June 1952 Journal of the SMPTE Vol. 58

Convtnnonol
r B&W Sound
Negotive

ANSCO COLOR
Negative Type 843

A B

Printing

Methods C D

Set of 3 Block-
and-White color
separation posi-
tives made on
panchromatic
dupe positive
stock for optical
printing

Set of 3 black-
and-white color
separation posi-
tives made on
panchromatic
dupe positive
stock for contact
printing

Color Master Posi-
tive produced on
Type 846 Color
Dupe stock by
successive expo-
sures through tri-
color separation
filters

•

~~ - — —

Original Color
Negotive inter-
spliced with opti-
cal dupes (from
tricolor separa-
tion positives) on
COLOR DUPE
STOCK TYPE 846
for
contact printing

Color Dupe Nego-
tive on Type 846
Color Dupe stock
(including opti-
cals)

for
contact printing

Set of 3 black-
and-white color
separation nega-
tives on special
finegrain pan-
chromatic dupe
negative stock—
for optical print-
ing through nar-
row band filters

Color Dupe nega-
tive (including
opticals) pro-
ducedonType846
Color Dupe Nega-
tive stock by suc-
cessive exposures
through tricolor
separation filters
—for optical
printing

Release Prints on
Color Postive
Type 848

Release Prints on

Color Postive
Type 848

DAILY RUSHES
on Color Positive

(Also for editing)

Release Prints on
Color Postive
Type 848

Release Prints on
Color Postive
Type 848

1

Fig. 10. Summary of methods which can be used for release printing.

these black-and-white separations by
printing on Color Dupe Negative Film
Type 846. Method A, which involves
a minimum of color printing by the use
of color negative originals except for
opticals and effect shots, leads to the
best color quality. This method would
be first choice for domestic releases.

Method B. Printing From Full-Length
Master Dupe Negatives

In Method B, as shown in Fig. 12,
release printing is done from master
color dupe negatives. This method is
recommended where the original color
negatives cannot be made available for
release printing. This is frequently
the case for foreign releases. Tricolor
separations on panchromatic duplicating
film are made from the cut negative.
Master color dupe negatives on Type

846 are made from all scenes, including
opticals and special effects. Scene-to-
scene conformance can be attempted in
making the separations, as well as in
printing the master dupe negatives, so
that only minor color balance and light
corrections have to be made during the
release printing steps. The black-and-
white separations also serve as protection
masters.

As in Method A, the release printing
is done by contact printing. In both
methods conventional equipment, such
as a Model D or Model E Bell & Howell
or similar printers, can be used.

The filters required to correct for the
overall and scene-to-scene color balance
variations in printing are determined
by the use of a color scene tester similar
to the one described by F. P. Herrnfeld.8

On the Model D printer provisions

Herman H. Duerr: Color Negative-Positive

473

OPERATION OR
PROCESS

Convcntionol
Comera Exposure

PRINTING METHOD A

C-Olor

Development

llock-ond-White
Postive Color
Seporotions

Optical Print from
Three Originals

Through Successive
Tricolor Filters

Contact Printing,

Color Development
and Track Develop-
ment

FILM TYPE USED

Ansco Color

Negative Type 843

Finegrain Pan
Dupe Film

Ansco Color
Dupe Type 846

Ansco Color
Positive Type 848

* Black-and-White Protection Masters

Fig. 11. Method A: Printing from original color negatives interspliced with
opticals on color dupe negatives.

should be available for the insertion of
color balance filters.8 Following a sug-
gestion made by the Metro-Goldwyn-
Mayer Laboratory, a special material
for colored traveling mattes for the
Model E Bell & Howell printer has been
made available. Film base dyed uni-
formly to produce various color filter
combinations, coated with positive fine-
grain emulsion, is exposed and processed
by the Laboratory to produce a "variable
width" type light control strip in the
center of the film, as shown in Fig. 13.

Appropriate lengths of different
colored matte negatives, representing the
various light and color balance changes
are spliced together. This colored travel-
ing matte automatically corrects for
scene-to- scene variations in color balance
and density and the full speed of the
printer can be utilized.

In the preparation of master color
dupe negatives on Type 846 film, the
following sensitometric conditions are
representative:

Negative-Positive Duplication Control Gammas

G R

B

1.15 1.00 1.00
0.75 0.75 0.75

1.20 1.10 1.10

Color Negative Type

843

Separations* ....
Color Dupe Negative

Type 846 ....
Color Positive Release

Print Type 848 . . 3.00 2.70 2.70
Sensitometric test strips exposed with
a light source approximately 3200 K,
using an intensity scale sensitometer and
measured on a Macbeth-Ansco Model 12
Color Densitometer.

* Black-and-white separations exposed on
an Eastman Type lib Sensitometer and
measured on Western Electric RA-11QOB
Densitometer.

474

June 1952 Journal of the SMPTE Vol. 58

OPERATION OR
PROCESS

Conventional
Camera Exposure

Color Negative
Development

Block-ond-Wh.te
Postive Color
Separations

Optical Print from
Three Originals
Through Successive
Tricolor Filters

Contact Printing,
Color Development
and Track Develop-
ment

PRINTING METHOD B

FILM TYPE USED

Ansco Color
Negotive Type 843

Finegroin Ron
Dupe Film

Ansco Color
Dupe Type 846

Ansco Color
Positive Type 848

•Block-ond-Wh.te Protection Masters

Fig. 12. Method B: Printing from full-length master dupe negatives.

Fig. 13. Traveling matte for light and color balance control on Model E-type
printer. Film on the left side of the splice is color correction filter density CC10Y.
Film on right side is CC filter density 15M-(-0.05Y.

Herman H. Duerr: Color Negative-Positive

475

OPERATION OR
PROCESS
Convenhonol
Comera Exposure

Color Negative

PRINTING METHOD C

Block-ond-White
Postive Color
Seporations

Block-ond-whire
Negotive Color
Separations

Opticol Printing
Color Development
ond Track Develop-

FILM TYPE USED

Ansco Color

Negotive Type 843

Finegrain Pon
Dupe Negotive

Finegrain Pon
Dupe Film

Ansco Color
Positive Type 848

'Block-and-White Protection Master*

Fig. 14. Method C: Release printing
from black-and-white separation negatives.

The method described next requires
optical printing and is shown schemati-
cally in Fig. 14.

Method C. Release Printing From Black-
and-White Separation Negatives

In Method C three-color separation
positives on fine-grain Pan Duplicating
Film are made from the color negative
originals. These positives are printed
on the same fine-grain duplicating film,
this time developed to a lower gamma
negative. Optical effects can be intro-
duced during this printing step. The
black-and-white three-separation nega-
tives, including the optical effects, are
used for release printing on Ansco Color
Positive Film Type 848, preferably using

multihead printers with good registra-
tion.

Method C avoids one color printing
step as compared with Method B, and
if very carefully controlled allows some-
what higher color brilliance. However,
due to the fact that optical printers have
to be used, the release printing is con-
siderably slower and the method requires
great accuracy in sensitometric and
registration control, and for that reason
is not generally recommended. A
fourth method not requiring tricolor
separations should also be mentioned.
Although the color degradation produced
by this printing Method D is definitely
noticeable, results have been better
than expected. This method is briefly
outlined in Fig. 15.

476

June 1952 Journal of the SMPTE Vol. 58

OPERATION OR
PROCESS

Conventional
Camera Exposure

Color Negative

Development

Optical Printing
Color Postive
Development

Optical Printing
Color Negative
Development

Contact Printing,
Color Development
and Track Develop

PRINTING METHOD D

Blue Filter Green Filter Red Fitter

FILM TYPE USED

Ansco Color
Negative Type 843

Color POSITIVE
Release Prints

Conventional
B&W Sound
Negative

1

Ansco Color
Dupe Type 846

Ansco Color
Dupe Type 846

Ansco Color
Positive Type 849

Fig. 15. Method D: Release printing from
color dupe negatives via color dupe positives.

Method D. Release Printing From Color
Dupe Negatives via Color Dupe Positives

In Method D color positive prints on
Color Dupe Film Type 846 are made
from the original color negatives using
sharp cutting filters. The filters recom-
mended are Ansco UV-16 for all print-
ing steps in addition to the three-color
separation filters:

Wratten Filter No. 70;

Wratten 16 plus Wratten 61; and

Wratten 23 plus Wratten 48A.

These filters are also recommended for
making the three-color separations in
Methods A, B and C.

The Color Dupe Film Type 846 is
developed as a color positive. Optical
effects can be introduced at this step
or the next one, in which the color
positive dupe is again printed on Color

Dupe Stock 846 with the same sharp
cutting filters. This time the 846 Film
is developed as a low-contrast color
negative. The contrast of this dupe
negative should be kept as closely as
possible to the same contrast as the orig-
inal color negative. This second
generation color dupe negative can be
used for release printing on a con-
ventional contact printer. The Method
D does not provide for black-and-white
protection masters. For this reason this
method is not recommended for feature
pictures. A description of this method
has been included because there may
be occasions where this procedure may
offer certain advantages; also, the fact
that color rendition is still quite accept-
able is a good indication of the flexibility
of the Ansco Color Negative-Positive
Process.

Herman H. Duerr: Color Negative-Positive

477

Sound on Ansco Color Release Printing
Film Type 848

The reproduction of sound from multi-
layer color films using developed dye
images has for some time presented a
problem, especially in connection with
the red sensitive photocell, which is
today the standard for 35mm motion
picture projection.9 In order to obtain
a track which is efficient in absorption
in the infrared region of the 868-type
phototube, a method to produce a
combination silver-plus-dye track having
response characteristics similar to the
conventional black-and-white silver
tracks has been worked out.

Sound Track Development

As shown in Table I, the Color Posi-
tive Release Printing Film Type 848,
after color development, fixing, bleach-
ing and washing, is surface-dried by
effective air squeegeeing. At this
stage the sound track area carries a
sound image consisting of a dye image
from the original color developing step
plus a silver ferrocyanide image, pro-
duced in the silver bleaching step. Using
an applicator wheel or a pen-type ap-
plicator, a high viscosity rapid developer
solution is applied to the sound track
area only. This developer reduces the
silver ferrocyanide-plus-dye sound image
to a metallic silver + dye image. For
the selective treatment of the sound
track area, the following steps are
important.

1. Effective air squeegeeing to remove
surface moisture. The air squeegee
should be close to the applicator station
to prevent diffusion of moisture to the
surface of the emulsion before developer
solution is applied in the form of a bead
covering the sound area only.

2. Application of high viscosity sound
track developer, treating time approxi-
mately 30 sec.

3. To accelerate the development of
silver track, infrared heat lamps at this
stage are advantageous.

Cross-modulation and listening tests
have indicated that variable-area sound
negatives used for printing Color Posi-
tive 848 should have about the same
densities as used for printing on black-
and-white positive fine-grain film.
Densities between 2.40 and 2.70, as
read on a Western Electric R A- 11 00
Densitometer, are satisfactory. Sound
printing with filtered light to confine the
sound image to the two top layers is
preferable. The top layer alone may be
used for variable-area tracks.

The variable-area silver-plus-dye track
of the edge-treated color positive film
shows very good cancellation, fully
equal to black-and-white tracks. The
contribution of, and the effect of the dye
image underlying the silver track image
is insignificant in terms of the 868-type
phototube response. A yellowish stain
in the track area reduces the volume only
by about 2 db.

Experience with variable-density re-
cording is still somewhat limited, al-
though satisfactory recordings have been
made. In order to produce satisfactory
gradation and resolution characteristics,
the sound track should be confined to
the two top layers, with equal contribu-
tions by both layers.

Acknowledgments

The development work reported in
this paper represents the combined
efforts of many people of the Ansco
Research and Development Dept., as
well as the Ansco technical staff in
Hollywood. The valuable assistance of
the Metro-Goldwyn-Mayer Laboratory,
in particular J. M. Nickolaus, J. Arnold
and D. Shearer, in cooperating on the
various phases of the process and in
supplying sample negatives and dupes
for this presentation, is gratefully
acknowledged.

References

1. H. H. Duerr and H. C. Harsh, "Ansco
Color for professional motion pictures,"
Jour. SMPE, 46: 357-367, May 1946.

478

June 1952 Journal of the SMPTE Vol. 58

2. F. Wing, "Ansco Color Film," Ansconian,
3-11, Sept.-Oct. 1943. J. L. Forrest,
"Machine processing of 16mm Ansco
Color Film," Jour. SMPE, 45: 313-326,
Nov. 1945.

3. W. Schneider, A. Frohlich and H.
Schultze, "Die diffusionsfesten Farb-
bildner des Agfacolor Films," Chemie,
57: 113-116 (DEZ.) 1944.

4. L. A. Sayce, Photographic Journal, 80:
454, 1940.

5. J. E. Bates and I. V. Runyan, "Proc-
essing control procedures for Ansco
Color Film," Jour. SMPE, 53: 3-24,
July 1949.

6. A. H. Brunner, Jr., P. B. Means, Jr.,
and R. H. Zappert, "Analysis of de-
velopers and bleach for Ansco Color
Film," Jour. SMPE, 53: 25-35, July
1949.

7. A Report of the Color Sensitometry
Subcommittee, "Principles of color
sensitometry," Jour. SMPTE, 54: 653-
724, June 1950. (Reprinted as a book-
let.)

8. F. P. Herrnfeld, "Printing equipment
for Ansco Color Film," Jour. SMPTE
54: 454-463, Apr. 1950.

9. R. Gorisch and P. Gorlich, "Reproduc-
tion of color film sound records," Jour.
SMPE, 43: 206-213, Sept. 1944.

A. M. Glover and A. R. Moore, "Photo-
tube for dye image sound track," Jour.
SMPE, 46: 379-386, May 1946.
R. O. Drew and S. W. Johnson, "Pre-
liminary sound recording tests with
variable-area dye tracks," Jour. SMPE,
46: 387-404, May 1946.
A. B. Jennings, W. A. Stanton and J. P.
Weiss, "Synthetic color-forming binders
for photographic emulsions," Jour.
SMPTE, 55: 455-476, Nov. 1950.
J. L. Forrest, "Metallic-salt track on
Ansco 16mm Color Film," Jour. SMPE,
53: 40-49, July 1949.

Discussion

/. G. Frayne: Dr. Duerr mentioned a
variable density track density of 0.85.
This would be high for unmodulated
density and would yield low level output.
Do you propose using variable density
tracks that are that dark?

H. H. Duerr: The density of 0.85
obtainable in the top layer alone referred
to, is the maximum density. The un-
modulated, unbiased operating density
would, of course, be considerably lower
and closely related to regular black-and-
white practice. The best median density
has to be established by further tests.

Dr. Frayne: Does this density figure
0.85, include the base?

Dr. Duerr: No. This is the maximum
density obtainable in the top layer.

C. R. Daily: Do you intend to produce
a tungsten-type film for use with a color
temperature of approximately 3350 K?

Dr. Duerr: We are now producing only
a film for daylight-type illumination, but
expect to have a tungsten-type film avail-
able later on. Whether it will be balanced
for 3350 K or a somewhat lower color
temperature is not yet certain.

Frank E. Carlson: You referred to a color
temperature of 5400 K for the negative
film. Is the film balanced to the spectral
emission of a black body radiator at that
color temperature?

Dr. Duerr: Yes.

Richard H. Ranger: I have no question,
but would like to compliment Dr. Duerr
on his presentation, because the work shown
here tonight represents great strides over
the results demonstrated to a group of
engineers in Wolfen shortly following the
end of hostilities in Germany several years
ago.

Herman H. Duerr: Color Negative-Positive

479

Multiple-Image Silhouette Photography
for the NOTS Aeroballistics Laboratory

Bv ERNEST C. BARKOFSKY

A technique of multiple-image silhouette photography has been developed
for the NOTS Aeroballistics Laboratory. Six or more silhouette images of
missile models are imposed at a high rate upon a single photographic plate.
A series of such plates is used in precision photogrammetry to determine the
orientation and position of the models in transonic and supersonic flight.
While neither stroboscopic nor silhouette photography is unique in itself, it
is believed that the combination of the two, as described in this paper, is a
new technique.

I\ TECHNIQUE of multiple-image sil-
houette photography has been de-
veloped by the Ballistics Division, Re-
search Department, of the U. S. Naval
Ordnance Test Station, Inyokern, for
utilization in the NOTS Aeroballistics
Laboratory. This technique is used in
precision photogrammetry to determine
the position and orientation of missile
models in transonic and supersonic flight
through the Laboratory. In order to
obtain the desired accuracy in the aero-
dynamic and ballistic coefficients of the
missile models, it is necessary that the
mean deviation of a number of com-
parator measurements on the photo-
graphic images does not exceed a few
ten-thousandths of an inch. It was
found that this accuracy in measurement

Presented on October 16, 1951, at the
Society's Convention at Hollywood, by
Ernest C. Barkofsky, Ballistics Div., U.S.
Naval Ordnance Test Station, Inyokern,
China Lake, Calif.

could be attained only by photographing
the models in silhouette with micro-
second-duration light flashes.

Considerations of economy and effi-
ciency, however, demanded that a
minimum of six silhouette images be
recorded (at rates up to 3000/sec) on
a single photographic plate; and experi-
mental development of this technique
has resulted in multiple-image silhouette
photography of the desired quality.

The NOTS Aeroballistics Laboratory

The NOTS Aeroballistics Laboratory
is a high-precision, enclosed range; an
exterior view of the Laboratory is shown
in Fig. 1. Inert missile models will be
launched from a 3-in. gun and will pass
through the 500-ft-long range building.
The missiles will be photographed at
4-ft intervals during their flight, with
photographic coverage provided by 23
pairs of precision ballistics cameras
positioned so that the fields of view of

480

June 1952 Journal of the SMPTE Vol. 58

Fig. 1. The NOTS Aeroballistics Laboratory; Gun Platform and Control Room
at right; Missile Stop at left.

adjacent cameras are overlapping. This
relationship of the cameras is shown in
Fig. 2. Each camera will photograph
the missile six times, so that each photo-
graphic plate will bear six images for
assessment. Figure 3 is a cross-sectional
schematic drawing of the Aeroballistics
Laboratory, and shows the orientation
of the pair of cameras at each of the 23
Stations of the Laboratory. Figure 4 is
a down-range view. Associated with
each camera is a bank of three electrical-
discharge flash lamps, the sources of
microsecond-duration illumination of
the missile models in transonic and super-
sonic flight. Much instrumentation was
required to make possible the high-speed,
multiple-image silhouette photography.

Instrumentation

The desired accuracy in the determi-
nation of the variation of position and
orientation with time of the missile
models to be studied in the Aeroballistics
Laboratory imposed severe specifications
in the performance requirements of the
necessary instrumentation. In particu-
lar, detailed consideration had to be
given to the many factors contributing
to the errors of the photogrammetry:
camera and lens; camera survey;

photographic technique, including emul-
sion, developer and development combi-
nation; geometry of the camera-rocket
model array; and exposure time and
timing. An ultra-precision ballistics
camera was designed and developed;
a high-quality, wide-angle lens was
selected for the camera; and a catenary
system was designed to permit the
calibration of the camera plates. A
new photographic technique, that of
multiple-image silhouette photography,
was developed. The production of
microsecond-duration flash illumination,
precisely timed, required a complex
electronic and electrical system includ-
ing: (1) a photoelectric triggering
system which provides for the start of the
flash lamp illumination by the passage
of the missile model through a light
screen; (2) an electronic "gate" which
stops the lamp flashing at each station
after the rocket has passed from the
field of view of the camera; (3) a light-
flash counting system; (4) a master
electronic timing system; (5) an elec-
tronic monitoring system: (6) a high-
voltage power source; and (7) an elec-
tronic system for simulating the transonic
and supersonic flight of a missile model
through the Laboratory.

Ernest C. Barkofsky: Multiple-Image Photography

481

Fig. 2. 45° longitudinal schematic presentation of the
NOTS Aeroballistics Laboratory.

.^-CIRCLE OF
, __ TRANSONIC

RAr) / AERODYNAMIC
CLEARANCE

Fig. 3. Transverse schematic presentation of the NOTS Aeroballistics Laboratory.
482 June 1952 Journal of the SMPTE Vol. 58

Multiple-Image Silhouette Photography

The technique of multiple-image sil-
houette photography was developed
expressly for use in the NOTS Aero-
ballistics Laboratory. Since the location
of the transverse components of the center
of gravity of the rocket model must be
determined to within 0.001 ft, and since
the magnification of the ballistics camera
is I/ 39th, measurements on the photo-
graphic plate must be made to within
less than 0.0003 in. Because of the
other factors which contribute to the
error in the position determination, the
accuracy of the comparator measure-
ments on the photographic image may
be only a fraction of the 0.0003 in.
(= 7.5 fj.). The accuracy of measure-
ment possible with the best of com-

parators is of the order of one micron,
hence the quality of the photographic
image must be such that it in itself
introduces practically no error. The
characteristics of a photographic image
which make possible precise comparator
measurements upon it are sharply
defined edges and proper contrast be-
tween the image of the object and the
background against which it is photo-
graphed. It required but very little
experimentation to reveal the fact that
only by silhouette photography could
satisfactory photographs be obtained of
the highly polished missile models.

The photography of the missile models
in silhouette at each four feet of their
travel through the Aeroballistics Labora-

Fig. 4. Down-range view of the interior of the NOTS Aeroballistics Laboratory.
Ernest C. Barkofsky: Multiple-Image Photography 483

Fig. 5. Preliminary mul-
tiple-flash silhouette photo-
graphs, showing decrease in
contrast with increase in
number of superimposed
flashes.

tory would have been impossible be-
cause of the instrumentation cost in-
volved, if only one silhouette image on
each photographic plate were possible.
Investigation was therefore made of
the feasibility of obtaining more than a
single silhouette image on each plate.
A serious difficulty with such a procedure
is evident: each of JV silhouette images
on a single photographic plate will
have N — I flash exposures of the bright
background superimposed upon it, a
condition that may result in very little
or even no contrast between the sil-
houette images and the background.
Exploratory work revealed, however,
that even with an appreciable number
of silhouette images on a single plate,
discernable and even measurable images
could be obtained. Figure 5 shows a
series of photographs of a static model
obtained with ever-increasing amounts
of light superimposed upon the sil-
houette image. Measurements made by
an inexperienced comparator operator
upon one of the first of such a series of

test exposures had the following values
for the standard deviation of measure-
ment for ten readings on each image:

Ratio of Silhouette Image Illumination
to Background Illumination

0:1 1:2 3:4 9:10
Std. deviation, /* 8.6 3.3 6.9 9.2

An interesting fact revealed by these
preliminary measurements is that a true
silhouette image (a negative with an
opaque background and a transparent
image) is less amenable to accurate
measurement than an image with less
contrast to the background. On the
other hand, it is also evident that as
the contrast is further decreased, ac-
curate measurement again becomes
difficult.

The decrease in contrast with increase
in the number of superimposed light
flashes is shown by manipulation of the
simple equation of the straight-line por-
tion of the photographic characteristic
curve:

484

June 1952 Journal of the SMPTE Vol. 58

If

where
and
then

D
E
N
F
D
AD

y (Log£ - Logz).
NF,

Number of flashes
Energy per flash,
y (Log N F - Log i)
DN - Z>tf_i
N

JV- 1

(within

10%

JV =

for

6).

This equation shows that the difference
in density between a multiple-flash
silhouette image and its background
decreases as the total number of flashes
is increased. Despite this predicted
decrease in density difference, the
technique of multiple-image silhouette
photography has been developed to the
degree that six-image photographs can

i.O

0.8

>- 0.6

0.4

0.2

0.0

PANOTOMIC-X

17 MINUTES IN MICRODOL

r=0.85

2345678
NUMBER OF FLASHES

9 10

Fig. 6. Characteristic curve of emul-
sion-developer combination for multiple-
image silhouette photography.

I

I

J

J

Fig. 7. Flash-by-flash development of six-image silhouette photograph
in the NOTS Aeroballistics Laboratory.

Ernest C. Barkofsky: Multiple-Image Photography

485

Fig. 8. Five-image silhouette photograph of 20-mm projectiound.

at 2800 ft/sec, flash rate of 2000/sec; Scotchlite backgrle in flight

be assessed with an accuracy of only a
few microns.

This photography is done with
Panatomic-X emulsion developed for
17 min in Microdol. Figure 6 is the
characteristic curve obtained for this
emulsion-developer combination under
the conditions of flash photography in
the Aeroballistics Laboratory.

Insufficient density was obtained with
the use of flat-white paint or even
movie screen as the silhouette back-
ground at the distance of 30 ft from the
flash lamps, dictated by the geometry of
the Aeroballistics Laboratory. It was
found that No. 30 Wide-Angle Silver
Scotchlite sheeting was very satisfactory
for use as the silhouette background in
the Aeroballistics Laboratory. By the
use of Scotchlite, sufficient intensity was
obtained with the superposition of six
flashes on a single plate to yield nega-

tives of satisfactory density. Figure 7
shows the flash-by-flash development of
a six-image silhouette photograph.

Figure 8 is a high-speed, five-image,
silhouette photograph of a 20-mm pro-
jectile in flight at 2800 fps. This photo-
graph is representative of those to be
obtained in the NOTS Aeroballistics
Laboratory.

Summary

The technique of multiple-image sil-
houette photography, when employed
in conjunction with the instrumentation
of the Aeroballistics Laboratory, permits
the accurate determination of the aero-
dynamic and ballistic characteristics of
missile models. The information con-
cerning the performance of the models
can be extrapolated for utilization in
the design and development of improved
missiles.

486

June 1952 Journal of the SMPTE Vol. 58

Optical Problems

in High-Speed Camera Design

By JOHN C. KUDAR

C,

CONCERNING THE optical theory of the
rotating prism in high-speed cam-
eras, there have been Letters to the
Editor published in the Journal in July
1951. As the Letters dealt with more or
less controversial interpretations of the
problems involved in conventional high-
speed camera design, I am glad to have
this opportunity to expand upon the
applications of the basic principles to the
definite possibility of a promising and
perhaps unexpected development in this
field. It is necessary, however, to start
with a survey of the existing commercial
types.

The rotating polygonal prism as an
optical component, such as a cube or an
octagonal prism, is the generalized case
of the rotating plane -parallel plate.
One well-known high-speed camera is
based on this principle of the plane-
parallel plate. The image projected
through the rotating plane-parallel plate
moves with the same speed as the con-
tinuously traveling film, at least during
the short exposure time. Another well-
known camera differs in its optical con-
struction in that the plane-parallel plate
is replaced by a rotating cube or by an
octagonal prism. In all cases the expo-

Presented on October 16, 1951, at the So-
ciety's Convention at Hollywood, by
John C. Kudar, 1809f Las Palmas Ave.,
Hollywood 28, Calif.

sure is limited to small incidence angles
by blacking edge pieces which act as
shutters during the rotation.

The precise thickness of the plane-
parallel plate (or polygonal prism) must
be designed in accordance with the re-
fractive index, as well as in consideration
of the maximum incidence angle allowed
for exposure. The correct mathematical
formula for the thickness was well known
long before the design of the first high-
speed camera, since the principle of the
rotating prism projector came up at the
very beginning of motion picture films.

The two types of cameras discussed
have a common feature in their mechani-
cal design, which is that the film trans-
port at the place of exposure is geared
to the prism rotation. The development
of high-speed cameras depended essen-
tially on improvements in mechanical
manufacturing and design, while the
problem of optical precision was con-
cerned only with suitable limitations for
the incidence angles and with the precise
thickness of the plate or prism.

The refractive index of the glass ma-
terial can be shown to be without any
notable influence on the quality of image
formation by the rotating prism. Even
the dispersion is irrelevant, due to the
deliberate limitation of the incidence
angle during exposure. As the Letters
to the Editor in the July 1951 Journal

June 1952 Journal of the SMPTE Vol. 58

487

deal with these controversies in detail,
there is no need to go into this aspect of
the optical theory of the rotating prism.
As the choice of glass material for the
prism is left free from the point of view of
optical design, the final decision can be
made on the basis of other considerations.
With regard to the choice of the glass ma-
terial, the centrifugal stress in the rotat-
ing prism is worth careful consideration.
For the same angular velocity, the cen-
trifugal force acting on the peripheral
volume element is proportional to the
product of the thickness of the prism and
the specific gravity. The thickness of
the prism must be designed proportional
to:

(1)

n - 1'

if n is the refractive index. Then the
centrifugal force is proportional to:

(2)

n - \

if S is the specific gravity. For low-
index glass, we may assume an ordinary
crown glass with n = 1.5 and S = 2.5.
A representative high-index glass may
be a rare-earth glass with n = 1.8 and
S = 4.6. Consequently, the product (2)
is 7.5 for low-index glass, and 10.35 for
high-index glass. This shows that the
centrifugal force on the periphery of the
high-index prism is, instead of being
smaller, nearly 40% greater than on the
periphery of the low-index prism.

The optical and mechanical principles
underlying the construction of commer-
cial high-speed cameras with rotating
prism or plate have been known for half
a century. The task of exploring the
possibly hidden potentialities of the polyg-
onal prism method, beyond the scope
of conventional constructions, has long
been overdue. With respect to the
necessary high rate of rotation, it is of
particular interest to investigate whether
the mechanical properties could be essen-
tially simplified by introducing some un-
conventional optical means. It would

be desirable, without doubt, to avoid any
gearing between prism rotation and film
movement at the place of exposure.
The precise position of the film relative
to the prism facet could be maintained
much better with a single rotating unit,
serving the double purpose of mechanical
and optical movement. Such a single
rotating unit should incorporate the ro-
tating prism into a film sprocket. In the
conventional case, however, this is not
possible, as the correctly designed polyg-
onal prism is always larger than the
corresponding sprocket, supposing equal
number of frames and facets. Neglect-
ing minor design corrections, the thick-
ness of the prism, i.e., the distance D
between opposite parallel facets, is:

HNn

(3)

in which H is the full frame height, i.e.,
the film length per frame; N is the num-
ber of facets; and n is the refractive index.
In comparison, the diameter of the corre-
sponding film sprocket must not be
larger than:

HN

7T

(4)

These two formulas show that the polyg-
onal prism could be reduced to or below
the periphery of the film sprocket, but
only under the condition that the refrac-
tive index is not less than 2. That
means that all commercial high-index
glass is useless for this purpose.

In order to arrive at a practical solu-
tion involving a single rotating unit at the
place of exposure, we must abandon the
idea of the solid prism as a conventional
unit, which was found to require a refrac-
tive index higher than 2, if the prism has
to fit within the sprocket periphery.
The new polygonal prism has a composite
structure. It consists of rotating and
stationary components. Its rotating
peripheral part is polygonal on the out-
side and cylindrically hollow inside.
The periphery of the polygon fits into the
periphery of a corresponding sprocket.

488

June 1952 Journal of the SMPTE Vol. 58

The stationary components form a sepa-
rate unit, which fills the cylindrical
cavity inside of the rotating polygonal
ring, without obstructing the rotation.
The film is guided around part of the
periphery by sprocket teeth, where the
exposure takes place. The curvature of
the film on the periphery of the rotating
component is compatible with the optical
system.

The purpose of this device is to get rid
of the conventional gearing in the central
part of the camera, where the exposure
takes place. The feed and take-up
parts can remain similar to those already
in use. The making of hollow polygonal
prisms with twelve or more facets does
not represent great manufacturing diffi-
culties. Their use in high-speed cameras
will have the advantage that for the same
film speed the rate of rotation (and the
centrifugal force) is slowed down in pro-
portion to the increase of the number of
facets.

Film

Figure 1.

Many different designs are possible,
based on the principle of the composite
polygonal prism. The simplest possible
optical construction (Fig. 1) is shown dia-
grammatically in cross section through
the rotation axis of the polygon, and
parallel to the optical axis. In the cylin-
drical cavity of the rotating polygonal

prism two plane-convex cylindrical lenses
are placed in stationary position. The
plane-parallel air gap between the flat
faces of the cylindrical lertses increases
the image displacement by refraction, as
illustrated by two parallel rays coming
from the camera lens and refracted sev-
eral times. The diagram represents an
air thickness between the flat sides of the
two cylindrical lenses approximately
equal to the total glass thickness of the
composite polygonal prism, in which case
the refractive index of the glass (for the
polygon and the cylindrical lenses) must
be about 1.6. The narrow cylindrical
air menisci between the rotating polyg-
onal component and the cylindrical
lenses can be designed for zero power.
(In a similar model, actually built for
35mm film, the zero power air menisci
have been made 0.5 mm thick.)

It should be noted that in conventional
high-speed cameras a substantial distance
is kept between the film and the prism,
while in the case of the new device the
film under exposure is perhaps very close
to the prism facets. Therefore, black
shutter strips on the edges of the new
polygonal prism should not be used, as
they would limit the field vertically.
But the required shutter effect can be
achieved by suitable vertical limitations
on the internal surfaces of the composite
optical device.

Some important users of high-speed
cameras have a particular interest in
highest possible speeds. For such a pur-
pose, it is certainly desirable to have a
single rotating unit, and no gearing, in
the central part of the camera.

It is in the common interest of the
users and manufacturers of high-speed
cameras to show clearly the inherent po-
tentialities of the polygonal prism prin-
ciple. A related development in another
field is already going on. It is to be
hoped that the inadequate theoretical
approach, which has prevailed in this
country for the last few years, will not
prove a permanent obstacle in this par-
ticular field of high-speed photography.

John C. Kudar: High-Speed Optical Problems

489

Discussion

John H. Waddell: (an abstract of the re-
marks which preceded the projection of the In-
stitute oj Medical Research high-speed motion
picture) Fastax prism design has been
studied extensively and there are, of course,
constant improvements being made in the
cameras as they exist today. When one
realizes that the cameras have been de-
veloped without benefit of Government
sponsorship, but entirely privately, the
advances which have been made have been
noteworthy.

The photographic quality of the rotating
prisms can be observed in the pictures
which are going to be projected which

wefe taken at the Institute of Medical
Research by Dr. Myron Prinzmetal and his
associates. It will be seen that the day of
the rotating prism camera is not over.

Centrifugal force as discussed previously
has been misinterpreted. We have never
seen a prism itself explode. Failure of
the prism housing has occurred however,
and, in our design problems, the housing
has to be constructed so that it does not
fly apart when using at ultra high speeds.
It is interesting to note that the camera as
constructed today will take over twenty
g's, and that picture taking rates far in
excess of advertised rates have been suc-
cessfully achieved.

490

June 1952 Journal of the SMPTE Vol. 58

Effective Sum of Multiple
Echoes in Television

By A. D. FOWLER and H. N. CHRISTOPHER

Observers compared the interfering effect of multiple echoes with that of single
echoes in black-and-white television pictures. The multiple echoes were 2,
4 or 8 echoes of equal strength but different delays. The single echoes were
40, 35 or 30 db weaker than the main signal. A method for estimating addi-
tion effects of several echoes is presented and demonstrated to be consistent
with the test results.

JL HIS PAPER reports the results of tests
comparing the interfering effect of
multiple echoes with that of single echoes
in black-and-white television pictures.
The present study supplements an earlier
one in which the interfering effect of
single echoes was considered.1 Al-
though general in application, the test
results have a special bearing on the
design of television transmission systems,
where echo requirements are rather
severe and sometimes difficult to meet.

In the tests to be described, observers
viewed a standard black-and-white tele-
vision picture on which was super-
imposed, for ready comparison, either
a single echo or multiple echoes. The
single echo was fixed in level (a little
above threshold) during a given test;
the multiple echoes were then uniformly

Presented on April 21, 1952, at the So-
ciety's Convention at Chicago, 111., by
A. D. Fowler and H. N. Christopher,
Bell Telephone Laboratories, Murray
Hill, NJ.

adjusted in level until they were judged
to have the same interfering effect as
the single echo. The multiple echoes
comprised 2, 4 or 8 echoes of equal
strength, and each was assigned a delay
in the range of 3 to 14 /xsec. The spac-
ings, or delay differences, were uniform
in a few tests and random in most.

Over 100 comparison tests, each em-
ploying eight or more observers, were
made. This rather large number of tests
was necessary in order to explore the
effects of such things as number of echoes,
levels of reference echo, weightings of
echoes, spacings between echoes, poling
of echoes, and types of picture material.

The results of these tests yielded an
empirical relation by means of which
the effective sum of multiple echoes can
be estimated with reasonable precision.
This relation depends upon: (a)
weighted echo power; (b) number of
echoes; and (c) average spacing be-
tween echoes.

June 1952 Journal of the SMPTE Vol. 58

491

DELAY
NETWORKS

REFERENCE
ECHO SIGNAL

2 VOLTS
PEAK-TO-PEAK

Fig. 1. Simplified schematic of test setup.

Apparatus and Circuit Arrangement

The circuit arrangement for the tests
is shown schematically in Fig. 1. In
this diagram, various buffing and mixing
amplifiers have been omitted in the
interests of simplicity. It should be
understood, however, that each branch
of the circuit was properly isolated and
that each echo signal, apart from the
indicated delay, was essentially a replica
of the m#in signal.

Referring to Fig. 1, it will be seen
that the output of the scanner, which
derives composite picture signal from a
slide or film, provides three signals:
(a) main picture signal; (b) single
reference echo signal, delayed TO micro-
seconds; and (c) multiple-echo signal,
comprising 2, 4 or 8 component signals
(four are shown for illustration), each
delayed 7*i, 7"2, etc., microseconds,
respectively. The main signal is fed
to the viewing monitor via a three-way
mixing pad where echo signals are
introduced. A comparison switch, oper-
ated by the observer at will, selects
either single reference echo or multiple
echo for transmission via attenuator
No. 1 to the mixing pad. The multiple-
echo path includes attenuator No. 2,
by means of which an adjustable loss
may be added to that path as required.

Most of the apparatus was laboratory
constructed and conventional. The
viewing monitor was equipped with a
10-in., black-faced, metal-backed kine-
scope operated at about 11 kv. The
overall transmission, which otherwise
would have extended somewhat higher,
was limited to 4.3 me by a phase-
equalized low-pass filter.

Picture Material

Most of the tests were made with a
slide called Model White Hat. This
picture shows a close-up of a girl model-
ing a large white hat against a plain and
rather dark gray background. It was
known from previous experience that this
picture was very sensitive to single
echoes which were delayed by more
than two microseconds.

Other slides, although known to be
less sensitive to single echoes, were used
in the tests. These were used because
it was suspected that they might exhibit
unusually severe addition effects of
multiple echoes.

A motion picture film of Model White
Hat was employed in one test. This
was used to see if motion, itself, caused
some of the cloudlike multiple echoes to
be more readily noticed.

492

June 1952 Journal of the SMPTE Vol. 58

Procedure

A picture from the scanner was
established on the viewing monitor with
the proper highlight luminance and con-
trast ratio. Reference echo was then
set at a fixed level (40, 35 or 30 db
weaker than main signal) by adjustment
of attenuator No. 1. The observer
was then asked to view the picture with
reference echo present and, upon switch-
ing alternately from reference to multiple
echoes, to declare whether multiple
echoes had less or more interfering
effect. The experimenter would then
adjust attenuator No. 2 appropriately
to make the interfering effects more
nearly equal. When the interfering
effects were judged to be equal, the values
of the attenuators were recorded, No. 1
registering the level of reference echo,
and No. 2, the relative interfering effect
of the multiple echoes. The test was
repeated for each of the other observers.

Viewing was done in a darkened room
and from a distance of four times the
picture height. In two of the tests, the
viewing distance was changed to 13f
times the picture height.

As a check on the results obtained by
the comparison method described above,
a Comment Test was made. The general
procedure for this kind of test has
already been reported in some detail.2'3
A series of intermixed conditions, viz.,
single and multiple echoes at various
levels of each, were displayed in a
randomized sequence. Ten experienced
observers rated each condition by
choosing one of seven preworded com-
ments listed for the purpose.

It was apparent that the relative
interfering effect of single echoes of
different delays would play an important
part in the results. Accordingly, a
comparison test was made using single
echoes of various delays in the path
shown in Fig. 1 for multiple echoes.
The reference echo was delayed 7
jusec and was set at a level of 40 db below
main signal. The same procedure and

viewing conditions were employed as in
the multiple-echo tests.

Summary of Results

The results of each test, together
with the conditions under which the
test was made, are shown in Table I.
The essential test results are tabulated
under the heading Relative Interfering
Effect (db} in columns labeled Meas.
(measured) for the appropriate level of
reference echo. Entries in those col-
umns are the average values of the
settings of attenuator No. 2 for the
number of observers listed. Entries
in the columns labeled Calc. are corre-
sponding calculated values to be dis-
cussed below.

In summarizing the results of the
tests, it will be convenient to employ
the terms effective echo power, weighted
echo power and advantage. These terms
are defined as follows:

Effective echo power, expressed in db
above the physical power of a single
reference echo, denotes the relative inter-
fering effect as determined by subjective
tests. In db, its value is given numeri-
cally by the loss in attenuator No. 2, as
determined by judgments.

Weighted echo power, also expressed
in db above the physical power of ref-
erence echo, is the sum of the weighted
physical powers of the component echo
signals. The weightings are time (delay)
weightings of single echoes referred to
that of the reference echo.

Advantage is the ratio, expressed in
db, of weighted echo power to effective echo
power. When effective echo power is
less than weighted echo power, a positive
advantage (over power addition) ob-
tains; when the addition effects are
more severe than power addition, a
negative advantage obtains.

The principal result of these tests
may be summarized in an approximate
rule for estimating effective echo power
of multiple echoes:

effective echo power (db) = weighted echo power
(db) — advantage (db).

Fowler and Christopher: Multiple Echoes in Television

493

bo

.

II

.sp|

1

1

1 8

4;

1

I

Jll

•s

1

6

*i

°°

oo oo oo

"c

3

Z

*s S

8

o

y

r- r- T-H o\

r^-

C

o

CO CM CM T-H

CM

CO

t^

S

(hi

•o

'•3

(M

^S

CM CM 00 vO

t^

bo

c

O
en

u

Tj- CO CM CM

CM

1

"I

^

>

cO

.

CM oo o r^-

00 0 ON

.

s

3

j=3

CO T-H T-H O

T— 1 T-H O

•^

X

y

TD

«s

^tt

m .

cO 22

T-H CM CM m

T-H Tj- CM

OS

T3

d

8?
•d

CQ

1

TJ- CM CM T-H

CM O o'

3

CM

1

JJ

o o co vo o T-H m o m ^- T-H o m ^t- T-.
>o m -^ co co CM T-I co T-H o o co T-H o o

"1

^

i'

1

c^

1"

oo r-. ON o o vo CM co o oo i- T-H T-. ^- inmsoo CM ^ r- CM

^

£

a

*4J
P<

1

^7T?°++

™

H

14

•o o

r

0000000 00000 00 0-00

0 CO T-H

1

*J CJ

> -D

cO CO cO cO cO cO cO cO CO cO CM cO CO CO CM

CO O CM

*^

M

a"°

i

c

^>

3

O

o

. I

- 0

«

00 ^

00 ^)

„—_-,---— _~, ^H|C4 rH|c^

HN

H

j"

< w1

.5 S.

o T-HT-HCM cM^m T-iCM-^-m

CM CM

1

tx

X 0

i- t- r- r- t- t- r- ^H^H^IC, H«HNH«HI«

T-Ht^r^

^N

Jc

,

CO

2

fbo

o

K

|

|

CM

(

1

•^- ooooo OOCNOO OOCAOO

*

1

CM

Tm

u

-l«

£

•8J

0<5

494

June 1952 Journal of the SMPTE Vol. 58

O 0 0 00 00 00 O
CN O SO O

m in ^ Tt-

* '*

^g § si g> g>

if ill i| ? iii

Q 2,* oo oo

O oOcOCNr-iOcOOOOcO T-I

tance
distance

cO

distance

be
C

"o

a

V
(4

1

O

tance

*

\
V

CN

ho

•o

if

—

T;

T3

1 CN r- so r~-

•rt- so mcNr-ooofMO tnm T-H

.S

<Q«

i

iO

bo

m

cO

c

c

^fTt-Tj- CO

co* co oooor-oooo^vo OCN ^-

•£

CN

so

00

d

.jjj

'^

• m T-I oo o o ^j- oo

CNCNOOO^OCNO

]T

4-

01

d

' CO cO CO Lf> cO CN T-I

oooor-r^r-ir5voiocN

CN

UT

jQ

-o

^

=§

01

•B

- CN m CN oo Tt m T-I

(\i so inooootv-'^"u^CNooT-< o T-I

i'

CO®

§

coo

~

0
01

CN

> CO cO rf CN cO T-I CN

CN- CN' oo^sor^so^-vocoo CN ^:

CN

_o"

m

8

00

> oo T-H m so

COfO^^^OT- ^

i

I

0

' CN CN CO O

OOOOsOsOsOU"!^ 00

to

ft

ft

2

n

•_

42

.J

JO

"O

c

TJ

S

-a

"^ CN O tn

00 CN T-iT-ir^sOsOOm 00 CO CN

o
y

CO<M

8

COO

c

LO

^*-

> cO CN cO O

o' -' CN" oo m' m ^ co ^ cs' oo m'

£

J

^-

0

00

i

£

j"

> 00 CN sO CO CO rj- sO

so CN CN OO OO 00 t^ 00 t — O O 00

CN

CN

oj

00

oo

> m so CN CN so m m

in oocooooooot-^t^r^CNso m

^

oo

•M

oo

o

oo

ho

Jf

J_?

5 O so oo m CNin so

^o T-! T-. m tn 10 m m m T- o o

2

T-l

"far

u->

"i>

m

4 CS cN O O CN CN cO

^

bo

.-*"

"^

"rn

JH

^^

_c

*

_

.rT-1

^ -I'C-I'* rHl-*

00 OOOOOOOOOOCNf-CNCN T-I

|

0

1

00

J

oo

' T»- cO CO CO

•* ^ CO CO CO CO CO CO CO Tf Tf

•§•

^

£

cO

I

CO

#

1

-|

_

*

#

J CN ^ CN CN

cOcOT-ir-T-iT-icOCNcO CO CN

K

CO

^

Hie.

si

T-l

o CN h- r^ o"

*
CN CN 0 O O CN O h- CN CN O

u

3

CN

3

-IN

O

1

•

O

n|-4im|-4in|-4i -N «!•*

ffi

£

nl-*

«!•*

' oo so m

T- T-, CN CN CN so r- 10 T- £ 00

y|

CN

C\

*

*

*

m ^f m

oooooooomsOTt-o

o

T-l

— !•*

oo

00

•* -^- -<t oo

00 CN CN 00 00 00 •* m Tj- CN

CN

oo

oo

«|-<(I-|N

ml* mNi n|* n|*i — HI H|e« H N

«!•*

«!•*

CO CO

oooommmcocococN

00

ir>

10

CO CO cO

cO so so cO cO cO cO cO CO sO

so

CO

cO

•r^****

^ oooooooooooooooooo ^- ^

00

00

00

Fowler and Christopher: Multiple Echoes in Television

495

O~ = STANDARD DEVIATION
OF OBSERVATIONS

0.5 0.6 0.8 1.0 2 3 456789 10

ECHO DELAY IN MICROSECONDS

Fig. 2. Relative interfering effect of single echoes for picture Model White Hat.
Reference echo 40 db below main signal.

Weighted echo power, for the purpose
of checking the test results obtained with
Model White Hat, can be calculated
using the weighting curve given in
Fig. 2. Advantage, which depends on
the number of echoes and the average
spacing between echoes, is given ap-
proximately by the family of curves
(or straight lines) in Fig. 3. The rela-
tions shown in Fig. 3 apply when the
reference echo is 40 db weaker than the
main signal; at higher levels of reference
echo, there is progressively less ad-
vantage. This will be seen in Figs.
4 and 5, which apply to 35- and 30-db
criteria, respectively. The method of
determining the empirical advantage
curves from the test data is described in
Appendix I.

To show how well the test results are
reflected in the empirical advantage
curves, the latter have been used, to-
gether with the time weightings of Fig.
2, to "predict" the test results. A
comparison of the measured and com-
puted values of effective echo power,
expressed in db, is shown in Fig. 6. On
the whole, the correlation is very satis-
factory: 75% of the data points fall
within + 1 db of the predicted values;
91% fall within ±2 db.

The picture, Model White Hat, chosen,
as stated above, for its sensitivity to
single echoes, proved to be the most
sensitive of several pictures to multiple
echoes. The results were substantially
the same for a motion picture of Model
White Hat as for the slide taken from the
same film and used for most of the tests.

In a few tests the polarity of about half
of the echoes was reversed. This pro-
duced no significant change in the
results.

Increasing the viewing distance re-
sults in more severe addition effects.
This is more than offset, however, by
the accompanying decrease in interfering
effect of either single or multiple echoes.

In a single check test, it was found
that the "Comment Method" of rating
picture impairments gave the same re-
sults as the comparison method used in
this series of tests.

Discussion of Results

The approximate rule of addition of
several echoes, as deduced from the
data, is an empirical one with limited
applications. It applies to eight echoes,
or less, to echoes having individual
interfering effects differing by no more
than about 6 db, and to echo spacings

496

June 1952 Journal of the SMPTE Vol.58

-10

2 34

NUMBER OF ECHOES

Fig. 3. Advantage
vs. number and aver-
age spacing of echoes.
Reference echo 40 db
below main signal.

-10

2 3 ' 4

NUMBER OF ECHOES

Fig. 4. Advantage
vs. number and aver-
age spacing of echoes.
Reference echo 35 db
below main signal.

Fowler and Christopher: Multiple Echoes in Television

497

t -4

2 3456

NUMBER OF ECHOES

Fig. 5. Advantage
vs. number and aver-
age spacing of echoes.
Reference echo 30 db
below main signal.

MEASURED VALUES IN DECIBELS (TEST RESULTS)

/

0 2 ECHOES
D 4 ECHOES
A 6 ECHOES

A

A

/

/•

XA

a
6 —

A

A
A

^

y

A

-
/

/

:

IDEAI <

CORRELATION"".

^/

S

4>t •

^

Oj

L 2S

!•

7

H

f

1

> 0

/

D

A

' a

a

123456769 10 11
CALCULATED VALUES IN DECIBELS

Fig. 6. Correlation
of measured and cal-
culated values of rela-
tive interfering effect.
Combined data for 40,
35 and 30 db criteria.

498

June 1952 Journal of the SMPTE Vol.58

not excessively * different from the aver-
age of those spacings. Although a
general law of addition of echoes was
not discovered, it is not clear that it
would be very useful for present pur-
poses. Such a law would probably be
either very complicated, if applicable
to specific cases, or too general for
specific applications, if based on statis-
tical averages.

It is generally conceded that as the
number of echoes becomes large, "ad-
vantage" should tend to vanish, and the
effective power of the many echoes
should approach weighted power. (This
presupposes a random distribution of
amplitudes, polarities and delays, which
tend to make the total echo signal like
noise.) In this study of echo addition,

there was no noticeable trend toward
decreasing advantage as the number of
echoes was increased (up to eight echoes)
unless there was an accompanying de-
crease in average spacing.

In Fig. 6, showing the correlation of
the test results with computed values,
the data points are lumped together
for the three kinds of tests where the
reference echo was 40, 35 and 30 db
below the main signal. Although the
overall correlation is very satisfactory,
a still better correlation is obtained
for the 40-db data taken alone. The
correlations for the 35-db and 30-db
data are progressively worse. This
trend is probably related to the greater
difficulty observers had in comparing
multiple with single echoes as the
visibility of the echoes was increased.

APPENDIX I
Determination of Empirical Advantage Curves

Assume several echoes, identical in
every respect except amplitude and
delay. The peak-to-peak amplitudes
are taken as /i, 72, 73, etc., and are
expressed in terms of unit amplitudes of
a single reference echo. Suppose, for
the moment, the delays are all rather
large, so that the interfering effects of
the several echoes taken singly are about
the same. If all the delays are the same,
the echoes will fall precisely on one
another, and the effective amplitude,
7e, of the sum of the echoes will be,
simply:

7. - 7i + /•+...+

(1)

If the delays increase progressively
by about 1 jusec, the effective amplitude
appears (from the test results) to be
expressible as:

7,2 = tf + 722 + . . . + 7*2. (2)

This suggests a simple rule of addition
of the form:

* The standard deviation of the spacings
should not exceed the mean spacing.

S = 7ir + hr + . . . +

(3)

where the exponent, r, depends upon
the spacing (or delay difference) of
successive echoes. Although uniform
spacing is assumed, moderate departures
from uniformity can be tolerated.

In the tests, the physical amplitudes,
7], 72, etc., were equal, but in general
had different time-weightings. Let the
weighing of the nth echo be wn db, where
positive values of w n mean greater inter-
fering effect than an equal-amplitude
reference echo. The interfering effect
of N echoes of unit physical amplitude
will exceed that of the reference echo by
M db, where M is given implicitly by:

rM xr rwn

1020 = S 1020,
n-1

or explicitly by:

(\T TWn

£ m20
*•
n=l

l()

(4)

(5)

The special value of M when r = 2,
i.e., when power addition obtains, is

Fowler and Christopher: Multiple Echoes in Television

499

designated A/o- It represents weighted
echo power, expressed in db above the
power of reference echo, and is given by

A/o = 10 logu

Ufa

ior°

N

(6)

When the weightings are small — and
they were purposely made so by making
the delay of the reference echo about
equal to the mean delay of the JV
echoes — first-order approximations can
be used with fair accuracy in computing
(5) and (6). Approximately, then,

and

M

A/o = 10 log,0 JV

1 AT

2 wn.

(7)

(8)

The difference, A/0 — A/, is defined
as advantage and is designated by A,
where

A = l - 101og,0 A".

(9)

Note that when r = 2, A = 0 for power
addition; when r = 1 (zero spacing
between echoes), A = —10 logio A",
indicating current (or voltage) addition;
and as r — »• », A — »• +10 logio N,
indicating no addition effects, i.e., M
is determined by the single most effective
echo of the group.

The approximate expression, (9), for
advantage, depending only on spacing
(as r does) and number of echoes, N,
was used as the basis for processing the
data. The test data gave values of M
directly; A/o (relative weighted echo

power) was computed using the weight-
ing curve of Fig. 2; subtracting M
from A/o gives advantage, A. Dividing
A by 10 logio JV gives the factor (1 -
2/r). At this point the data are segre-
gated according to criterion, i.e., ref-
erence echo at 40, 35 and 30 db below
main signal. The factor (1 — 2/r) was
then plotted separately for each criterion
and, in all cases, against AT", the average
spacing of echoes. The relations, (1 —
2/r) vs. AT" could fairly accurately be
represented by smooth curves. By
choosing AT as parameter, and using
values of (1 - 2/r) taken from the
smooth curves, expression (9) can be
plotted in the form shown in Figs. 3,
4 and 5.

In each of the above three figures,
one of the advantage curves is indicated
for AT" = oo. These represent the
apparent limits of positive advantage
approached as A Tis increased somewhat
beyond 5 or 6 jusec. The limiting values
of advantage were derived from the smooth
curves of the factor, (1 — 2/r), which
appeared to have asymptotic values as
AT increased.

References

1. P. Mertz, A. D. Fowler and H. N.
Christopher, "Quality rating of tele-
vision images," Proc. IRE, 38: 1269-
1283, Nov. 1950.

2. A. D. Fowler, "Observer reaction to
low-frequency interference in television
pictures," Proc. IRE, 39: 1332-1336,
Oct. 1951.

3. A. D. Fowler, "Observer reaction to
video crosstalk," Jour. SMPTE, 57:
416-424, Nov. 1951.

500

June 1952 Journal of the SMPTE Vol. 58

ABSTRACT

The Image Iconoscope —

a Camera Tube for Television

By P. SCHAGEN, H. BRUINING and J. C. FRANCKEN

The oldest television camera tube, the iconoscope, is now used only for trans-
mitting still pictures (e.g. the signal picture of a certain station) and film
pictures. Further development of camera tubes in Europe has followed a
course different from that in America. In the U.S.A. the image orthicon has
become predominant, whilst in Europe the image iconoscope is widely used.
Of the latter there are British and French versions and also one that has been
developed in the Philips Laboratory at Eindhoven (Netherlands). This
Philips image iconoscope is described here and compared with other camera
tubes.

JL HE OBJECT of television is to transmit
moving pictures via electrical means.
This is achieved by "measuring" in
succession the brightness of the very
large number of picture elements into
which the picture to be transmitted is
imagined as being divided. This meas-
uring consists in the conversion of the
brightnesses into corresponding fluctua-
tions of an electric current which in
some way or other govern the signal
transmitted.

However the time available for meas-
uring the brightness of one picture
element is very small, actually only
10~7 sec. A method can be imagined,
whereby the imaged scene is illuminated
continuously on a photosensitive plate,
while for each picture element in
succession in the space of time of 10~7

Abstract by Pierre Mertz of a paper in
Philips Technical Review, 13: 119-133,
Nov. 1951.

sec a signal is transmitted which corre-
sponds to an illumination that was present
during the whole of the time (^ sec)
available per picture. This idea is to
be found materialized in all present-day
television camera tubes. With this
method there is a continuous accumu-
lation of charge during a frame period,
and thus these tubes have come to be
known as "storage tubes."

The oldest form of storage tube is the
iconoscope, designed by Zworykin
(1933). In the main this article will
be devoted to a modern camera tube
named the image iconoscope. Some
other types will be mentioned in passing.

Classification of Modern Camera Tubes

In the most important camera tubes
of modern design there is a plate
("target" or "mosaic") on which is
projected an electrical image consisting
of a two-dimensional pattern of electric

June 1952 Journal of the SMPTE Vol. 58

501

V2 eV

Fig. A. Secondary-emission coefficient 5 of an insulator, as a function of the energy
Fpr of the primary electrons. At two values of Fpr (Fi and 1 2) 5 is equal to 1. This
curve applies when the collector potential is high compared with F2.

potentials corresponding in amplitude
and position to the luminance in the
optical image of the scene to be trans-
mitted. This electrical image is scanned
point by point by a focused beam of
electrons (the scanning beam), the po-
tentials being thereby reduced to a
certain "stabilizing potential" which in
some way or other produces an electric
signal.

The target is, of course, made of an
insulating material, e.g. mica. When
an electron beam is directed upon it the
rule is that for every surface element,
in the stable state, on an average just as
many electrons have to be emitted as
impinge upon it.

When the primary electrons impinge
upon a surface element of the plate they
release secondary electrons from the
material. The secondary-emission coeffi-
cient 6, i.e. the average number of
secondary electrons released by one
primary electron, depends upon the
material and the velocity (thus the
energy) of the primary electrons at
the plate. If VcM is so high that the
collector attracts all the secondary
electrons towards it, then the variation
of 6 as a function of the energy Fpr (ex-
pressed in electron-volts) of the primary
electrons is as represented in Fig. A.
In the case of most materials there are
two values for Fpr where 5 = 1; the

smaller of the two is denoted by V\, the
larger by F2. For mica, for instance,
these material constants are in the
order of 10 volts and some thousands of
volts, respectively.

Upon reducing Fcou, the potential of
the surface will be stabilized at a value
F3, where the current intensity of the
secondary electrons actually reaching
the collector (z'coii) is equal to the current
intensity ipr of the primary beam. As
a rule F3 is slightly higher than Fcoii
(Fig. B); in contrast with V\ and F2,
F3 is therefore not a material constant.

Thus, when bombarded with slow
electrons (Fpr < FI) the surface potential
becomes stabilized at zero, and when
bombarded with electrons of high
velocity it becomes stabilized at the
value F2 (provided Fcon > F2) or at
V$ (wFcoii < F2). For the target of a
camera tube however no use is made of
the value F2, for practical reasons; it is
strongly influenced by the condition of
the surface and thus is too variable from
point to point.

It is according to these possibilities
that camera tubes are classified as:

(1) low velocity tubes, where the
target is stabilized at cathode potential,
and

(2) high velocity tubes, where the
target is stabilized at the potential
V* « Fcoii (e.g. 1000 v).

502

June 1952 Journal of the SMPTE Vol. 58

Fig. B. Effective secondary-emission coefficient 5eff of an insulator, as a function of the
energy Vpr of the primary electrons when the collector potential VCQ\\ is smaller than
V2 (cf. Fig. A). 5eff = 1 at Vpr = V\ and at Kpr = V3, the latter value being a few volts
higher than Fcoll. To the left of F0 (slightly lower than Vco\\} the curve is identical to
that in Fig. A.

Among the first belongs the image
orthicon, which is the type of tube
mainly used in the U.S.A., while be-
longing to the second group are the
iconoscope and the image iconoscope,
the latter often being given preference
in European television circles. One of
the reasons for this preference is related
to the large number of lines adopted on
the West-European continent (625, and
in France 819): with a high electron
velocity it is easier to satisfy the high
requirements for the focusing of the
scanning beam which are demanded for
the definition required for such a large
number of lines.

The Iconoscope

The iconoscope is the camera tube
which in its time gave such an impetus
to television.1 It is schematically rep-
resented in Fig. C, while in Fig. D a
photograph is given of the Philips
iconoscope, type 5852.

A lens (objective) projects an image
of the scene onto a target of thin mica
coated on the front with a mosaic of
minute, mutually insulated, photosen-
sitive elements. On the reverse side is

1936.

e.g., Philips Tech. Rev., 1: 18-19,

a coating of metal, called the signal
plate, forming the output electrode
and externally connected to earth via a
resistor. A ring-shaped coating of metal
on the inside of the envelope serves as
collector and is connected to earth
direct.

The action of the iconoscope is some-
times explained in the following (in-
adequate) way. The incident light
causes the photoelectric elements of the
mosaic to emit photoelectrons, which are
taken up by the collector. Thus a
positive electrical image is formed on
the mosaic. The photoelectric elements
together with the target form as many
minute capacitors. As the scanning
beam moves across the mosaic the group
of capacitors belonging to a certain
picture element are discharged.
Through the resistor via which the
signal plate is earthed there then flows
a small current corresponding in intensity
to the charge of the picture element,
thus corresponding to the local luminance
of the optical picture. Thus in the
scanning of the electrical image a series
of current impulses are generated which
together form the video current.

Actually the position is not so simple
as this. Such a description does not
take into account the part played by

Schagen, Bruining and Franckcn: Image Iconoscope (Abstract)

503

-1000V

Fig. C. Iconoscope. L, a lens projecting the scene on the mosaic M of the target T.
SP, signal plate ; /?„, load resistor ; C, collector ; K, cathode ; Z>, deflection coils ; E, scanning
beam. The (electrostatic) focusing is not shown.

Fig. D. The Philips Iconoscope, Type 5852.

secondary emission.2 Not all the sec-
ondary electrons reach the collector,
firstly because the potential of a bom-
barded surface element is higher than
FCoii. The secondary electrons which
do not reach the collector fall back on
other parts of the mosaic. This dis-

2 V. K. Zworykin, G. A. Morton and L. F.
Flory, Proc. I.R.E., 25: 1071-1092, 1937;
and W. Heimann and K. Wemheuer, Z.
tech. Phys., 19: 451-454, 1938.

tribution of the secondary electrons is
called the redistribution effect, and it
is of essential importance for the action
of the iconoscope.

After the surface element in question
has been scanned, it will continue to
receive secondary electrons originating
from other surface elements, until it is
scanned by the beam again. Thus its
potential V begins to drop (Fig. E).

During a considerable part of the

504

June 1952 Journal of the SMPTE Vol. 58

Fig. E. Curve of the potential V of a picture element on the mosaic of an iconoscope,
as a function of the time t. Fully drawn line: mosaic not illuminated; broken line:
mosaic illuminated. T0 = scanning period for the whole image (0.04 sec), T\ = scanning
time for one picture element (10~7 sec; in the drawing highly exaggerated). For the
meaning of Vco\\ and V-A see Fig. B, and for F0, F0', and VQ" see the text.

scanning period the potential V of the
element is higher than Fcoii, and the
photoelectrons do not possess sufficient
energy to overcome this potential differ-
ence. Photoemission begins, therefore,
when — owing to the redistribution
effect — the potential V has been
sufficiently reduced.

The most important features of the
iconoscope will now be briefly discussed.

As already explained, it is due to the
redistribution effect that photoemission
can take place, but this is only possible
during a fraction of a scanning period.
Thus we are still far removed from a
continuous photoemission such as was
imagined in the case of an ideal storage
tube! This is one of the reasons for
the iconoscope's rather low sensitivity.

A second cause of the lack of sensitivity
lies in the mosaic form of the light-
sensitive layer. The insulation between
the elements does not contribute towards
photoemission, so that a considerable
part of the surface of the target is photo-
electrically inactive.

The main cause of spurious signals
(see the literature quoted in footnote 2)
is that the redistribution does not take
place in the same way all over the

mosaic, owing to the surroundings of
the elements not being the same every-
where.

When the iconoscope is illuminated
the spurious signal is superposed on the
picture signal and only if the latter is
of a reasonable strength is the spurious
signal not very disturbing. It is for
this reason that with the iconoscope very-
high intensities of illumination are
needed.

The stronger the illumination on a
certain part of the mosaic, the higher
is the potential VQ" at that spot just
before it is scanned by the beam. This
has two consequences: there is slightly
less chance of further photoelectrons
escaping, and there is a somewhat greater
attraction of redistributed secondary
electrons. Both these effects result in
the amplitude of the signal increasing
less than proportionately with the
illumination. This nonlinearity is rather
an advantage than a disadvantage in
that it compensates fairly well an inverse
nonlinearity between the beam current
and the control voltage in the picture
tube of the receiver. Thus there is no
need to take steps to compensate the
latter nonlinear effect.

Schagen, Bruining and Francken: Image Iconoscope (Abstract)

505

Fig. F. Image iconoscope. P, photocathode ; S, coil of the magnetic
electron lens; 7 and 2, paths of photoelectrons ; FOC, focusing coil.
Other letters have the same meaning as in Fig. G.

506

Fig. G. The Philips Image Iconoscope, Type 5854.
June 1952 Journal of the SMPTE Vol. 58

The Image Iconoscope

The greatest disadvantage of the
iconoscope is its lack of sensitivity, and
it is for that reason that attempts have
been made to develop camera tubes with
greater sensitivity, while still retaining
the good picture quality obtained with
the iconoscope when the scene is suffi-
ciently illuminated.

A year or two prior to 1940 a more
sensitive version of the iconoscope,
called the image iconoscope, was de-
veloped in the U.K. and in the U.S.A.3
Some improvements on this have since
been made in the Philips Laboratory
at Eindhoven, as will appear in the
course of this article.

In the case of the image iconoscope
(Fig. F) a lens (objective) projects an
optical image of the scene to be televised
onto a continuous, transparent photo-
cathode. The local density of emission
of the photoelectrons corresponds to
the local luminance of the optical image.
This photoemission image is focused
by an electron lens onto a target con-
sisting in this case of a thin layer of
insulating material applied to the signal
plate. The metallized inner wall of
the envelope serves as collector. An
electron gun mounted in an arm of the
envelope supplies the beam of electrons
scanning the target.

The differences, compared with the
conventional iconoscope, which are
mainly responsible for the gain in
sensitivity, are the following:

(1) The surface of the photocathode is
continuous, so that none of its effective
area is lost in insulation between the
separate photoelectric elements.

(2) The stream of photoelectrons
reaching the target is reinforced by
secondary emission, each photoelectron
releasing on an average more than two
secondary electrons.

(3) The secondary electrons released

sSee, e.g., H. lams, G. A. Morton and
V. K. Zworykin, "The image iconoscope,"
Proc. I.R.E., 27: 541-547, 1939.

from the target by the photoelectrons
have a much greater energy than the
photoelectrons in the ordinary icono-
scope, so that secondary emission from a
surface element begins immediately
after that element has been stabilized
by the scanning beam. This means a
considerable gain in storage action.

Let us now consider more closely the
principal parts of the image iconoscope
and also the important question of
electron-optical projection. The Philips
type of image iconoscope is illustrated
in Fig. G.

Contrary to ordinary photoelectric
cells, an image iconoscope must have a
photocathode which is semi transparent,
because the light enters from the outside
while the photoelectrons have to emerge
on the inside.

The requirements greatly restrict the
choice of photoelectric material to be
used. The photocathode in the Philips
image iconoscope consists of a very thin
coating of cesium, antimony and oxygen
applied to a flat part of the glass enve-
lope. The sensitivity for the light from
an incandescent lamp with color tem-
perature 2600 K is about 45 jua per
lumen. The spectral sensitivity curve,
compared with the relative luminosity
curve for the normal eye, is slightly
displaced towards the blue (Fig. H).

The optical image of the scene is
converted into a corresponding photo-
emission image on the photocathode.
The next step is to produce on the target
an electrical image which is a faithful
replica of the photoemission image.
This requires that the small beams of
photoelectrons emitted from points of
the photocathode are focused on corre-
sponding points on the target. For this
electron-optical image formation an
electron lens is needed.

An electric field has to be employed.
This is obtained by means of a metal
cylinder (e.g. the metal coating A on
the inner wall of a glass tube, Fig. I)
facing the photocathode P and applying
a potential difference of, say, 1000 v

Schagen, Bruining and Franckeh: Image Iconoscope (Abstract) 507

4000

5000

6000

7000

8000A

Fig. H. Relative spectral sensitivity of the Type 5854 image iconoscope

(curve /), compared with the relative luminosity (curve //),

as functions of the wavelength X of the light.

Fig. I. Formation of the electron-optical image of the photocathode P on the target
T with the aid of an electric field (between P and the cylinder A} and a magnetic field.
The latter (lines of flux density B} is produced by a focusing coil S. Dimensions are in
millimeters.

between these electrodes. Since the
cylinder forms, electrically, one whole
with the earthed collector, the photo-
cathode is given a potential of — 1 000 v
with respect to earth.

This electric field alone, however,
does not suffice; a magnetic field has
to be added which focuses each electron
pencil. Such a field can be produced
by means of a coil placed concentrically

around the tube. The coil has to be of
such dimensions and in such a position
as to minimize aberrations, whilst the
magnetic field must not disturb the
movement of the scanning beam.

The movement of the electrons de-
pends not only upon the two fields
mentioned but also upon the velocities
of the electrons leaving the photo-
cathode. Some of them have zero initial

508

June 1952 Journal of the SMPTE Vol. 58

velocity, and the paths they follow are
called the principal rays. Generally,
however, the electrons leave the cathode
with a certain velocity, with the result
that they follow a more complex path.

Briefly, the course of a principal ray
is as follows: at first the path is ap-
proximately parallel to the axis^ of the
tube (the £ axis), then it diverges
farther and farther from that axis,
turning about the £ axis first clockwise
and later counterclockwise in the form
of a widening helix.

Although most of the electrons which
leave the photocathode have velocities
greater than zero and thus do not follow
any principal paths, still it is the prin-
cipal rays which determine the geometry
of the electron-optical image. Each
forms the axis of a small electron pencil.

The axial component of initial velocity
gives rise to a certain "chromatic"
aberration: a point of the photocathode
from which electrons emerge with axial
velocity does not result in a point being
formed on the target but a small circle
(scattering circle), the diameter of which
is so small — thus the image so sharp —
that the image iconoscope can quite
well be worked with more than 600
scanning lines. In the image orthicon,
on the other hand, the electric field at
the cathode is ten times smaller,4 so
that with this type of tube the formation
of the electron-optical image is a limiting
factor for the number of lines.

Owing to the predominance of the
diverging forces acting upon the elec-
trons following the principal path the
image on the target is magnified, and
owing to the tangential forces the
electron image is rotated with respect to
the optical image on the photocathode,
the angle of rotation being about 30 to
40°.

With our image iconoscope the magni-
fication is normally 3.75, which means
to say that the scanned part of the

H. B. De Vore, Proc. I.R.E., 36: 335-345,
1948.

Fig. J. A BCD is an image on the photo-
cathode, A'B'C'D', the corresponding
electrical image on the target. The latter
is magnified and turned with respect to
A BCD and also shows some S distortion,
which always occurs when magnetic
lenses are used (straight lines are projected
with a slightly S-shaped curve). If the
magnification is too small the S distortion
becomes so pronounced that it can no
longer be sufficiently corrected.

target, which always covers an area of
45 mm X 60 mm, corresponds to an
area of 12 mm X 16 mm on the photo-
cathode (the diameter of the active part
of the photocathode is 20 mm). By
exchanging the coil for another of
different dimensions it is also possible,
however, to work with a larger or a
smaller magnification, thus projecting
a smaller or a larger part of the photo-
cathode on the target. The choice as
regards the size of the effective photo-
cathode is governed by requirements of
an optical, light-technical and camera-
technical nature. The limits for the
magnification are 2.75 and 7.5 (diameter
of the projected part of the cathode,
respectively, 27 mm and 10 mm).

With a magnification greater than
7 to 8, owing to the "chromatic" aberra-
tion of the photoelectrons emerging
with axial velocity (see above) there is
too great a loss in resolving power.

Schagen, Bruining and Francken: Image Iconoscope (Abstract)

509

Fig. K(l). Picture showing a marked field curvature, pin-cushion distortion and S
distortion. In Fig. K(2), there is only a slight S distortion, which can easily be corrected
electrically.

These photographs have been taken with the aid of an experimental tube in which a
fluorescent screen was used instead of a target. On the photocathode a test pattern was
projected, as used in television, for detecting aberrations and checking the definition
and gradation. The heavy black circle and the thick horizontal line in the middle corre-
spond to markings on the photocathode for determining the magnification.

The lower limit of 2.75 is due to various
other aberrations, which with a smaller
magnification can no longer be suffi-
ciently compensated. As such may be
distinguished: field curvature, pin-
cushion distortion and so-called S dis-
tortion. The first two are known from
light-optics.6 By S distortion is meant
the effect of the image of a straight line
being projected as a line curved some-
what in the shape of the letter S (Fig. J).
If the magnification is not too small the
S distortion can be sufficiently corrected
by electrical means (which we cannot
enter into here), but if it is less than 2.75

6 A review of various optical aberrations is
to be found, for instance, in : W. de Groot,
Philips Tech. Rev., 9: 301-308, 1947, in
particular pp. 304 and 306.

this is no longer possible. In Fig. K(l)
a picture is given showing all three
aberrations to a marked extent. The
picture in Fig. K(2), however, has only
a scarcely perceptible S distortion, which
is not troublesome.

The electron gun supplies the scanning
beam. Just as is the case with most
picture tubes, in the image iconoscope
the beam is focused and deflected with
the aid of magnetic fields.

In regard to the sharpness of the
scanning, there are two things to be
considered. The non-deflected beam
is focused on the center of the target,
where its diameter must be so small that
the lines do not overlap when being
scanned. If it is desired to work for
instance with 1000 lines then, if the
height of the scanned part of the target

510

June 1952 Journal of the SMPTE Vol. 58

Figure K(2).

is 45 mm, the effective diameter of the
focus must not be more than 45ju. This
requirement is all the better fulfilled
the higher the acceleration voltage is
chosen, but this should preferably not
exceed 1000 v.

Further, account has to be taken of
the fact that in the image iconoscope
the electron gun has to be mounted
with its axis at an angle to the target.
Consequently when the beam is deflected
upward or downward the focus is no
longer situated on the target. There-
fore, to obtain sufficiently sharp scanning
also away from the center, the beam
must have a good depth of focus, which
means that it has to be extremely narrow.
Hence the angle of divergence 2at (see
Fig. L) has to be kept very small.

It is, in general, difficult to obtain a
fine focus with a very narrow beam on
account of the mutual repulsion of the
electrons, but fortunately the intensity
of the beam current required is very
low, in the order of 0.1 na.

Fig. L. Assuming that the nondeflected
beam E0 has been focused onto the center
of the target 7", when the beam is de-
flected the focus will no longer be in the
plane of T. This gives rise to blurring,
the extent of which increases with the angle
of divergence 2af.

Schagen, Bruining and Francken: Image Iconoscope (Abstract)

511

In addition to this space-charge re-
pulsion there is another factor limiting
the spot size obtained with a very
narrow beam: there is a very funda-
mental relationship between the angle
of divergence 2aj and the current density
in the beam. In the case where the
space charge is negligible this relation-
ship is:

sin2af = p • J4-, . . . .(1)

where VQ = ^mv^/e (with m = mass,
VQ = initial velocity and e = charge of
an electron), V = the potential difference
traversed by the electrons, jt = density
of the beam current in the focus, and
*'o that at the cathode of the gun.

What has to be found is an optimum
value for a{ at which, on the one hand,
the focus- is not too large and, on the
other hand, the sharpness at the edges
of the image does not differ too much
from that in the middle. With our
image iconoscope the position is such
that this optimum value of ai lies at
about 3 X 10~3 radians.

This small angle of divergence, com-
bined with a low beam current intensity
(about 0.2 /*a), has been obtained by
placing two diaphragms in the beam.
The first, with a narrow aperture,
confines the beam within the desired
small angle. The second one, with a
wider aperture, allows the beam to pass
through without hindrance but inter-
cepts the low-velocity secondary electrons
formed round the edge of the first dia-
phragm.

With the focus of 45 M already men-
tioned and a beam current of 0.2 /xa,
the average current density in the focus
isjt = 12 ma/sq cm. Substituting this
in Eq. (1), and for V the value 1000 v,
and for VQ the value corresponding to
the average initial velocity (« 0.1 v),
we find for the average current density
at the cathode of the gun jo « 120
ma/sq cm. The peak value of the
current density is in fact several times
greater. Although an ordinary oxide-

coated cathode may indeed be con-
tinuously loaded with such a current
density, it is better to use what is known
as an L cathode,6 since this has a much
longer life. It would be quite un-
desirable if the useful life of a costly tube
such as the image iconoscope were to
be dependent upon the life of a com-
ponent like the cathode of the gun.

The glass arm of the envelope con-
taining the electron gun has been kept
as narrow as possible (internal diameter
11 mm, external 14 mm), so that also
the focusing coil and the deflection coils
may be small. A camera with an image
iconoscope is shown in Fig. M.

In practical use the resolving power
of the Philips image iconoscope is found
to be 900 to 1000 lines in the middle of
the image and about 700 lines at the
edges. (These limits are set by the
electron gun; the resolving power of
the electron-optical projection is very
much greater.)

An improvement has been reached by
coating the mica target with a thin
layer of MgO, which leads to a con-
siderable gain in secondary emission and
hence sensitivity. Furthermore, owing
to the coating of MgO, stains on the
mica which cannot be removed and
otherwise show up clearly in the picture
are thereby made invisible.

The capacitance of a surface element
of the target with respect to the signal
plate is an important factor, and an
increase of this capacitance must lead
to greater sensitivity.

New tubes were therefore made with
a mica sheet of only about 25 M thickness
(also with a layer of MgO, thin compared
with the mica) instead of the original
sheet thickness of 50 M.

The reproduced picture of a scene
televised under the normal studio light-
ing, or of an outdoor scene in daylight

6 H. J. Lemmens, M. J. Jansen and R.
Loosjes, "A new thermionic cathode for
heavy loads," Philips Tech. Rev., 11: 341-
350, 1950.

512

June 1952 Journal of the SMPTE Vol. 58

Fig. M. One of the cameras used for the experimental television broadcasts at Eind-
hoven. One side panel and a screen have been removed. /, image iconoscope Type
5854; 5", image coil; FOC, focusing coil; D, deflection coils; G, time-base generator;
V, chassis with monitor picture tube and accessories; M, microphone and T, telephone
for communication between the operator and the control room; 1C, knob for exchanging
the objective; P, playbook.

(even in bad weather), with the image
iconoscope last described, is almost free
of "noise" and shows excellent gradation.
In the image iconoscope spurious
.signals arise from the same cause as in
the case of the conventional iconoscope:
the various surface elements of the target
are not all in the same position with
respect to the scanning beam. In the
image iconoscope, however, the situation
is more favorable: with the tube de-
scribed (mica 25 M thick, beam current
0.2 /xa) and with an illumination pro-
ducing a photocurrent of more than
0.1 /ia, the spurious signals are so weak
that there is hardly any need of com-
pensating measures. In practice a
photocurrent of 0.1 pa can be obtained
with an illumination of the scene of
about 1000 lux, when using a non-

diaphragmed, normal objective with
aperture f/2.

Comparison of Different
Types of Camera Tubes

Let us now compare, briefly, the two
main types of camera tubes, the high-
velocity and the low-velocity types.

In the first place there is the question
of sensitivity. This res.olves itself into
two factors (disregarding the efficiency
of the optical system), viz. the sensitivity
of the photocathode (photocurrent 7Ph
in relation to the light flux falling on the
cathode) and the sensitivity of the
scanning mechanism (ratio of signal
current 7S to photocurrent /ph).

As regards the sensitivity of the photo-
cathode of the two high-velocity tubes —
the conventional iconoscope and the

Schagen, Bruining and Francken: Image Iconoscope (Abstract)

513

image iconoscope — the latter has very
much the advantage, owing to the con-
tinuity of the photocathode. Among the
low-velocity tubes there are likewise
types with a mosaic cathode and others
with a continuous cathode, the latter
including the image orthicon, which as
regards photocathode sensitivity is equal
to the image iconoscope.

The scanning sensitivity of low-
velocity tubes is simply 1 jua signal
current per »a photocurrent. In high-
velocity tubes the phenomenon of re-
distribution complicates matters, but
the scanning sensitivity of the ordinary
iconoscope can he put at -fa jua/jua
and that of the image iconoscope, at
about 1 Ata/Va.

Although, therefore, the image icono-
scope has about the same scanning sensi-
tivity as the simple low-velocity tube, the
78 = f(/Ph) curve is not linear, whereas
in the case of low-velocity tubes it is
linear; the nonlinear curve is favorable,
as explained when dealing with the
iconoscope.

There is a means, however, of appre-
ciably increasing the scanning sensitivity
of low-velocity tubes. The electrons
from the scanning beam which are not
taken up by the target and return to the
gun can be collected in a multiplier,

placed around the gun, which works
with secondary emission and thus mul-
tiplies them. This is what takes place
in the image orthicon, commonly em-
ployed in the U.S.A. In this way the
scanning sensitivity may be raised to a
value of 25 to 100 Ma/Ma, which is of
course valuable when scenes have to
be televised in poor light. However, the
current of the returning beam can be
modulated only up to about 20% and
consequently contains a relatively large
amount of noise.

It has already been explained that in
regard to spurious signals the image
iconoscope has a decided advantage over
the ordinary iconoscope. The image
orthicon is free of spurious signals of
this nature, but on the other hand it is
subject to another interference connected
with the fact that the secondary-emitting
surfaces of the multiplier do not have
exactly the same secondary-emission
coefficient over the whole area ("dynode
spots").

Electron-optically, high-velocity tubes
have undeniably the advantage over
those of the other group, in that with
electrons of a high velocity it is easier
to obtain a scanning beam with a high
resolving power, and there is much less
trouble from interfering electric and
magnetic fields.

514

June 1952 Journal of the SMPTE Vol. 58

TelePrompter —
New Production Tool

By FRED BARTON and H. J. SCHLAFLY

The TelePrompter is a device now being used extensively in motion picture
and television productions, and by public speakers as an aid in delivery of a
prepared script. It is a production tool of great flexibility; its technical
features and applications are described.

JL HE PROBLEM of presenting entertain-
ment to the public has been accen-
tuated by the very nature of the tele-
vision industry because of its continual
requirement for new material. Whether
this material is "live," that is, performed
directly in front of television cameras or
filmed before it is translated into a
television signal, the continual demand
for new material of an expected quality
presents many pressing problems. One
of these problems is the fundamental
necessity of memorization. Added to
the tension that normally accompanies
the production of a television or motion
picture presentation is the very real and
continuing chore of memorization of
lines for the actor, and the constant threat
of fluffs, delays and retakes for the pro-
ducer. A professional performer ac-
cepts, as part of his vocation, the neces-
sity of studying his lines. Such a
professional can substantially memorize
new material with comparatively few

Presented on April 21, 1952, at the So-
ciety's Convention at Chicago, by Fred
Barton and H. J. Schlafly, TelePrompter
Corp., 270 Park Ave., New York 17, N. Y.

readings, but in order to reach the point
of perfection, the point which makes
the difference between a smooth and a
ragged performance, this same per-
former may spend many tedious hours
of study. Those who are not normally
engaged in the entertainment or public-
speaking professions find this problem
of memory so much the more difficult.
The necessity for accurate memorization
breeds a second evil which may be even
more devastating to a good performance
than the mere fault of forgetting the line.
This second evil is the fear of forgetting
the lines and the resulting tension,
tightness and unnaturalness uncon-
sciously generated by such fear.

The TelePrompter is a modern ap-
proach to the old problem of prompting.
It is a new production tool whose
intelligent use can save hours of rehearsal,
help to promote a smooth and relaxed
performance, and reduce film studio
retakes. Not only does its use result in
the 100% perfect script without hours
of laborious work and mental exercise,
but it serves also as a guardian against
fear. Paradoxically its mere presence

June 1952 Journal of the SMPTE Vol. 58

515

Fig. 1. A group of four TelePrompter reader units illustrating various
methods of mounting for studio use.

in a studio greatly reduces the probability
that prompting of any nature will be
required.

Prompting devices are not new —
many methods have been tried through-
out the history of entertainment and
public speaking. The most familiar of
these methods, as far as the public is
concerned, is typified by the old prompt-
er's box. Unfortunately, the prompt-
er's hoarse whisper was often as audible
to the audience as was the actor's
uncomfortable predicament. Even so,

the prompter was one of the higher-paid
members of the cast. Projection devices
of one sort or another have been tried
repeatedly, but because of their bulk
and the difficulty of making changes
they are not in common use. The
electronic age contributed a tiny radio
receiver with a speaker hidden in the
ear of the performer. This method
removed the audibility of the cue, but
more often than not it distracted the
performer with unwanted promptings
and confusing instructions, and it bur-

516

June 1952 Journal of the SMPTE Vol. 58

Fig. 2. Master control, operator and monitor TelePrompter.
The operator's hand is on the throttle-type speed control.

dened him with equipment that must be
hidden on his person.

A careful and lengthy study of the
problem pointed out the requirements
for a prompting device which would have
the features and flexibility necessary to
contribute to modern production tech-
niques. This study revealed that, as
a primary requisite, a prompting device
must be always ready and available for
the performer's use if he desired to refer
to it without betrayal of that fact to the
audience, but it must be a device which
could be completely ignored by the
performer if he did not need a reference.
This basic demand governed the choice

of the fundamental TelePrompter de-
sign, a design which has successfully
withstood the trial by fire of studio use,
with only an evolution of details to meet
the great variety of individual studio
situations that have been met. The
TelePrompter is in effect a multiple
reader. A number of reading units,
usually four for normal studio work as
shown in Fig. 1, which are small enough
and light enough to be strategically
positioned about a set or moved with
the action during the shooting, contain
the script and acting and production
cues. A master control unit, seen in

Barton and Schlafly: TelePrompter

517

Fig. 2, comprises the fifth item of
equipment in an operational group.

The script is written in large clear
black letters on a yellow, glare-free,
noncrinkle paper especially developed
for this application. The type is front
illuminated by self-contained in-
candescent lights which can be varied
in intensity. Front illumination was
chosen because it preserves contrast
ratio and because it is helped rather
than hurt by other studio illumination.
Extensive readability tests have been
conducted to determine the most favor-
able combination of the variables affect-
ing vision. The large type is readable
without effort by persons having normal
or corrected normal vision at 25 ft, a
distance that has been found to be quite
adequate for the great percentage of
studio shots. For an occasional cue
many performers have found that
distances considerably in excess of this
figure are quite satisfactory. The use
of low-power enlarging lenses in front
of the script, while possible, is dis-
couraged, except on rare occasions.
Not only do such lenses add to the
bulk and weight of the unit, but they pick
up surface reflections, introduce flare
and geometric distortions, and so limit
the viewing angle that readability gener-
ally decreases in spite of the apparent
increase in letter size. In those cases
where letters larger than the standard
Videotype size are required, it is pre-
ferred that larger original type be used.
Such type is now in the process of being
prepared for the electric "Videoprinter"
which has been developed for Tele-
Prompter Corporation.

Eight or more lines of script, depend-
ing on spacing, are visible to the actor
at any one time with the "hot" line,
that is the line currently being spoken,
indicated by a large red arrow. Thus,
the performer is at liberty te precue
himself and is not limited to the "hot"
line or word only. The script in each
reader is printed on a continuous strip
of paper which normally passes over the

flat reading surface from a supply roller
at the bottom to a take-up roller at the
top of the machine. In special cases
this direction of motion and the direction
of script continuity can be reversed.

The use of multiple units is basic to
this prompting technique because it
frees the performer from the necessity of
referring to a particular location on the
set. He is at liberty to allow his glance
to fall here or there or wherever the
director has determined during rehearsal
would be the most natural direction for
such a glance. Even when the per-
former is playing to the camera lens,
one of the TelePrompter units can be
positioned so close to the lens, as shown
in Fig. 3, that the slight angular differ-
ence in his glance is not significant. In
fact, it is convenient to position the
"hot" line immediately adjacent to the
taking lens itself. The use of multiple
units, however, imposes an operational
requirement for line-by-line synchronism
of the script in the several units so that
quick reference may be made from one
to the other without losing the place
(see Fig. 4).

Several means of synchronizing motion
of the script in the individual machines
were considered. The most obvious
method perhaps would be a pin and
sprocket hole arrangement with friction
drive of the take-up roll similar to the
motion picture projector. This method
was discarded for several reasons: it is
not suitable for operating over the great
range of speeds that are required; the
pins demand precise location of the paper
on the rollers and precise alignment of
inserts and changes; stretch and shrink-
age of paper as a function of weather
make loading difficult; full drive power
must be transmitted to the paper by
small-diameter pins in a few sprocket
holes; and misalignment of the paper or
rollers can cause noise and even tearing
of the paper. In the TelePrompter the
take-up roller is driven by direct gearing
to the motor pinion. Tension is applied
to the paper by friction braking of the

518

June 1952 Journal of the SMPTE Vol. 58

supply roller. For reversing, the motor
pinion is caused by remote control to
disengage the take-up gear train and to
engage the gear train of the supply
roller. In either forward or reverse,
one-way clutches remove friction from
the drive roller and apply it to the
feeding roller.

Torque for each machine is provided
by a small-size selsyn motor connected
electrically to a large self-synchronous
generator in the master control unit.
The generator is motor driven through
a mechanical speed changer which is

Fig. 3. The small Model 3 TelePromp-
ter reader with camera mount attach-
ment on the friction head can position
the "hot" line immediately adjacent to
the camera lens.

Fig. 4. Dramatic shows such as the First Hundred Tears make
use of the TelePrompter.

Barton and Schlafly: TelePrompter

519

continuously variable from zero to
maximum speed. Various simplifica-
tions of this drive system are, of course,
available at the cost of certain opera-
tional features. This drive system was
chosen not only because of the syn-
chronous rotation of all rotor shafts in
the system, but because the motors
operate well over a great range of speed
and are quiet, both mechanically and
electrically.

While the self-synchronous system
provides the basic unit synchronism, it
does not protect against insertion errors
or cumulative errors of line spacing in
the printing or slight differences in
paper tension. A further precaution
was added not only to guarantee the
line-for-line synchronism required, but
to provide an additional degree of
flexibility of individual unit control.
This feature consists of small conductive
strips placed at short intervals along
the script. Normally these so-called
synchronizing marks are placed in
identical positions on the several script
copies. As the paper travels past the
reading aperture a pair of contacts
engages these marks making a temporary
electrical circuit. Should one machine
reach a particular synchronizing mark
on its script before that point has been
reached in the other machines, a relay
circuit operates causing that machine to
stop. When all machines have reached
that identical point in the script, a reset
relay operates and they continue re-
synchronized. Normally synchroniza-
tion marks are located at periodic inter-
vals of about four feet.

Mention has been made of the "great
range of speed" required in operation
of the machine. Possibly some explana-
tion is required on this point. An
operator, located at the master unit, has
control not only of the TelePrompter
lighting, cue lights, synchronization and
mode of operation, but also of the speed
of the script motion. This operator is,
if you will, the accompanist. He
regulates the speed of the script so that

the "hot" line is always on the red
arrow indicator of his monitor unit.
If there is no dialogue at a particular
moment or if the actor chooses to ad-lib,
script speed may be zero. With normal
monologue the speed may be very slow.
A lively dialogue may require a speed
several times normal, and a program
cut or change in continuity (even that
happens occasionally while "on the air")
may require a speed many times normal.
Rehearsals, of course, place a maximum
demand on speed variations since the
script must be brought forward or
reverse to any particular spot for a
run-through in less time than the other
studio preparations can be made for
that run-through.

An important feature of a modern
prompting device is speed not only in
preparation of the script, but speed in
making changes. TelePrompter scripts
are prepared by an electric "Video-
print" typewriter with paper supplied
in fanfold packs for an original and
three carbon copies. The carbons have
been chosen by laboratory tests for good
contrast and the copies are in fact hardly
discernible from the original. Although
the paper and type size of the "Video-
printer" is large, it has a standard type-
writer keyboard and is operated easily
and conveniently by any typist.

Small script changes can be made
simply by an ink-brush write-in either
directly on the paper or on a cloth
adhesive tape placed over the changed
word or line. Larger changes are made
by removing or inserting complete
panels in the paper. A panel consists
of one fold of the original paper, an
eight-inch length for the standard Tele-
Prompter. The paper is supplied by the
manufacturer with accurate horizontal
perforations between panels for ease in
tearing out sections.

Accessories for studio operation include
various types of mobile and stationary
stands or stand attachments for mounting
the TelePrompter units. A newly de-
veloped camera mount, used to support

520

June 1952 Journal of the SMPTE Vol. 58

the TelePrompter, shown in Fig. 3,
is the most recent addition to the ac-
cessory line. This mount is installed
between the friction head of the pedestal
or tripod and the camera. It permits
quick and easy adjustment of the center
of gravity of the camera so as to balance
the weight load of any other camera
accessory with respect to the axis of tilt.
A TelePrompter unit can be supported
by this mount so that it is fixed with
respect to the taking lens, regardless of
camera motion. Furthermore, it is
supported in this position without
hindering the operation or accessibility
of the camera in any way, and without
obstructing the view of the cameraman,

while retaining perfect balance of the
camera on the friction-head assembly.
Although developed primarily for the
TelePrompter, the camera mount al-
ready is a ready and excellent solution
to other camera-balancing and accessory-
mounting problems and is now being
used for that purpose.

The TelePrompter is a modern solu-
tion to an old problem that has been
aggravated by urgent production sched-
ules. It is a new tool, flexible enough
to be fitted to particular production
problems, placed at the disposal of the
director to save time, to reduce tension
and to help achieve a smooth per-
formance.

Barton and Schlafly: TelePrompter

521

The Synchro-screen as a Stage
Setting for Motion Picture
Presentation

By BENJAMIN SCHLANGER, WILLIAM A. HOFFBERG
and CHARLES R. UNDERHILL, Jr.

The Synchro-screen* is described as consisting of a motion picture screen
with contiguous reflecting side wings, top and bottom panels. The picture
surround surfaces synchronously fluctuate in light intensity and color with
the changes in picture light and color adjacent to the reflecting surround
areas. There is an appreciable increase in the subtended angles of the
luminous field of view of the theater patron. A luminous, maskless stage
setting is thus created for the viewing of motion pictures.

I

T is A WELL-KNOWN FACT that ophthal-
mologists, and others concerned with
the care of the eyes, do not advise any
condition where there is a high central
but inadequate peripheral illumination.
Reading a book or working under a
shaded lamp in an otherwise darkened
room is known to cause eye fatigue and
even injury. The conventional black
masking used with the motion picture
screen is a glaring example, on a larger

Presented on April 22, 1952, at the Society's
Convention at Chicago, by R. H. Heacock
for the authors, Benjamin Schlanger and
William A. Hoffberg, Theatre Consultants,
35 W. 53d St., New York ;9, N.Y., and
Charles R. Underbill, Jr., Radio Corpora-
tion of America, RCA Victor Div., Engi-
neering Products Dept., Camden, N.J.

* Manufactured for and distributed by
RCA.

scale, and has been denounced by
lighting specialists for years.

Luckiesh and Moss1 have proven
that certain eye muscles become fatigued
from strain under the condition of dark
surroundings and that the strain is
relieved when there is some general
peripheral lighting available without
changing the brightness of the central
objects. Employees in modern industry
are no longer required to work under
purely localized light sources, because
employers know that the effects of dark
surroundings for any visual task are
psychologically depressing and that
associated eyestrain and fatigue will
result. The armed forces have given
much thought to avoiding a dark sur-
round on certain instrument panels and
radar screens by incorporating a lumi-

522

June 1952 Journal of the SMPTE Vol. 58

TOP PANEL

SIDE
WIWQ

BOTTOM
Fig. 1. The RCA Synchro-screen. PANEL

SIDE
WING

EDGE

DE- FOCUSING

SURFACE

nous surround area in the design of
military equipment.

Manufacturers of home television
receivers were quick to recognize the
importance of having the most favorable
viewing conditions for a highly com-
petitive product. An obvious difference
in the viewing of home television and
theater motion picture images is that
home television receivers generally have
light gray masks, whereas the conven-
tional motion picture screen has black
masking — a device as old as the motion
picture industry.

The viewing of motion pictures in
theaters is one of many visual tasks that
are performed daily for prolonged periods
by millions of people. Audiences, from
the earliest days of the art, have never
complained about prevailing lighting
conditions until they have had better
conditions for comparison. It is then
that the older conditions are judged
primitive. Therefore, the public has
been inclined to accept black masking
on motion picture screens, just as it
has accepted primitive lighting condi-
tions in other fields.

The Synchro-screen completely elimi-
nates the black masking and complies
very effectively with two distinct steps in
the disposition of lighting as suggested
by Luckiesh and Moss1:

(1) the attainment of maximal visi-
bility within the central field without
regard to the surroundings and

(2) the lighting of the surroundings
in such a manner as to produce maximal
comfort and ease, and minimal loss of
visibility in the central field.

The first step is effectively attained by
the use of a gradationally perforated
sound screen known commercially as
the Evenlite screen.2 The design of
the surround, which is the outstanding
feature of Synchro-screen, adequately
meets the requirements specified in the
second step. The necessity of co-
ordinating the requirements of both steps
in the design of a screen stage setting is
emphasized in the words of Luckiesh
and Moss1:

"The important positive contribution
to the lighting of the surroundings is the
creation of conditions that will provide
maximal visual and mental relaxation,
minimize eye strain and fatigue without
causing undue distraction from the
picture, and promote safety. This
means a proper balance or compromise
of these factors."

In connection with these conditions,
they have noted that the use of fixed,
colored light appears unadvisable in
the surrounding field for the same
pyscho-physiological reason as darkness
and glaring sources. By reflecting the
light and color of the projected picture
to the surround, the Synchro-screen has
avoided this disadvantage.

Fundamentally, a Synchro-screen
stage assembly consists of the main

Schlanger, HoiFberg and Underbill: Synchro-screen

523

--WING

SUBTENDED ANGLE OF
LUMINOUS FIELD (L)

SUBTENDED
PICTURE
ANGLE (P)

SUBTENDED ANGLES L$P

N° IO WING

M° SWING

Nl£7 WING

24'W

20'W

20'W

10 'W

IG'W

i?'W

L

4b°32

40C4G!

36*52'

32*2'

30" 14

25°20

P

27-0'

22°38

22*38

i&°ld

!8°ICi

I3°42

L/P

\.<fi

I.8O

\.G3»

1.77

1.G7

1.85

AVERAGE INCREASE = 73/£= (rf-}

Fig. 2. The proportion of luminous surround area to picture area varies from 90%

to 120% based on subtended horizontal and vertical angles of Synchro-screen

from above seating position.

picture screen and frame to which is
added a surround screen area comprising
two side wings, a top panel and a bottom
panel, as shown in Fig. 1. The two
wings, as well as top and bottom panels
are basically screens and frames of
special designs, so that the stage setting
assembly consists essentially of five
screen surfaces and frames arranged in
a definite relation to each other and to
the projected picture light. In the
same manner that the conventional
black masking reduces slightly the
maximum projected picture area for
the purpose of concealing the projected
edges of the picture aperture, so are
the four surround panels of Synchro-
screen set to enclose a picture area
slightly smaller than that projected, in
order that the edges of the projected
picture will fall on the surround screens
very close to their inside edges. The
design and positions of the surround
screens cause the image of the picture
edges to be so completely out of focus
that no borders are discernible from the
seating area. Neither colored aberra-
tion edges nor borderline picture jump
are noticeable.

The screen surfaces of the surround
area receive and simultaneously reflect
diffused light from the relatively adjacent
areas on the picture screen in such a

manner that the intensity of light and
the predominant color of the adjacent
picture area are reflected synchronously
from the surround panels to the audience
as a blended extension of the picture
in light intensity and hue. At no time
during the presentation of a picture on
the Synchro-screen is the brightness of
the surround area as bright as the nearest
area of the picture. Therefore, there
is never any distraction of the attention
from the picture itself.

It can be seen from Fig. 2 that there
is an appreciable and dramatic increase
in the luminous field of view produced
by the Synchro-screen. The picture
seems larger and therefore closer to the
audience because the subtended picture
angle is related by the eye to the syn-
chronous, luminous surrounding field
which has a subtended area of from
90% to 120% of the picture area. The
average increase in subtended horizontal
angle to the luminous field is about
73% and produces a strong horizontal
extensional effect which approximates
the relatively subdued luminosity of the
monocular portion of the field of view.

Synchro-screen Stage Setting As-
semblies are manufactured as a complete
package in 17 sizes, beginning with a
minimum projected picture width of 12
ft and increasing by 1-ft width steps
up to about 30 ft as a practical limit.

524

June 1952 Journal of the SMPTE Vol. 58

APPROX. OUTLINE OF
PROJECTED WHITE
LIGHT ON SIDE WINGS,
TOP a BOTTOM PANELS

HEIGHT OF FRAME • H + 30

HEIGHT OF PROJECTED WHITE
IGHT FOR MAX. PICTURE SIZE

2"X 3" INSIDE FRAME MEMBER
OF PANEL IN CONTACT WITH PICTURE
SCREEN

NOTE' PICTURE FRAME MEMBERS
2"X 4" FOR OVERALL SCREEN SIZES
TO AND INCLUDING 12 FEET BY 16 FEET.
2"X 6" FOR OVERALL SCREEN SIZES
OVER 12 FEET BY 16 FEET.

Fig. 3. Relation of panels to picture screen.

The picture screen, as previously men-
tioned, is the Evenlite screen described
in a paper presented at the April 19512
meeting of this Society. The fact that
this screen has no perforations in the
side areas has three important effects:
firstly, the screen material of the wings
exactly matches the sides of the picture
screen; secondly, the side areas of the
picture screen reflect the maximum
amount of light so essential for optimum
illumination of the surround; and
thirdly, the Evenlite screen has been
found to be a practical solution to the
problem of obtaining an optimum
screen brightness of an essentially uni-
form value in foot-lamberts from all
points on this screen's surface.

Synchro-screen shipments are made

knocked down. Total shipping weight
for a picture size of 1 5 by 20 ft is about
800 Ib. The five screen surfaces used
in the complete assembly are shipped
in tubes and the structural frame mem-
bers are strapped into bundles with the
exception of the top and bottom panel
frames which are prefabricated into
half sections for more rapid assembly on
the job. Hardware is included in a kit
of assembly parts and instructions. All
members are lettered for ready identi-
fication and match marked for easy
assembly. Basically, the installation of
a Synchro-screen requires the erection
of five frames and screens of various
sizes (comprising the picture screen, the
left and right side wings, the top and
bottom panel) with wings and panels in

Schlanger, Hoffberg and Underbill: Synchro-screen

525

Fig. 4. Synchro-screen installed in the Plaza Theatre, New York City.

proper relation to the picture screen
(Fig. 3). The picture screen is face-
laced to bring the screen surface in a
plane where the wings and panels can
be in close contact. The screen ma-
terial is Firestone Velon made to RCA
specification. It is wrapped around
those edges of the wings and frames
which are in contact with the picture
screen or exposed to view.

Architecturally, the Synchro-screen
presents the appearance of an orderly,
organized and attractive stage setting for
motion picture presentation. It offers
an economical method of creating a new
atmosphere in the zone of maximum
patron attention in the auditorium. The
use of a draw curtain in front of the
Synchro-screen has been found to have
dramatic value, especially when the
large luminous reflecting surfaces grad-
ually come into view as the performance
begins. A photograph of a Synchro-

screen installation (Fig. 4) in the
Plaza Theatre, New York, illustrates
the use of a curtain and valance of a
light gold-tint material.

Without an interest in or prolonged
attention to the picture presentation
(conditions which a theater patron
assumes), an initial and momentary
casual appraisal of the surround surfaces
may result in a negative first impression,
such as one of distraction. However,
it must be remembered that the patron
desires to concentrate on the main pic-
ture dramatic development. At such
times, the Synchro-screen surround areas
are in the peripheral field of view since
the human field of view tends to become
narrower when pyschological concentra-
tion exists. Distraction is completely
eliminated when concentration occurs.

That the Synchro-screen is easier on
the eyes, gives the effect of better color,

526

June 1952 Journal of the SMPTE Vol. 58

causes an illusion of added width and
height, and creates an atmospheric
effect is repeatedly attested to by patrons
in unsolicited statements to theater
personnel. From an audience reaction
viewpoint, estimated as at least 100,000
persons to date, Synchro-screen has
proven to be a major improvement in
the presentation of motion pictures in
theaters.

References

1. M. Luckiesh and F. K. Moss, "The
motion picture screen as a lighting
problem," Jour. SMPE, 26: 578-591,
May 1936.

2. G. R. Underbill, Jr., "Practical solution
to the screen light distribution problem,"
Jour. SMPTE, 56: 680-683, June 1951.

3. B. O'Brien and C. M. Tuttle, "Experi-
mental investigation of projection screen
brightness," Jour. SMPE, 26: 505-517,
May 1936.

4. B. Schlanger and W. A. Hoffberg,
"New approaches developed by relating
film production techniques to theater
exhibition," Jour. SMPTE, 57: 231-237,
Sept. 1951.

5. B. Schlanger, "A method of enlarging
the visual field of the motion picture,"
Jour. SMPE, 30: 503-509, May 1938.

Discussion

R. H. Heacock: You know, Barton, I
have a confession to make about Synchro-
screen. I, like most of you, have heard
Ben Schlanger talk about light surround
areas as applied to motion picture screens
for some years. Since there are very few
new theaters that can be constructed like
that at Framingham (which was originally
designed and built for a light surround
screen), as compared to our twenty-
thousand existing theaters, my interest
was only academic.

From a practical viewpoint, another
factor colored my reaction to Ben's ideas.
I think most of you have experienced
what I refer to. We are told most seriously
of the marked advantages of the dissolving
action of double shutter projectors; of
the greater, whiter light of a certain arc
lamp; of the longer, flatter response
curve of a particular theater sound system.
And yet we who tell you of these wonders

would be hard pressed to tell you with
absolute certainty specifically what equip-
ment was in the booth of a particular
theater where you might ask us to see and
hear a current Hollywood release with you.

Last fall I saw the first installation of
Synchro-screen. Each time I watch an-
other picture on this screen I become
more convinced that this packaged stage
setting will do more to improve an existing
theater's film presentation than the same
amount of money could if spent for any
other improvement.

Each of you has been given a courtesy
admission card at the time of registration
to the Palace Theatre to see an actual
operating theater equipped with Synchro-
screen. If you want to see how a technical
idea has been adapted to actual operating
conditions, take a ten-minute cab ride,
see one reel of the feature, and then we
believe you will understand the great
interest in this practical modern method of
stage setting for presentation of our current
Hollywood productions.

W. R. Cronenwett: I'd like to ask whether
Synchro-screen equipment can be so
designed to be flown to a grid?

Mr. Heacock: It probably ' can, but at
the present time we do not have such
equipment. Our current installations are
set up and are rigidly assembled to the
stage floor. With any special require-
ments like that, we would be very happy
to discuss them in detail with you and I
think it's always possible to overcome
difficulties of that kind if there is money
enough in the kitty to pay for this special
feature.

/. A. Tanney: Does the name Synchro-
screen come from the fact that the lighting
in the surround varies with the brightness
of the image on the screen?

Mr. Heacock: That's exactly right. In
England I saw a write-up in the Inter-
national Projectionist where they took the
light from the frame that was four frames
above the actual projected picture. Some
of the light was diverted up through an
associated optical system and then pro-
jected around the edges of the main
picture screen. My interpretation of that
article would be that if we had a Techni-
color feature, we would have a surround
area that would be a blend of all of the
colors in that particular frame. One
other thing, it is out of sync by the amount

Schlanger, Hoffberg and Underbill: Synchro-screen

527

of four frames. For those of you who
may have seen the Desert Fox, there was
quite a sequence of an artillery barrage
where you would see a blast of light and in
silhouette, a gun crew in this corner of the
picture screen and in a fraction of a second
there would be another flash of light up
here, and finally there were flashes here,
here and here. I happened to see that
on Synchro-screen and, of course, in exact
synchronization the whole area of the
side wings and bottom wings flashed into
view here and tip here, then here, and then
over here, so that truly the word Synchro-
screen is not just a trick name, but has
meaning. The surround areas are in
perfect synchronization not only with
light variation but also with variation in
color intensity.

If the area at the left side is blue, the
wing on that side is blue. If the upper
righthand corner is dark red, the surround
area there is dark red.

Anon: Where can this screen be seen
in New York?

Mr. Heacock: The Plaza Theatre on
58th Street and Madison Avenue has the
Synchro-screen, as does the Plaza in
Scarsdale, New York.

R. L. Estes: I'm interested in knowing
how the light surround reflected from the
wings affects the shadow detail and also
what is the effect of viewing from different
angles the width of the wings?

Mr. Heacock: Although I have seen six
or eight different feature films in various
theaters equipped with Synchro-screens,
I have not been aware of any washing out
of the picture at the edges. Although
theoretically a very small amount of light
might be diffused back on to the edges
of the main picture screen, from a practical
viewpoint this effect is not perceptible.
I have not heard any criticisms of the
screens in this respect from any others
who have seen them.

When I first saw Synchro-screen, I
moved around from the center of the
balcony to the sides of the balcony, to
close-up on the right, and close-up on the
left. There is a variation. There is a
change, but I didn't sit in any seat where
I had any but a favorable reaction and my
own feeling was that here was something
that would really have popular appeal.

I believe that your question depends a
great deal on your own personal reaction ;
consequently, I strongly recommend that
you go down to the Palace Theatre here
in Chicago and check into each of these
points, and I shall be very much interested
in hearing of your own personal conclu-
sions.

You'll notice in the advertisements of
the Palace Theatre — I cut this out of
the paper this morning — it says here in
capital letters: SEE THIS GRAND
PICTURE ON OUR NEW THREE-
DIMENSIONAL SYNCHRO-SCREEN.
We know that this is not three-dimensional,
but I think if you happen to see a picture —
the ballet sequence in An American in Paris
was mentioned here this morning by Mr.
Handley — you would be almost willing
to declare that you did see a three-dimen-
sional picture. But it's the same picture
on the same plane, flat picture screen.
The psychological effect of the light sur-
round area, I do think, tends to help in
your feeling of a three-dimensional effect.
But technically, of course, it is not.

C. L. Greene: Is there, in Chicago, an
installation of Ben Schlanger's original
type of screen, where the margins of the
screen transmitted some picture light to a
reflective surface behind it, a specular
reflecting surface which then reflected that
light out on to side wings.

Mr. Heacock: I do not believe there
is. I don't know if there is such an
installation.

Mr. Greene: That was demonstrated
with a model in New York, I believe, in
1937. I heard the comment repeated
many times at that meeting, that that was
the greatest advance in picture presenta-
tion since the coming of sound. Person-
ally, I still subscribe to that opinion. It
is not flattering, I think, to the exhibition
branch of the industry that it has shown
an almost psychopathic fear of advance-
ments such as this. I am in a way sorry
to see it come in a package form which I
fear is not the equal of Mr. Schlanger's
original design, but at the same time I
am happy to see the industry accept any
improvement.

Mr. Heacock: Well, go down to the Palace
and see how you like it. I'll be very much
interested in talking to you about it.

528

June 1952 Journal of the SMPTE Vol. 58

Resolution Test Chart

of the Motion Picture Research Council

THE MOTION PICTURE RESEARCH COUN-
CIL has announced the preparation of a
resolution test chart designed primarily
for the use of studio camera departments.
When mounted in front of a 35mm
camera lens at a distance in inches
equivalent to the focal length of the lens
in millimeters, its image exactly fills a
standard 35mm aperture (American
Standard Z22.59-1947). Resolution test
figures in key positions over the field
will then indicate limits of resolution

Articles describing the development and
construction of the chart have been
published :

1. Armin J. Hill, "A new resolution test
chart for motion picture camera lenses,"
Phot. Set. and Tech. (Sec. b, PSA Jour.),
77: 68-70, Sept. 1951.

2. Armin J. Hill, "A resolution test chart
for motion picture cameras," Am.
Cinematographer, 32: 402, Oct. 1951.

directly in lines per millimeter. Other
test patterns give qualitative indications
of serious aberrations or other lens
defects while carefully designed focus
figures at the center and each corner
assist in studies involving depth of focus
and curvature of field.

Although designed primarily for use
with 35mm camera lenses, the chart
may be used satisfactorily with any
photographic lens. It may not fill the
field at the specified distance, but simple
conversions of the indicated scale units
can be used without difficulty. An
illustration of the chart is included. It
is printed on white card and has an
overall dimension of approximately 16
by 22 in. Copies may be obtained for
$3.00 each, postage prepaid, directly
from the Motion Picture Research
Council, 1421 North Western Ave.,
Hollywood 27, Calif. "Directions for
Use" accompany each chart.

10-

= 111

.5-=|ll

20-=IH

*>-=»«

Fig. 1. Detail of focus figure.

June 1952 Journal of the SMPTE Vol. 58

Fig. 2. Enlarged detail of the
resolution pattern.

529

530

f =111111111111= V

Fig. 3. Motion Picture Research Council Resolution Test Chart
June 1952 Journal of the SMPTE Vol. 58

Laboratory Practice
Committee Report

By JOHN G. STOTT, Committee Chairman

_L HIS COMMITTEE was organized under
its present Chairman and Committee
personnel early in 1949. The original
agenda of this Committee was thought
to be concerned primarily with chemical
and chemical engineering problems
associated with the motion picture
industry. The proposed agenda was
directed into these particular channels
primarily because the membership of
the Committee was composed largely
of chemists and chemical engineers
working in motion picture laboratories.
The first meeting of this Committee,
however, brought out rather clearly that
there were many problems to be con-
sidered before too much effort was
expended on purely chemical and
chemical engineering problems.

At the Committee's first meeting the
following proposed agenda was laid out:

1. Design of a special leader for
television films to replace the Academy
theater leader.

2. Aid in the standardization of screen
brightness for 16mm projection.

3. Establishment of a standard for
the notching of 35mm and 16mm nega-
tive films.

4. Investigation of the possibility of
modifying sound and picture reduction

Presented on April 24, 1952, at the So-
ciety's Convention at Chicago, 111., by
John G. Stott, DuArt Film Laboratories,
245 W. 55 St., New York 19, N. Y.

printers to print forward and backward
and to employ 2000-ft negative feed and
take-up.

5. Investigation of the standardization
of edge numbering of 16mm films.

6. Study recommendations for the
splicing of 16mm films.

7. Study 16mm projection emulsion
position.

8. Study methods of bringing data
on chemical and chemical engineering
developments to the attention of Society
members.

This Committee is able to report as
follows on the agenda laid out early in
1949:

1. A special leader for television
films to replace the Academy theater
leader has been designed, submitted to
the membership of the Society for
comment, has been approved, and is
currently in widespread use throughout
the industry. The work on this leader
has been carried out by the Leader
Subcommittee of the Films for Tele-
vision Committee under the Chairman-
ship of Charles Townsend. The repre-
sentative from the Laboratory Practice
Committee on the Leader Subcommittee
was V. D. Armstrong.

2. A subcommittee on 16mm review-
room screen brightness of the Labora-
tory Practice Committee under the
Chairmanship of O. E. Cantor has

John G. Stott: Laboratory Practice Report

531

drafted a Proposed American Standard
for screen brightness of 1 6mm laboratory
review rooms. One criticism of the
Proposed Standard concerns the dis-
crepancy between the established stand-
ard for 35mm and the proposed standard
for 16mm. It is felt, however, that
due to the extremely difficult conditions
under which 16mm films are projected,
particularly in Armed Forces installa-
tions, a screen brightness standard
identical to the 35mm standard would in
most cases be unattainable and hence,
unrealistic. The problem is further
complicated by the amateur and tele-
vision fields.

3. The Proposed American Standard
on Size and Location of Notches for
35mm Negatives was drawn up by Paul
Kaufman and was submitted to the
Laboratory Practice Committee for bal-
loting. Considerable opposition to the
Standard arose from the Committee
members because many printers in the
industry could not be made to conform
to this Standard without extensive and
expensive alterations. It was also
learned that many thousands of reels
of negatives are presently stored in
laboratory vaults from which prints
are ordered from time to time, these
negatives being notched in accordance
with past practice and hence, in many
cases being notched contrary to this
proposed Standard.

At a later meeting of the Laboratory
Practice Committee it was decided that
it would, at the present time, be im-
possible to arrive at any compromise
which would make it desirable to adopt
this Standard. It was also felt that it
would be much simpler to establish
a standard for 1 6mm films since the need
for a notching standard in 16mm is
so urgent. Further observations were
that the method of "notching" would
have to be totally different and distinct
from past methods and also essentially
noninterfering with presently notched
films in order to make the establishment
of a standard even remotely possible.

Lloyd Thompson had been working
on other means of "notching" 16mm
films and offered to turn over to the
Society any methods, patents or licenses,
free of charge, providing this informa-
tion would aid in the establishment of
an American Standard. Mr. Thomp-
son, after a great deal of research and
development, has devised a means of
actuating light-change mechanisms by
means of applying an electrical con-
ducting material to the emulsion of the
film which may be used as a cueing de-
vice for such light-change mechanisms.
Mr. Thompson has prepared a report
outlining the history and background of
printer light-change cueing methods
and proposals and has drawn up a
Proposed American Standard. This
report is complete in all respects includ-
ing drawings of the mechanical elements
required, amplifier circuits, parts list
and approximate prices. This report
was submitted to the Laboratory Practice
Committee for balloting early in April
1952.

This Proposed American Standard
makes it possible to provide printer
light-change cueing of 16mm negatives
whether they have been notched before
or not. It is hoped that when this
method of cueing of 16mm negatives
has been adopted as an American
Standard, the same technique may be
applied to 35mm, thus solving the
35mm notching problem.

4. The manufacturers of motion
picture printers were contacted re-
garding the possibility of modified sound
and picture reduction printers to print
forward and backward and to employ
2000-ft negative feed and take-up. It
was the unanimous opinion of the equip-
ment manufacturers that modification
of sound and picture reduction printers
now in use to print forward and back-
ward would entail extensive modifica-
tion of present equipment and would
involve considerable increase in price
in future models. It was the opinion

532

June 1952 Journal of the SMPTE Vol. 58

of the Committee membership that it was
not a function of the Laboratory Prac-
tice Committee to propose to equipment
manufacturers changes of this type, but
was rather a matter concerning negotia-
tion between equipment manufacturers
and the purchasers of the equipment.

With regard to the use of 2000-ft
negative feed and take-up assemblies
on these printers, it was learned that
some manufacturers of this type of
equipment were already turning out
printers so equipped. Existing equip-
ment in laboratories in many cases had
been altered by their engineering de-
partments. It was thus decided that
this matter should be tabled.

5. At the time the Laboratory Prac-
tice Committee investigated the stand-
ardization of edge numbering of 16mm
films, a Proposed American Standard
was already in the process of being
drafted by the 16Mm and 8Mm Com-
mittee of the Society. This Standard
has now been submitted to ASA for
balloting.

6. In spite of a great deal of discus-
sion in Laboratory Practice Committee
meetings, it was impossible to arrive at
any conclusions regarding recommenda-
tions on the splicing of 16mm films. A
standard for splicing of 16mm positive
films was drafted by the 16Mm and 8Mm
Committee and has just been approved
as an American Standard.

7. The matter of 16mm projection
emulsion position has been discussed
thoroughly in several engineering com-
mittees of the Society including this one.
A symposium on this problem was held
at the Hollywood Convention in Oc-
tober 1951, but the matter seems still
to be completely up in the air. It is
certain that the entire membership of
the Society is familiar with the problems
of 16mm projection emulsion position
and hence, no further amplification is
required here.

8. Under the editorial direction ol
Irving Ewig, the Laboratory Practice
Committee has been contributing to the
Journal under the heading of "Chemical
Corner" which includes suggestions, tips,
new methods, new equipment and new
techniques related primarily to chemical
and chemical engineering problems.
This project requires a great deal of
time on the part of Mr. Ewig and it is
hoped that Committee members and
the membership at large of the Society
will aid Mr. Ewig in this work by sub-
mitting interesting material.

Since this agenda was proposed, other
projects have been referred to the
Laboratory Practice Committee for com-
ment and action.

At the Lake Placid meeting of the
Committee on Color, it was pointed out
that a definite need exists for the estab-
lishment of a standard magnification
ratio for the production of 35mm color
prints from 16mm preprint material.
The Committee on Color recognized
this as a laboratory problem and rec-
ommended that this project be assigned
to this Committee. This project was
undertaken by Gordon Chambers, who
prepared a proposed 16mm optical
printer aperture for enlargement print-
ing to 35mm. This Proposed American
Standard examines several possible solu-
tions and recommends what appears to
be the most logical choice. The Labora-
tory Practice Committee has been
balloted on this Standard and, as a
result of the voting, this Proposed
Standard has been submitted to the
Standards Committee for processing.

At the Convention held in Chicago
in April 1952, another meeting of the
Laboratory Practice Committee was
held. At this meeting the Proposed
Standard on screen brightness for 16mm
laboratory review rooms was discussed
at quite some length. Considerable
opposition to the proposed brightness
was raised because of the wide divergence
from usual projection screen brightness

John G. Stott: Laboratory Practice Report

533

for 16mm amateur films. It was rec-
ommended that the Screen Brightness
Committee as well as this Committee
be balloted and that further action be
held up until the results of the voting are
known.

At this meeting it was also decided
that this Committee would embark on
a program of drawing up definitions or a

system of nomenclature for all chemical
operations found in motion picture
laboratories in order to clarify terms in
technical discussions and literature.
Should this chemical glossary prove
successful, future plans would include
expansion of this glossary to include all
laboratory products, equipment and
functions.

Standards for Splices and
Projection Reels

ON APRIL 30, 1952, the American
Standards Association approved one
new standard, PH22.77-1952, Splices
for 8 MM Motion Picture Films,
and revision of two previous stand-
ards, PH22.24-1952, Splices for
16MM Motion Picture Films for
Projection, and PH22.11-1952,
16MM Motion Picture Projection
Reels.

All three standards were initiated
and/or revised by the 16MM and
8MM Committee. The splice
standards were published for trial
and comment January 1951 and
the reel standard first in February
1950 and again in February 1951
when the initial publication re-
sulted in comments indicating a
need for further revisions.

534

June 1952 Journal of the SMPTE Vol. 58

American Standard

ASA

Keg. U. S. Pat. Off.

for

PH22.11-1952

Revision ol

16-Millimeter Motion

Picture

Z22.II-I94I
and

Projection Reels

Z52.33-I945

•UDC 778.55

Pa0.lof4po9M

•w-*

-AT PERIPHERY \ ^ I

' Qi ,

S ^\

•w*

-AT CORE /

/ /'*"~^~[ 1

ENLARGED VIEW

OF HOLE IN

R

^

FLANGE ON LEFT IN SECTIONAL

VIEW SHOWN ABOVE

VV

T

* rrr

( r<M

U-S-4 ENLARGED VIEW OF HOLE IN
FLANGE ON RIGHT IN SECTIONAL
VIEW SHOWN ABOVE

Table 1 See page 3 for notes.

Dimension

Inches

Millimeters

A

0.319

+0.000
-0.003

8.10 +

B

0.319

+0.000
-0.003

8.10 +

R1

0.790

maximum

20.06 maximum

S2 (including flared,

rolled, or beveled

0.962

maximum

24.43 maximum

edges)

T (adjacent to

0.027

minimum

0.69 minimum

spindle)

0.066

maximum

1.68 maximum

U

0.312

±0.016

7.92 ±0.41

V

0.125

+0.005
-0.000

31 Q "^
• 1 O

X13

W, at periphery3

0.660

+0.045
-0.025

16.76 +

1.14

at core4

0.660

±0.010

16.76 ±0.25

at spindle holes

0.660

±0.015

16.76 ±0.38

Flange and core
concentricity5

±0.031

±0.79

Approved April 30, 1952, by the American Standards Association, Incorporated

Sponsor: Society of Motion Picture and Television Engineers

«Univrr»«t Decimal CUxiRcltloo

Copyright, 1952, by American Standards Association, Inc.; reprinted by permission of the copyright holder.

June 1952 Journal of the SMPTE Vol. 58

535

American Standard
for

ASA

Rff. V. S. fat. Off.

16-Millimeter Motion Picture
Projection Reels

PH22.11-1952

Page 2 of 4 pages

Table 2

Capacity

Dimension

Inches

Milli-
meters

Capacity

Dimension

Inches

Milli-
meters

200 feet"
(61 meters)

D, nominal
maximum
minimum

5.000
5.031
5.000

127.00
127.79
127.00

1200 feet
(366 meters)

D, nominal
maximum
minimum

12.250
12.250
12.125*

311.15
311.15
307.98*

C, nominal
maximum
minimum

1.750
2.000*
1.750

44.45
50.80*
44.45

•

C, nominal
maximum
minimum

4.875
4.875
4.625*

123.83
123.83
117.48*

Lateral
runout,7
maximum

0.570

1.45

Lateral
runout,7
maximum

0.140

3.56

400 feet1'1
(122 meters)

D, nominal
maximum
minimum

7.000
7.031
7.000

177.80
178.59
177.80

1600 feet
(488 meters)

D, nominal
maximum
minimum

1 3.750
14.000*
13.750

349.25
355.60*
349.25

C, nominal
maximum
minimum

2.500
2.500
1.750*

63.50
63.50
44.45*

C, nominal
maximum
minimum

4.875
4.875
4.625*

123.83
123.83
117.48*

Lateral
runout,7
maximum

0.080

2.03

Lateral
runout,7
maximum

0.160

4.06

800 feet
(244 meters)

D, nominal
maximum
minimum

10.500
10.531
10.500

266.70
267.49
266.70

2000 feet
(610 meters)

D, nominal
maximum
minimum

15.000
15.031
15.000

381.00
381.79
381.00

C, nominal
maximum
minimum

4.875
4.875
4.500*

123.83
123.83
114.30*

C, nominal
maximum
minimum

4.625
4.875
4.625

117.48
123.83
117.48

Lateral
runout,7
maximum

0.120

3.05

Lateral
runout,7
maximum

0.171

4.34

*When new reels are designed or when new tools are made for present
reels, the cores and flanges should be made to conform, as closely as prac-
ticable, to the nominal values in the above table. It is hoped that in some
future revision of this standard the asterisked values may be omitted.

536

June 1952 Journal of the SMPTE Vol. 58

American Standard Rf,v.s.Pal.ol.

for

16-Millimeter Motion Picture
Projection Reels

PH22.il -1952

Pag* 3 of 4 pages

Note 1: The outer surfaces of the flanges shall be flat out to a diameter
of at least 1 .250 inches.

Note 2: Rivets or other fastening members shall not extend beyond the
outside surfaces of the flanges more than 1 /32 inch (0.79 millimeter) and
shall not extend beyond the over-all thickness indicated by dimension S.

Note 3: Except at embossings, rolled edges, and rounded corners, the
limits shown here shall not be exceeded at the periphery of the flanges,
nor at any other distance from the center of the reel.

Note 4: If spring fingers are used to engage the edges of the film, dimen-
sion W shall be measured between the fingers when they are pressed out-
ward to the limit of their operating range.

Note 5: This concentricity is with respect to the center line of'the hole for
the spindles.

Note 6: This reel should not be used as a take-up reel on a sound projector
unless there is special provision to keep the take-up tension within the
desirable range of 1 V'2 to 5 ounces.

Note 7: Lateral runout is the maximum excursion of any point on the flange
from the intended plane of rotation of that point when the reel is rotated
on an accurate, tightly fitted shaft.

June 1952 Journal of the SMPTE Vol. 58 537

American Standard
for

16-Millimeter Motion Picture
Projection Reels

Kef . V. S. Pat. Off.

PH22.1 1-1952

Pag* 4 of 4 page*

Appendix

(This Appendix is not a part of the American Standard for 16-Millimeter
Motion Picture Projection Reels, PH22.1M952.)

Dimensions A and B were chosen to give sufficient clearance between the
reels and the largest spindles normally used on 16-millimeter projectors.
While some users prefer a square hole in both flanges for laboratory work,
it is recommended that such reels be obtained on special order. If both flanges
have square holes, and if the respective sides of the squares are parallel, the
reel will not be suitable for use on some spindles. This is true if the spindle
has a shoulder against which the outer flange is stopped for lateral position-
ing of the reel. But the objection does not apply if the two squares are ori-
ented so that their respective sides are at an angle.

For regular projection, however, a reel with a round hole in one flange is
generally preferred. With it the projectionist can tell at a glance whether or
not the film needs rewinding. Furthermore, this type of reel helps the pro-
jectionist place the film correctly on the projector and thread it so that the
picture is properly oriented with respect to rights and lefts.

The nominal value for W was chosen to provide proper lateral clearance
for the film, which has a maximum width of 0.630 inch. Yet the channel is
narrow enough so that the film cannot wander laterally too much as it is
coiled; if the channel is too wide, it is likely to cause loose winding and ex-
cessively large rolls. The tolerances for W vary. At the core they are least
because it is possible to control the distance fairly easily in that zone. At the
holes for the spindles they are somewhat larger to allow for slight buckling
of the flanges between the core and the holes. At the periphery the toler-
ances are still greater because it is difficult to maintain the distance with
such accuracy.

Minimum and maximum values for T, the thickness of the flanges, were
chosen to permit the use of various materials.

The opening in the corner of the square hole, to which dimensions U and
V apply, is provided for the spindles of 35-millimeter rewinds, which are
used in some laboratories.

D, the outside diameter of the flanges, was made as large as permitted
by past practice in the design of projectors, containers for the reels, rewinds,
and similar equipment. This was done so that the values of C could be made
as great as possible. Then there is less variation, throughout the projection
of a roll, in the tension to which the film is subjected by the take-up mech-
anism, especially if a constant-torque device is used. Thus it is necessary to
keep the ratio of flange diameter to core diameter as small as possible, and
also to eliminate as many small cores as possible. For the cores, rather widely
separated limits (not intended to be manufacturing tolerances) are given in
order to permit the use of current reels that are known to give satisfactory
results.

538 June 1952 Journal of the SMPTE Vol. 58

American Standard

Splices for
16-Millimeter Motion Picture Films
for Projection

ASA

ficg. V. S. I'al. Off.

PH22.24-1952

Revision of
Z22.24-I94I,
Z22.25-l94l.and
Z52.20-I944

*UDC 778.55

Page 1 of 2 pages

Scope

Splices made in accordance with this standard are primarily for use with
films intended for actual projection, such as release prints and reversal films.
It is not intended that this standard be prejudicial to the use of diagonal-type
splicers, nor to the use of narrower splices for professional purposes. For
negatives and other laboratory films, narrower splices, sometimes with one
edge on the frameline, frequently are used.

Dimension

Inches

Millimeters

A

01 00 +°-000
—0.005

9 .. +0.00
h —0.13

B

4 +0.001

* -0.001

13.920 ijgy

C

°'324 ^0.003

8-23 ^:2s

D

+0.000
•4 -0.003

8-23 iSS

Note 1: In the plan view, the splice is arranged with the perforations at
the bottom in order to show them as they appear on most splicers. The
splice may be made with the film turned through an angle of 180 degrees,
or any other angle, but, of course, the emulsion surface should always be
up. It is customary to scrape the top (emulsion) surface of the left-hand .
film and to cement this scraped area to the bottom (base) surface of the
right-hand film. (Continued)

Approved April 30, 1952, by the American Standards Association, Incorporated
Sponsor: Society of Motion Picture and Television Engineers

Copyright, 1952, by American Standards Association, Inc.; reprinted by permission of the copyright holder.

June 1952 Journal of the SMPTE VoL 58

539

American Standard

Keg. V. S. Pal. Off.

Splices for

16-Millimeter Motion Picture Films
for Projection

PH22.24-1952

Note 2: Dimension A is given a negative, but no positive, tolerance be-
cause narrower splices are less conspicuous on the screen and are less
likely to affect the normal curvature of the film as it follows the bends in
its path through cine-machinery.

Note 3: Dimension B controls the longitudinal registration of the two films
being spliced. It is measured to the perforations that are most commonly
used for registration on splicing blocks, and to the nearer edges of these
perforations because they are edges that are generally used for the regis-
tration. This dimension is made the same as in the American Standard
Splices for 8-Millimeter Motion Picture Films, PH22.77-1952, because many
splicers are designed to accept either 16- or 8-mm film.

The nominal value of the B dimension was made 0.548 inch instead of
the usual 0.550 (for unshrunk film) because the films being spliced are
always shrunk to some extent. The 0.548 figure corresponds to a shrink-
age of 0.36 percent, while the 0.549 and 0.547 values, permitted by the
tolerances, correspond to 0.18 percent and 0.55 percent, respectively.
Thus, the tolerances include the range of shrinkage ordinarily encountered
when film is being spliced.

Note 4: Dimensions C and D were chosen to give a straight 0.100-inch
splice that is symmetrical about the included perforation (and, therefore,
the frame-line) when the film is shrunk 0.36 percent. (See Note 3.)

Note 5: The width of the film at the splice shall not exceed 0.630 inch.
If the film has been widened during scraping, the extra width shall be
removed.

Note 6: The overlapping perforations of the two films shall not be offset
laterally more than 0.002 inch.

Note 7: At the splice, the edges of the two spliced films shall not be offset
laterally more than 0.002 inch, unless a difference in the lateral shrink-
ages of the two strips makes it impossible to maintain that tolerance.
Shoulders formed by such misalignment shall be beveled after the cement
has dried.

Note 8: In the plan view, the angle between the respective edges of the
spliced films shall be 180 degrees, plus or minus 40 minutes. Thus, the
spliced film shall be aligned to the extent that when one portion of the
film is placed against a straight edge, the other portion will not deviate
more than 0.006 inch (approximately the thickness of the film) in 6 inches.

Note 9: In order to prevent the appearance of a white line on the screen,
the scraped area shall be 0.001 to 0.003 inch narrower than the area
covered by the overlapping film. The presence of this narrow uncemented
area w'JI not shorten the life of the splice.

540 June 1952 Journal of the SMPTE Vol. 58

American Standard

Splices for
8-Millimeter Motion Picture Films

Rtf . V. S. Pat. Of.

PH22.77-1952

•UDC 778.55

HM

Page 1 of 2 pages

DODO

ODD

^T

B

Dimension

Inches

Millimeters

A

•""SBS

254 +a°°
1 -0.13

B

-•SB

13920 +a°25
I3.5W _0025

C

0324 +0.000

°'324 -0.003

823 •fa00
8'23 -0.08

D

0324 +0.000

°""4 -0.003

823 •f0-00
8'23 -0.08

Note 1 : In the plan view, the splice is arranged with the perforations at the
bottom in order to show them as they appear on most splicers. The splice
may be made with the film turned through an angle of 180 degrees, or
any other angle, but, of course, the emulsion surface should always be up.
It is customary to scrape the top (emulsion) surface of the left-hand film
and to cement this scraped area to the bottom (base) surface of the right-
hand film.

Note 2: Dimension A is given a negative, but no positive, tolerance because
narrower splices are less conspicuous on the screen and are less likely to
affect the normal curvature of the film as it follows the bends in its path
through cine-machinery. (Continued)

Approved April 30, 1952, by the American Standards Association, Incorporated

Sponsor: Society of Motion Picture and Television Engineers *Univcrui Decimal ci«».ific«ii..n

Copyright, 1952, by American Standards Association, Inc.; reprinted by permission of the copyright holder.

June 1952 Journal of the SMPTE Vol. 58

541

American Standard

Splices for
8-Millimeter Motion Picture Films

PH22.77-1952

Page 2 of 2 pages

Note 3: Dimension B controls the longitudinal registration of the two films
being spliced. It is measured to the perforations that are most commonly
used for registration on splicing blocks, and to the nearer edges of these
perforations because they are edges that are generally used for the regis-
tration. This dimension is made the same as in the American Standard
Splices for 16-Millimeter Motion Picture Films for Projection, PH22. 24-1 952,
because many splicers are designed to accept either 8- or 16-mm film.
The nominal value of the B dimension was made 0.548 inch instead of
the usual 0.550 (for unshrunk 61m) because the films being spliced are
always shrunk to some extent. The 0.548 figure corresponds to a shrink-
age of 0.36 percent, while the 0.549 and 0.547 values, permitted by the
tolerances, correspond to 0.18 percent and 0.55 percent, respectively.
Thus, the tolerances include the range of shrinkage ordinarily encountered
when film is being spliced.

Note 4: Dimensions C and D were chosen to give a 0.100-inch splice that
is symmetrical about the included perforation (and, therefore, the frame-
' line) when the film is shrunk 0.36 percent. (See Note 3.)

Note 5: The width of the film at the splice shall not exceed 0.317 inch. If
the film has been widened during scraping, the extra width shall be
removed.

Note 6: The overlapping perforations of the two films shall not be offset
laterally more than 0.002 inch.

Note 7: At the splice, the edges of the two spliced films shall not be offset
laterally more than 0.002 inch, unless a difference in the lateral shrink-
ages of the two strips makes it impossible to maintain that tolerance.
Shoulders formed by such misalignment shall be beveled after the cement
has dried.

Note 8: In the plan view, the angle between the respective edges of the
spliced films shall be 180 degrees, plus or minus 40 minutes. Thus, the
spliced film shall be aligned to the extent that when one portion of the
film is placed against a straight edge, the other portion will not deviate
more than 0.006 inch (approximately the thickness of the film) in 6 inches.

Note 9: In order to prevent the appearance of a white line on the screen,
the scraped area shall be 0.001 to 0.003 inch narrower than the area
covered by the overlapping film. The presence of this narrow uncemented
area will not shorten the life of the splice.

542 June 1952 Journal of the SMPTE Vol. 58

71st Semiannual Convention

The several sources and aspects of the So-
ciety's overall interest were reflected from
the opening Monday noon Get-together
Luncheon right on through the eleven-
session technical program organized* by
Program Chairman Geo. W. Golburn and
held at The Drake in Chicago on April
21-25. Besides sessions arranged by topics
such as television, screens, high-speed
photography and usual motion picture
subjects, one session was labeled as to
source — the armed forces production
session. The complete roster of the
authors and their papers is given on later
pages of this Journal.

Excerpted below are the convention
introductory remarks by President Peter
Mole, who also introduced the principal
speaker, Dr. W. R. G. Baker, General
Electric Vice-President and General Man-
ager of its Electronics Div., Syracuse, N.Y.
Mr. Mole noted that Dr. Baker, known
for his pioneering in radio and television,
has, among his numerous contributions to

American industry, service as a key figure
in the coordination of technical develop-
ments through the work of industrial and
technical societies. He has been President
of the I.R.E. and R.T.M.A. His speech
on the work of the National Television
System Committee is given in part below.

President Mole also welcomed Jimmy
Frank as one who needs no introduction
to old timers of the Society. He spoke
as Deputy Director of the Motion Picture
and Photographic Products Div. of the
National Production Authority and his
speech is excerpted below.

A lighter part of the luncheon program
was the appearance of Leon Ames, motion
picture actor then starring in Chicago's
long-run stage performance of The Moon
Is Blue. He came to the luncheon through
the kindly offices of George Colburn to
assure engineers in a whimsical fashion
that even actors and engineers can get
along more and more agreeably as they
increase their appreciation of the other
fellows' side of the business.

Get-Together Luncheon Remarks by President Mole

It is encouraging to note that an interest
in engineering is developing quite generally
among motion picture people who have
not had technical training or experience.
Evidence appears in many places — in
plans of motion picture interests to appear
before the Federal Communications Com-
mission in the matter of theater tele-
vision — in current efforts to produce
stereoscopic motion pictures for theater
release — in the first steps toward extend-
ing the field of view of theater patrons
through changes in screen design or
illumination of the surrounding area —
and in widespread introduction of new or
enlarged facilities for release printing in
color. Another significant indication of
the new interest in engineering is the all-
industry research program proposed last
winter by Dr. L. A. du Bridge, President
of the California Institute of Technology.

The motion picture industry need not
wait for the institution of an extensive
research program, however, before further
progress can be made along technical

lines. A great deal can be done on a
small scale. Each manufacturer of equip-
ment, each producer of motion pictures,
and each exhibitor who possesses an
engineering or technical department can
conduct his own development program.

Young engineers can be hired to bring
new blood into an old organization, and
promising students can be employed during
summer vacations. Many students would
be glad to take temporary jobs on a trial
basis, with the chance for full-time work
after graduation. The more apt students
might well be given scholarships and
encouraged to pursue a course of study
relating directly to a particular industrial
problem.

Other steps which can be taken toward
the same goal include the placing of
development contracts with commercial
research organizations.

Our Society serves to coordinate those
activities of the industry that relate directly
to technical standardization and inter-
changeability. It also conducts studies

543

designed to assist entire branches of the
industry in learning more about their
technical operations. Through the
Journal, published twelve times each
year, it provides a valuable source of
technical data that helps in the orientation
of engineers just entering the field.

These are the means for steady sub-
stantial technical growth, and they are
available at nominal cost.

By working in concert with each other
and with the men who manage the business

of making and using motion pictures for
various purposes, our members provide
an effective and valuable service. For
that reason the Society deserves the con-
tinued active support of all related com-
mercial interests, as well as its members.

Given such backing, the Society will
continue to function as a fountainhead of
the technical progress which underlies
both service and profitable operation in
all branches of the motion picture and
television industries.

Excerpts From Address by Dr. W. R. G. Baker

Authorized by the Radio-Television Manu-
facturers Association, the present National
Television System Committee is a com-
mittee whose members are representatives
of organizations broadly interested in and
experienced in the television field, or of
technical organizations vitally interested
in research and development of television,
as well as qualified individuals not asso-
ciated with any organization, association
or company.

The Committee is charged with the
task of assembling technical data on:
(1) the allocation of channels in the ultra-
high-frequency band; (2) procedures
enabling the FCC to lift the "freeze" on
very-high-frequency allocation; and (3)
basic standards for the development of a
commercially practicable system of color
television ; and it is directed to undertake
such additional work as may be indicated
to provide more adequate television service
to the American public.

Various projects within the framework
of the charter of the NTSC are assigned to
individual members of the Committee or
to technical panels named by the chairman
with the concurrence of the vice-chairmen.
The members of the panels are drawn
from any company, association or organi-
zation regardless of affiliation. The only
requirement for membership on any panel
is recognized skill, interest and ability in
the assigned project. When each panel
completes an assigned project, it submits
a report stating both majority and minority
opinions to the NTSC.

The frequency spectrum and particularly
that portion utilized in communication,
is one of our national resources, as much a
national resource as are the supplies and

reserves of crude oil or forests or mineral
ores. Although the frequency spectrum
transcends all geographical or political
boundaries, most of the spectrum is of its
greatest value within such borders, and
therefore can be considered a physical
asset of political entity.

Misuse or inefficient use of the spectrum
is as great a physical loss as destruction
of our other natural resources. And effi-
cient use of the spectrum is difficult to
achieve for these reasons:

(1) The spectrum has physical limits.
At any one period in the development of
the electronic art, only a stated amount of
information can occupy the spectrum.

(2) The "lock and key" element is
present in every use of the spectrum.
Receiving equipment is useless without
transmitting equipment and vice versa.
Development of the two elements must
proceed simultaneously.

(3) Economics often prove a greater
factor in governing rate of development
than does scientific knowledge. The cost
cf moving a service from one section of the
spectrum to another may be so great that
it is not, at any stated time, in the public
interest to make the move even though it
would result in efficiencies throughout
the entire spectrum.

Although the Federal Communications
Commission maintains its own group of
technical experts, the competition inherent
in a private enterprise and within in-
dustry — and I'm sure you all will agree
there is no industry more competitive
than the electronics industry — would
normally be an obstacle to the FCC

544

obtaining free access to the latest technical
information, and particularly information
concerning promising avenues of develop-
mental research.

It is also true that in a competitive
business you may have persons or concerns
who will fight more vigorously for ac-
ceptance of an idea, or a development,
than will others with equally acceptable
ideas or developments. To reach a point
closest to a correct evaluation of ideas or
developments, to make such information
available to the FCC, to help focus re-
search and development toward the
quickest solution of a problem, an organi-
zation like the National Television System
Committee is necessary.

In 1939 the FCC held a hearing to
determine whether it was practical to
establish standards for monochrome tele-
vision. This hearing developed two im-
portant facts. First, monochrome tele-
vision had been developed to a point where
from the technical viewpoint, commerciali-
zation was a distinct possibility. Second,
at the hearing the industry had demon-
strated beyond a doubt that about the
only agreement on standards that could be
expected was an agreement to disagree.

As of January 15, 1940, the status of the
necessary standards for the establishment
of a national system of television was
substantially as follows: There was no
great difference of opinion in the industry
with respect to the frequency assignments
for television. However, it was clearly
apparent that in the industry there was
wide divergence of opinion concerning
the system standards. Furthermore, there
was every indication that unless and until
these divergent ideas could be reconciled,
progress toward a national system of
television was practically at a standstill.

Fourteen months later, after the forma-
tion of the first NTSC, full commercial
operation of monochrome television began
with the approval of the Commission. In
this short time the plan for the National
Television System Committee had been
formulated, the committee assembled, its
meetings held, its minutes recorded, tech-
nical reports compiled and its final report
delivered to the Commission. When the
Committee's recommendations were made
to the Commission, the complexion of the
industry had changed from a discord of
counterclaims to a concord of expert

opinion which persuaded the Commission
to acknowledge its value and to proclaim
the art open to the public.

From the time of the first NTSC,
color television has been under serious
consideration by the engineers of the
industry.

You are all familiar with the recent
hearing on color television. If one reads
the transcripts of the monochrome tele-
vision hearing preceding the establishment
of the first NTSC and the recent color
television hearing, certain facts are quite
evident :

(1) At the time of the monochrome
hearing several experimental transmitters
were in use providing television programs
to several hundred receivers. There was
little if any question on the ability of the
industry to provide a commercial mono-
chrome television service from the view-
point of providing products in the form of
transmitters, studio equipment, receivers
and certain specialized tubes. The public
had ample opportunity to see television
and accepted it enthusiastically.

(2) The industry was not faced with any
prior obligation such as obsolescence of
equipment in the hands of the public.
No television receivers had been sold,
hence the public had made no investment.

(3) While there was no disagreement
within the industry as to the availability
of the tools necessary to establish a com-
mercial monochrome television service,
there were wide differences of opinion as
to the standards on which such service
should be based.

The comparable situation with respect
to color television was as follows:

(1) There was no extensive experi-
mental broadcasting of color television
as was the case with monochrome tele-
vision. There were very few color tele-
vision receivers viewing such color tele-
vision broadcasting. A real question
existed as to the ability of the industry to
produce color television equipment capable
of rendering a commercial service. While
there had been some exposure of color
television programs to the public, it was
not in any sense as extensive as was the
case with monochrome.

(2) The public was confronted with
the possibility of the investment it had

545

made in monochrome television receivers
being obsoleted. There had been some
experience in the case of FM with the
mass obsolescence of products in the hands
of consumers and it was difficult to look
with favor on a repetition of this problem.
The entiie problem can be summed up in
the idea of compatibility with which you
are familiar.

(3) There was certainly real and justi-
fiable concern as to the availability of the
tools necessary to commercialize color
television.

Right or wrong, and personally I think
it was right, many engineers felt that
insufficient foundation had been laid to
warrant selection of a system of color
television.

With these facts in mind, it seemed to
the RTMA Television Committee that
one possibility of resolving the problem
was through the formation of a second
NTSG. Such action was recommended
to the Board of Directors of RTMA and
approved.

The NTSG proceeded under its assigned
charter but in the late fall of 1950 it be-
came clear that because of the many
advances and new proposals relating to

color television systems and components,
it would be necessary to have an up-to-
date appraisal of the state of the art. A
special, or ad hoc, committee of the NTSC
was established. Some months later, the
committee proposed a reorganization of
the NTSG to permit a more direct study
on the problems and outlined the broad
framework of a new standards for color
television achieved by combining the best
elements of the furthest advances of
existing proposals.

The NTSG accepted the ad hoc com-
mittee report because it stated a philosophy
without specifying details of a color tele-
vision system. The NTSC technical panels
were revised and the work of building
around the framework established by the
ad hoc committee has gone forward.

The value of the NTSG, to my mind,
has been the fact that, in almost every
case, when the technical panels sat down
to work, those forces which could have
distorted their vision, their deliberations,
were left outside the door. Engineers and
scientists tend to forget their pride in
their own idea or research when they are
joined in a conference with many others
whose technical competence they respect.

Excerpts From Address by James Frank, Jr.

The National Production Authority, under
the authority of the Defense Production
Act of 1950, created in the Department of
Commerce in September 1950, and the
Defense Production Administration estab-
lished in January 1951 were set up to
accomplish certain basic objectives. The
most important of these were:

(1) To set up programs and procedures
which would permit the most efficient
rapid progress of the defense mobilization
program and, in the event of the necessity
of full mobilization, to meet the require-
ments of the Department of Defense and
the Atomic Energy Commission. This
objective has been fully met by this time,
with emphasis on the fact that a full
Controlled Materials Plan has been put
into satisfactory operation.

(2) To get the defense mobilization under
way so that the military could, as quickly
as possible, build up its supply of necessary
material.

(3) To the extent possible, to permit the

accumulation of critical materials, for which
we depend on imports to a large extent,
as a stockpile in the event of full mobiliza-
tion.

(4) To encourage and effect expansion
of facilities for the production of raw
materials and military products to meet
the requirements of the Department of
Defense and the Atomic Energy Com-
mission in the event of full mobilization.

(5) To carry out all of the above objec-
tives without handicapping civilian in-
dustry to any greater extent than need be.

Let us get straightened out on some con-
fusing issues. The defense mobilization
program has been increased, not decreased,
and the 1953 Budget before Congress
now calls for an increase from 3,500,000
men and women in the armed services to
3,700,000 and, in the case of the Air Force,
from 95 Wings under the present program
to 143 Wings. The new program also
calls for certain increases in the Army,
Navy and Marine Corps. As a result of

546

world conditions at this time, our military
and civilian authorities believe that our
rearmament goals should be set somewhat
higher. At the same time, however, the
rearmament program has in fact been
lengthened from approximately three years
to approximately four years, but we are
now talking about the new increased
program.

There has been some cancellation of
military contracts but not to as wide an
extent as some are led to believe by the
newspapers. One of the reasons for this
is that the armed services at first placed
some contracts for military goods for the
increased program in the belief that that
program would be achieved in three years,
rather than four years. When the term
of four years was decided on, they cancelled
some contracts. Another explanation is
that the design of new machine tools,
substitution of materials, etc., has required
some changes in planning. Furthermore,
constantly improving design of weapons
calls for the altering of contracts and
cutting down or ceasing production of the
old types of products.

The amount of material that was re-
turned to DPA by the Department of
Defense for the second quarter can be
misleading. 25,000,000 pounds of copper
returned is 12,500 tons, or less than one-
half of one per cent of our annual con-
sumption of copper.

The easing of material shortages is
explained by the above minor return of
material and the fact that there has been
a considerable increase in our production
of certain basic metals, as well as electric
power, petroleum and the like. For
instance, steel production has been in-
creased from 100,000,000 tons a yearr
pre-Korea, to 109,000,000 tons, and will
continue to rise to 120,000,000 tons per
year. Aluminum production has in-
creased by 150,000 tons and eventually
will be double pre-Korea. Furthermore,
the operation of the Controlled Materials
Plan has really begun to work and manu-
facturers are able to book orders with
mills when they get such allotments.
That has minimized scare buying and has
encouraged reduction of large inventories.

However, there are now two serious
problems with respect to controlled ma-
terials. All of the copper products, as
well as nickel-bearing stainless steel, are

and will continue for some time, to be in
tight supply.

The constant flow of civilian products
during the past year, despite material
shortages, is a tribute to the ingenuity of
engineers such as those represented by this
Society. They have affected remarkable
conservation in the use of critical materials
to permit maximum production of end
products. However, at this time, certain
vigorous efforts are required to minimize
the use of all of the forms of copper and of
nickel-bearing stainless steel.

There is definite evidence that the
officials of DPA and NPA are anxious to
lift controls the moment that they can do
so safely. This is evident with respect to
rubber, lead, non-nickel-bearing stainless
steel, and others. However, we still
must travel a long road before we reach
the desired stage of decontrol. In the
third quarter, direct defense needs will
take 18% of all our carbon steel, 35% of
our alloy steel, 19% of our copper foundry
products, 27% of our copper wire mill
products, 46% of our copper brass mill
products, and 43% of our aluminum.
To some extent, moreover, expanding
production of materials in the months
ahead will be counterbalanced by ex-
panding military needs.

It is our feeling that until such direct
defense needs of any controlled material
reaches the level where they are not in
excess of 20 or 25% of total supply, they
should not be completely decontrolled.
It is hoped that certain forms of carbon
steel and aluminum may be decontrolled
early next year.

Some people have asked whether we
are not taking a terrific risk in extending
the Defense Mobilization Program from
three to four years. Unquestionably there
is a calculated risk, based on current world
conditions, but it is felt that it would be
unwise to break ourselves or cripple our
economy with a mad rush in rearmament
and then find we had a mountain of weap-
ons rapidly becoming obsolescent. We
are gathering strength in an orderly fashion,
which will keep us stronger over a longer
period of time than if we rushed now into
all-out mobilization.

One of the serious problems is military
design changes. Such design changes
mean improvements. Improvements
mean a finer fighting machine. For our

547

members of the armed forces they mean
more offensive power, more defensive
protection. It is a hard decision to have
to make between freezing designs and
pouring out the stuff, or allowing more
design changes and slowing up production.
If we thought the Soviets were going to
attack tomorrow, of course we would
freeze and pour it out, but if they did not
attack, we would have spent a lot of money,
chewed up a lot of much-needed copper
and aluminum, and have on our hands a
mountain of obsolescent weapons. By
allowing for a reasonable number of
design changes: we get better weapons
in the long run; we use up less material;
and we save some money for the be-
deviled taxpayer.

Since the establishment of the National
Production Authority, a year and a half
ago, a strong organization has been

"Chicago" was a very smoothly func-
tioning Convention, due to the cooperative
and coordinated efforts of the many
Chicago people who worked hard : George
Colburn for the Papers Program and the
Luncheon and Banquet and C. E. Hepp-
berger for the overall of Local Arrange-
ments. Supporting George Colburn on
the procuring of papers were Papers Com-
mittee Chairman Ed Seeley and Vice-
Chairmen J. E. Aiken, F. G. Albin,
G. G. Graham, R. O. Painter (for High-
Speed Photography), W. H. Rivers and
R. T. Van Niman, with some assists by
Editorial Vice-President John G. Ffayne.

To get the folks registered and oriented
is a real job at each convention and, though
high on the list of the many cares of the
Society's Convention Vice-President Bill
Kunzmann who is always on deck, regis-
tration is a major operation — at Chicago
well done under Chairman Jim Wassell
with a strong assist by Reid Ray and
help by Steve Hunter, Ken Mason,
Charles Nesbitt, Jack Powers and Paul
Ireland.

Public address and recording were the
bailiwick of Robert P. Burns who put and
kept the Society's equipment completely
under control to record a goodly lot of
convention discussion, while listening with
one ear to the Sound Dept. of Balaban &
Katz where he is Director.

developed, basic regulations and controls
have been created and improved as a
result of close cooperation with industry,
the Controlled Materials Plan has been
re-created, and is now working effectively,
and a system of priorities has been brought
into existence. While manv difficulties
have been encountered and mistakes made,
the job has been done. Military pro-
duction, industrial expansion, and all
defense supporting programs are moving
ahead at rapid rates. Moreover, the
organization, control techniques, and ex-
perience, all of which represent tremendous
resources, equal to the weapons of war,
will stand us in good stead should the
necessity for their increased use arise.
While all of us hope that this will not be
the case we must continue to emphasize
the importance of preparedness.

I. F. Jacobsen. Projection Supervisor
for Balaban & Katz, kept steady control
of the. convention projection, with E. W.
D'Arcy initially responsible for 16mm
projection plans. In charge of rounding:
up the motion picture shorts was L. E.
Weber, assisted bv R. J. Sherry.

An assist for television papers planning
was given by Wm. C. Eddy. Chairman for
hotel and transportation arrangements was
Wm. C. De Vry. Membership promotion
activities were headed by Col. S. R.
Todd who is the Membership Committee's
Vice-Chairman in Chicago. Ray Gallo,
Chairman of the Membership Committee,
was on hand with active meetings and
planning in the latter part of the week.

Publicity was handled on the spot by
Len Bidwell who kept a solid volume of
information going out through channels to
trade and general press, with some nice,
effective help from Mrs. Bidwell. Co-
hostesses for the Ladies' Program were
Mrs. C. E. Heppberger and Mrs. Geo.
W. Colburn who presented a highly
praised program of which one highlight
was a luncheon production of Miniature
Grand Opera by Fredrik A. Chramer
who presented Madame Butterfly, with
realistic puppets and with music recorded
by the Metropolitan Opera Company.

Committee meetings continued to be an

548

important part of the week's activities.
There were eight engineering committee
meetings which have been covered in the
May issue of the Journal under the usual
column "Engineering Activities." In
addition to the Membership Committee

Engineering Activities

Meeting mentioned above, there was a
papers and editorial meeting under the
aegis of Ed Seeley and John Frayne,
chiefly to pass on to Joe Aiken the main
mantle for the next convention — of
which much in detail in the next Journal.

Color As noted in the February 1952
Journal, the four-year limitation
term of office required appointment cf a
new chairman to the Color Committee.
To this end, Fred Bowditch, Engineering
Vice-President, recently appointed J. P.
Weiss of Du Pont to fill the post.

Television Studio The committee met
Lighting on June 4, 1952, and

further discussed spe-
cifications for incident light and brightness
meters, amending the former by placing
a 10% tolerance on the deviation from a
cosine response. Word was received from
Photo Research Corp. that they are
producing a brightness meter which meets
the committee specifications and the
committee is eagerly looking forward to
an opportunity to study and test it.

A subcommittee on terminology under
the chairmanship of Hank Gurin was
formed to start activity on this vital
project.

Glossary Glossary activity in the past
has depended primarily on
the activity of individuals and engineering
committees whose chief interest lay in
specialized subjects and projects. This
has produced much useful material but not
the industry-wide layman type of glossary
that is required. The Engineering Vice-
President has therefore established a
committee charged solely with the re-
sponsibility of bringing out a useful
glossary. Bill Offenhauser has accepted
the chairmanship of this committee and
is looking about for competent and eager
members. The Committee will first take
inventory of past work and then draw up
its plan of attack. It will correlate its
activities with similar activities of other
societies and call upon all branches of
industry and all SMPTE engineering
committees for any required assistance.
Those interested in participating in the

work of this new committee are asked to
so advise Bill Offenhauser or Hank Kogel
at Society Headquarters.

Television Film The main order of
Equipment business at the May

27, 1952, committee
meeting was the discussion of the film
recording and reproduction dimensions.
The committee has been attempting to
reach agreement on proposed standards
with no success as yet. Differences have
cropped up both on the East and West
Coasts. One claims an excessive loss cf
material in the kinerecording process;
the other is seeking to increase the size of
the reproduced cr scanned area. The
detailed discussion of these problems is
contained in SMPTE 461, a copy of which
is available upon request.

In addition, agreement was reached on
proposals regarding dimensions of slides
and opaques which must now go to a
letter ballot of the full committee.

ISO/TC 36 The Technical Committee
on Cinematography of the
International Organization for Standardi-
zation met as scheduled on June 9, rolled
up its sleeves and proceeded to do a very
workmanlike job. After establishing the
order of importance of the various items
on the agenda and extending the meeting
to three days instead of the two originally
planned, the committee broke up into
informal work groups for round-table dis-
cussion. The areas of agreement and
disagreement were thoroughly canvassed
(to the degree that time permitted) with
a considerable measure of success. The
delegates were agreed that an excellent
foundation had been laid for international
standards activity which could now be
fruitfully pursued through correspondence.
A more detailed report will be made avail-
able to the Society at a later date. — -Henry
Kogel, Staff Engineer

549

Current Literature

The Editors present for convenient reference a list of articles dealing with subjects cognate
to motion picture engineering published in a number of selected journals. Photostatic
or microfilm copies of articles in magazines that are available may be obtained from The
Library of Congress, Washington, D.G., or from the New York Public Library, New
York, N.Y., at prevailing rates.

American Cinematographer

vol. 33, Feb. 1952
Technicolor Cameras Now Ride the RO

Crane (p. 65) A. Rowan
Stereoscopic Movies With Any 16mm

Camera (p. 72) /. Forbes

vol. 33, Mar. 1952
"The Wild North" Introduces MGM's

New Ansco Color Process (p. 106) A.

Rowan
Stereoscopic Motion Pictures, Pt. II (p.

110) /. A. Norling

New Resolving Power Test Chart (p. Ill)
Bell & Howell Introduces 16mm Magnetic

Recorder-Projector (p. 112) E. C.

Hajduk

British Kinematography

vol. 20, Mar. 1952
A Projector-Family Programme — The

FP. 7 Projector — (p. 66) /. /. Kotte
Filming Radar (p. 77) /. R. F. Stewart
Exposure for Colour (p. 83) /. H. R. Coote

International Projectionist

vol. 27, Feb. 1952

Illuminated Screen Surround? No;
Rounded Masking Corners? Yes.
(p. 18) R. A. Mitchell

vol. 27, Mar. 1952
Better Film Care Improves Entertainment,

Pt. I (p. 7)
LaVezzi's Newly-Designed Intermittent

Movement (p. 16)

vol. 27, April 1952

What's New in Projection Screens (p. 5)
Three-dimensional Projection in Europe

(p. 9)
Better Film Care Improves Entertainment,

Pt. II (p. 10)

Kino-Technik

no. 3, 1952
Storschallgeber rings um das Mikrophon

(p. 56) E. Leistner
Die Internationale Spielfilm — Produktion

aus zwanzig Jahren (p. 50)
Schmalfilm — Theater maschine "Leitz

G 1" in drei Typen (p. 58) E. May

Der Farbfilm im Atelier, in der Kopier-
anstalt und im Theater (p. 63) W.
Behrendt

16mm-Tonfilmprojektoren auf dem
deutschen Mark( p. 66)

no. 4, April 1952
Ein neues Karnerasystem fur Stereoauf-

nahmen (p. 74) H. Luscher
Welche Anforderungen stellt das Fern-

sehen an den Rohfilm? (p. 80)
Eine leistungsfahige 1 6mm-Bilton-Kamera

(p. 84)

Proceedings of the l.R.E.

vol. 40, Feb. 1952

Improvements in Image Iconoscopes by
Pulsed Biasing the Storage Surface (p.
146) R. Theile and F. H. Townsend

Radio & Television News

vol. 47, April 1952

(Radio-Electronic Engineering Section)
High Speed Image Converter (p. 12) H.
Weil

RCA Review

vol. 13, Mar. 1952
Performance of the Vidicon, a Small

Developmental Television Camera Tube

(p. 3) B. H. Vine, R. B. Janes and F. S.

Smith
The RCA Color Television Camera Chain

(p. 11) /. D. Spradlin
Image Orthicon Color Television Camera

Optical System (p. 27) L. T. Sachtleben,

D. J. Parker, G. L. Allee and E. Kornstein.
The NBC New York Color Television

Field Test Studio (p. 107) /. R. DeBaun,

R. A. Monfort and A. A. Walsh

Tele-Tech

vol. 11, April 1952

A New All-Purpose Television Camera
(p. 38) A. Reisz

Tele- Vision Engineering

vol. 3, April 1952

Magnetic Sound and Negative Picture
Transmission (p. 10) R. Conner

550

New Members

The following members have been added to the Society's rolls since those last published.
The designations of grades are the same as those used in the 1952 MEMBERSHIP DIRECTORY.

Fellow (F) Active (M)

Honorary (H)

Abernathy, Lloyd B., Director of Photog-
raphy, Fotovox, Inc., 286 Monroe
Ave., Memphis, Tenn. (A)

Alsworth, Charles W., Jr., Chief, Cine-
matography, Edwards Air Force Base.
Mail: Box 124, Edwards, Calif. (A)

Andresen, Warren, Field, Studio Tech-
nical Director, KGOTV. Mail: 2420
Hilgard Ave., Berkeley 9, Calif. (A)

Aselstyne, John G., General Manager,
Benson-Wilcox Electric Co., 188 King
St., London, Ontario, Canada. (A)

Baran, Paul, Engineer, Audio-Video Prod-
ucts Co., 730 Fifth Ave., New York,
N.Y. (A)

Bates, Reginald, Engineer, John M. Wall,
Inc. Mail: 718 Irving Ave., Syracuse
10, N.Y. (M)

Bazaar, Walter T., Chemist, Signal Corps
Photo Center. Mail: 20-17—19 St.,
Long Island City 5, N.Y. (A)

Callen, Robert J., Audio Engineer, Glen
Glenn Sound Co. Mail: 3467 Adina
Dr., Hollywood 28, Calif. (M)

Capano, Dommick J., Stock Manager,
S.O.S. Cinema Supply Corp. Mail:
94 Knox PI., Staten Island 14, N.Y
(M)

Coston, Melvin L., Senior Photographer,
Humble Oil & Refining Co. Mail:
722 W. 42 St., Houston 18, Tex. (A)

Eagler, Paul, Process Cameraman. Mail:
2142 Prosser Ave., West Los Angeles
25, Calif. (M)

Frew, Patricia, University of Southern
California. Mail: 802 N. Kinglsey
Dr., Hollywood 29, Calif. (S)

Gardiner, J. H., Television Engineer,
National Broadcasting Co. Mail: 914
N. Isabel, Glendale 7, Calif. (A)

Gately, Frederick, First Cameraman,
Mark VII Productions. Mail: 522 S.
Barrington Ave., Los Angeles. (M)

Germain, J. Roger, Television Projec-
tionist, Canadian Broadcasting Corp.,
Box 6000, Montreal, Canada. (A)

Heiland, John G., Chief, Engineering
Laboratories, Bell & Howell Co. Mail:
1339 Center St., Des Plaines, 111. (M)

Hirasawa, Isao, Chief Engineer, Tokyo
Theatre Supply Co., Ltd. Mail: No.
86, Takaban-cho, Meguro-ku, Tokyo,
Japan. (A)

Huhndorff, Ervin P., Chief Engineer,
KPRC-TV, Lamar Hotel, Houston,
Tex. (A)

Associate (A)

Student (S)

Kragiel, Henry P., Chemist (Industrial),
Stanley Works. Mail: 97 Eastwick
Rd., New Britain, Conn. (A)

Lepage, John L., Electronics Technician,
National Research Council. Mail: 12
Pankdale Ave., Deep River, Ontario,
Canada. (A)

Lustberg, Stanley E., Television Camera-
man, Radio Belgrand. Mail: c/o
American Embassy, Buenos Aires, Ar-
gentina. (A)

Mangan, William J., Film Producer for
Television, National Television Guild.
Mail: 37-45—100 St., Corona 68,
N.Y. (M)

Mclnnes, Harold W., Radio Engineer,
International Broadcasting Co., Ltd.
Mail: 6560 East Hastings St., North
Burnaby, British Columbia. (A)

Merritt, Charles H., Engineering Section,
U.S. Dept. of State, 221 W. 57 St.,
New York 19, N.Y. (M)

Mowery, Raymond, Field Engineer, RCA
Service Co. Mail: 3705 Windom Rd.,
Brentwood, Md. (A)

Ott, John Nash, Jr., President, John Ott
Pictures, Inc., 85 Hibbard Rd.,
Winnetka, 111. (M)

Pesca, Frank, Junior Engineer, Federal
Manufacturing & Engineering Corp.
Mail: 2912—86 St., Brooklyn 23. (A)

Quentin, Charles F., Radio Engineer,
Cowles Broadcasting Co., KRNT.
Mail: 1120 Polk Blvd., Des Moines 11,
Iowa. (A)

Rasmussen, Hans Christian Erik, Chief
Sound Engineer, Cia: Cinematografica
Vera Cruz, 311, Rua Major Diogo,
Sao Paulo, Brazil. (A)

Sharpe, Robert K., Brown University.
Mail: 90 Crescent Dr., Glencoe, 111.
(S)

Sidlo, Thomas C., District Engineer, Lamp
Div., General Electric Co. Mail: 230
S. Clark St., Rm. 1233, Chicago. (A)

Usuf, Mohammad, Engineer, Western
Electric Co. (Near East), Karachi,
Pakistan. (A)

Walker, Alberto W., Territorial Manager,
Loew's International Corp., 1540 Broad-
way, New York, N.Y. (A)

CHANGES IN GRADE

Greenfield, J. C., (A) to (M)
Tickes, Samuel, (A) to (M)
Wilson, Brown, (A) to (M)

551

New Products

Further information about these items can be obtained direct from the addresses given.
As in the case of technical papers, the Society is not responsible for manufacturers' state-
ments, and publication of these items does not constitute endorsement of the products.

holds the reading until the button is again
depressed.

The sensitivity of the meter is broad,
having an overall range from 1 to 1,000,000-
ft-L over its five ranges. A logarithmic
scale meter is used to accomplish a per-
centage cf accuracy about the same over
the entire scale. The photocell is filtered
so that its response closely approximates
visual response.

Photo Research Corp. has designed the
meter with the idea that it will be most
useful for measuring brightness from a
given position, such as the camera position
in cinematography or television, and in
measuring screen brightness.

This new Spectra Brightness Spot Meter
will soon be available from Photo Research
Corp., 127 W. Alameda Ave., Burbank,
Calif. It is designed to measure the
brightness of a very small area at any
distance from 4 ft to infinity, through the
use of a vacuum p'lotDcell, amplifier and
microammeter. All users obtain the same
reading of a given area, for the meter is
independent of the sensitivity of the
observer's eye and requires no matching
of brightness.

The size of the area measured is 1°,
that is a 1-in. spot at 5 ft, 2-in. spot at
10 ft, etc. The subject to be measured is
viewed at considerable magnification
through the telescopic sight built into the
meter, a circle in the center of the field
of view indicating the actual area being
measured. The meter is self-contained,
having no external power source. It
weighs about 5 Ib.

To allow handholding the meter and
still reading the brightness, a locking-type
microammeter is used. The operator pushes
a button on the side when the desired
area is in the reticule circle and then
releases the button again. The meter

A new 16mm lens series has been an-
nounced by the Kinoptik Company of
Paris, France, which for several years
has had a 35mm series for motion picture
and television cameras. Available in
C-mounts for standard 16mm motion
picture cameras are 20-, 25-, 32-, 50- and
75-mm Kinoptik lenses. They have 7
as well as / scales, a new system of equi-
distant aperture markings, and a six-
element design. Shown here is a Kinoptik
for 16mm, 1-in., //2 Apochromat in
focusing C-mount. Complete descriptions
of both the 35mm and 16mm lenses are
available from the U.S.A. distributor,
Victor Kayfetz, 130 E. 56 St., New York
22, N.Y.

552

Lantern Slides and How to Make Them of the Biological Photographic Association;
is a 37-page reprint booklet of which 26 (2) "Homemade Slides by Photographic
pp. are from See and Hear magazine and Methods" and (3) "Filing Opaque Pro-
written by Mary Esther Brooks of the jection Material" by Harold F. Bernhardt
Bureau of Audio-Visual Aids at Indiana from The Educational Focus.
University. In these pages are a wealth

of details about techniques, materials and Copies are available, preferably in

sources, arranged in four chapters antl a orders of 50 copies or more, at 20 cents per

bibliography. Additional chapters are: copy from the Educational Sales Division.

(1) "Letter Height and Legibility" by Bausch & Lomb Optical Co., 786 St.

R. A. Sage and reprinted from the Journal Paul St., Rochester 2, N.Y.

Back issues of the Journal available: 5 years (1947-51) in perfect condition plus the
indexes for 1936-45 and 1946-50 and including the 1949 High-Speed Photography,
upon any reasonable offer to Vic Gretzinger, 3547 Suter St., Oakland 19, Calif.

Journals and Transactions available: Of the Transactions Nos. 11, 14, 20, 21, 23, 25,
27, 28 and 38; and 22 years of the Journal (1930-1951) except for Jan., Feb., Mar. and
Apr. of 1934, Jan. and Apr. of 1948, and Feb. 1950; also these extra single copies — Nov.
1930; Jan., Feb., July and Nov. 1931; June 1932; Mar. and Apr. 1933; Dec. 1934;
Jan. and May 1935; Oct. 1938; July and Dec. 1940; Oct. 1948 and Jan. 1950. These
are available upon any reasonable offer made to : Paul J. Larsen, Assistant to the Presi-
dent, Borg^Warner Corp., 310 So. Michigan Ave., Chicago 4, 111.

Meetings

72nd Semiannual Convention of the SMPTE, Oct. 6-10, Hotel Statler,

Washington, D. C.

Other Societies

National Audio-Visual Association, Convention and Trade Show, Aug. 2-5, Hotel

Sherman, Chicago, 111.

University Film Producers Association, Annual Meeting, Aug. 11-15, Syracuse Univer-
sity, Syracuse, N. Y.

Photographic Society of America, Annual Convention, Aug. 12-16, Hotel New Yorker,

New York

American Institute of Electrical Engineers, Pacific General Meeting, Aug. 19-22, Hotel

Westward Ho, Phoenix, Ariz.

International Society of Photogrammetry, Conference, Sept. 4-13, Hotel Shoreham,

Washington, D.C.

American Standards Association, Third National Standardization Conference, Sept.

8-10, Museum of Science and Industry, Chicago, 111.

Illuminating Engineering Society, National Technical Conference, Sept. 8-12, Edge-
water Beach Hotel, Chicago, 111.

Biological Photographic Association, Annual Meeting, Sept. 10-12, Hotel New Yorker,

New York

National Electronics Conference, Annual Meeting, Sept. 29-Oct. 1, Sherman Hotel,

Chicago, 111.

Optical Society of America, Oct. 9-11, Hotel Statler, Boston, Mass.

American Institute of Electrical Engineers, Fall General Meeting, Oct. 13-17, New

Orleans, La.
American Standards Association, Annual Meeting, Nov. 19, Waldorf-Astoria, New York

553

Papers Presented

at the Chicago Convention, April 21-25

BY SESSIONS

MONDAY AFTERNOON — Television Session

Robert E. Lewis, Armour Research Foundation, Chicago, 111., "A Color or Stereoscopic

Frame-Sequential Television Viewer."

Sam H. Kaplan, Consultant, Chicago, 111., "Theory of Parallax Barriers."
A. D. Fowler and H. N. Christopher, Bell Telephone Laboratories, Murray Hill, N.J.,

"Effective Sum of Multiple Echoes in Television."
Fred Barton and H. J. Schlafly, TelePrompter Corp., New York, "TelePrompter, New

Production Tool."
J. A. Norling (Committee Chairman), Loucks and Norling Studios, New York, "Report

of Stereoscopic Motion Pictures Committee."

MONDAY EVENING — Television Session

Nathan L. Halpern, Theater Network Television, Inc., New York, "Theater Television

Progress."
M. C. Banca, RCA Victor Division, Industrial Equipment Section, Camden, N.J.,

"Industrial Television."
Victor Trad and Ricardo Muniz, Trad TV Corp., Asbury Park, N.J., "Dual Theater

Television System."
John M. Sims, General Precision Laboratory, Pleasantville, N.Y., "Installing GPL

Video Film in Denver's Broadway Theater."
W. H. Offenhauser, Jr., Consultant, New Canaan, Conn., "Nomenclature for Motion

Pictures and Television."

TUESDAY MORNING — Screens and Control of Brightness

D. R. White (Committee Chairman), E. I. du Pont de Nemours & Co., Inc., Parlin,
N.J., "Report on International Standardization."

Charles W. Handley (Committee Chairman), National Carbon Co., Inc., Los Angeles,
Calif., "Report of the Progress Committee."

Arthur J. Hatch, Strong Electric Corp., Toledo, Ohio, "A-C High-Intensity Arc Slide
Projector."

Benjamin Schlanger and William A. Hoffberg, Theater Consultants, New York, and
Charles R. Underhill, Jr., RCA Victor Division, Camden, N.J., "The Synchro-
Screen as a Stage Setting for Motion Picture Presentation."

554

W. W. Jennings, W. Wheeler Jennings, Chicago, 111., and Pierre Vanet, A. Mattey,
Paris, France, "A New Direct- Vision Stereo-Projection Screen."

Ellis W. D'Arcy and Gerhart Lessman, De Vry Corp., Chicago, 111., "Objective Evalua-
tion of Projection Screens."

H. B. Brueggemann, Cinecolor Corp., Burbank, Calif., "Continuous Arc Projector Light
Meter."

TUESDAY AFTERNOON — Armed Forces Production

Max Beard and A. M. Erickson, Naval Ordnance Laboratory, Silver Spring, Md., "An

Auditorium Specifically Designed for Technical Meetings."
Phillip M. Cowett, U.S. Navy, Bureau of Ships, Washington, D. C., "16mm Motion

Picture Theater Installations Aboard Naval Vessels."
W. R. Cronenwett, U.S. Navy, U.S. Naval Photographic Center, Anacostia, D.C.,

"The Navy's Training Film Production Program."
Charles F. Hoban and James A. Moses, Motion Picture Branch, Army Pictorial Service

Division, Washington, D.C., "Cameo Film Production Technique."
P. C. Foote, Bell & Howell Co., Chicago, 111., and R. E. Miesse, General Scientific Co.,

Chicago, 111., "Military-Type Lenses for 35mm Motion Picture Cameras."

TUESDAY EVENING — Magnetic Protection; Film Inspection

M. G. Townsley, H. H. Brauer, J. P. Weber and F. J. Schuessler, Bell & Howell Co.,
Chicago, 111., "New Magnetic and Optical 16mm Sound Projector."

Robert Grunwald, Harwald Co., Evanston, 111., "The Inspect-O-Film Machine."

Carl E. Hittle, RCA Victor Division, Hollywood, Calif., "Automatic Torque Controller
for Torque Motors."

E. W. D'Arcy and J. S. Powers, De Vry Corp., Chicago, 111., "Magnetic Sound Applica-
tion to 16mm Armed Forces Projectors."

WEDNESDAY MORNING — High-Speed Photography Session

D. Muster and E. G. Volterra, Illinois Institute of Technology, Chicago, 111., "Rotating-

Drum Camera for High-Speed Experimental Work."
Brian O'Brien, Gordon Milne and William Covell, University of Rochester, Rochester,

N.Y., "Automatic Printing of High-Speed Image-Dissection Negatives."
J. L. Tupper, Eastman Kodak Co., Rochester, N.Y., "Practical Aspects of Reciprocity

Law Failure."
N. W. Rodelius, Armour Research Foundation, Chicago, 111., and Eugene L. Perrine,

Zonolite Research Laboratories, Evanston, 111., "Methods of Improving Visibility

in High-Speed Photography.'

WEDNESDAY AFTERNOON — High-Speed Photography Session

C. H. Winning, Du Pont Eastern Laboratory, Gibbstown, N.J., and H. E. Edgerton,
Massachusetts Institute of Technology, Cambridge, Mass., "Explosive Argon Flash-
lamp."

James Cooper, University of Michigan, Aero Research Laboratories, Willow Run Airport,
Mich., "Photography in the Wind Tunnel, Including Schlieren and Flame Photog-
raphy."

Ralph P. Sledge, International Harvester Co., Memphis, Tenn., "Portable Generating
Equipment for Use With the Fastax Camera."

555

THURSDAY AFTERNOON — Color and Laboratory Session

John Stott (Committee Chairman), Du-Art Film Laboratories, Inc., New York, "Labora-
tory Practice Committee Report."

Robert C. Lovick, Eastman Kodak Co., Rochester, N. Y., "Exposure of Kodachrome
Sound Tracks for Optimum Quality.

Robert C. Lovick, Eastman Kodak Co., Rochester, N.Y., "Densitometry of Silver Sulfide
Sound Tracks."

F. P. Herrnfeld, Frank Herrnfeld Engineering Corp., Culver City, Calif., "Integrating-
Type Color Densitometer."

A. A. Duryea, T. J. Gaski and L. Mansfield, Pathe Laboratories, Inc., New York, "Nega-
tive-Positive Color Processing by Pathe."

THURSDAY EVENING — General Session

Mauro Zambuto, Scalera Films, Rome and Venice, Italy, "Foreign Language Dubbing."
Thomas T. Hill, The Edwal Laboratories, Inc., Ringwood, 111., "Nonsilver Photographic

Processes."
H. L. Baumbach, Paramount Pictures Corp., Hollywood, Calif., "Effect of pH Upon

Photographic Developer Activity."
John K. Hilliard (Committee Chairman), Altec Lansing Corp., Beverly Hills, Calif.,

"Sound Committee Report."
Otto Bixler, Magnecord, Inc., Chicago, 111., "Commercial Binaural Tape Recorder."

FRIDAY MORNING — Sound and Editing Session

Chester E. Beachell, The National Film Board of Canada, Ottawa, Ontario, Canada,

"Simplified Preamplifier With High Gain, Low Distortion, and Exceptional Dynamic

Range and Frequency Response."
O. L. Dupy, Metro-Goldwyn-Mayer Sound Dept., Culver City, Calif., "A Method of

Direct-Positive Variable-Density Recording With the Light Valve."
Chester E. Beachell and G. G. Graham, National Film Board of Canada, Ottawa, Ontario,

Canada, "Dual-Purpose Optical Sound Prints."

R. M. Savini, Editola Corp. of America, New York, "The EditolaFilm Editing Machine."
Richard H. Ranger, Rangertone, Inc., Newark, N.J., "Tape-to-Film Editor."
Richard H. Ranger, Rangertone, Inc., Newark, N.J., "A Versatile Camera Car and
Recording Unit."

FRIDAY AFTERNOON — New Equipments Session

A. L. Holcomb, Westrex Corp., Hollywood, Calif., "Three-Phase Power From Single-
Phase Source."

Chester E. Beachell, The National Film Board of Canada, Ottawa, Ontario, Canada,
"Multiple-Camera Control and Automatic Synchronizing System."

Benjamin Berg, Benjamin Berg Agency, Hollywood, Calif., "Professional Motion Picture
Camera Which Utilizes Both 16mm and 35mm Film."

Lee R. Richardson and William N. Gaisford, Richardson Camera Co., Hollywood, Calif.,
"Follow-Focus Device and Camera Blimp for 16mm Professional Camera."

Willy Borberg, General Precision Laboratory, Pleasantville, N.Y., "Buckle Reduction
in 35mm Film."

SMPTE Officers and Committees: The roster of Society Officers and the
Committee Chairmen and Members were published in the April Journal.

556

INDEX TO SUBJECTS

January — June 1952 • Volume 58

ACOUSTICS

Recent Improvements in Silencing Engine-
Driven Generators, L. D. Grignon

Jan. pp. 43-52

ARCS

Studio Lighting

Recent Improvements in Silencing Engine-
Driven Generators, L. D. Grignon

Jan. pp. 43-52

BIOGRAPHICAL NOTE

Downes, A. C. Mar. p. 266

BOOK REVIEWS

Agfacolor Process, a Short Bibliography, com-
piled by Alexis N. Vorontozoff (Re-
viewed by Lloyd E. Varden)

May p. 462

Transmitting Valves, by J. P. Heyboer and
P. Zijlstra (Reviewed by Richard H.
Dorf) May p. 462

Application oj the Electronic Valve in Radio
Receivers and Amplifiers (Vol. II), by B.
G. Dammers, J. Haantjes, J. Otte and
H. Van Suchtelen (Reviewed by Richard
H. Dorf) May p. 461

Television Principles, by Robert B. Dome
(Reviewed by Otis S. Freeman)

May p. 461

liases Techniques de la Television, by H.
Delaby (Reviewed by S. W. Athey)

May p. 460

Basic Electron Tubes, by Donovon V. Gep-
pert (Reviewed by Harry R. Lubcke)

May p. 460

IES Lighting Handbook (2d ed.} (Reviewed
by M. S. Wright) May p. 459

Standards for Single -Line Diagrams

Apr. p. 361

Dynamics oj the Film, by Joseph and Harry
Feldman (Reviewed by George L.
George) Apr. p. 361

Prism and Lens Making (2d ed.}, by F.
Twyman (Reviewed by R. Kingslake)
Apr. p. 360

Television Engineering (2d ed.), by D. G.
Fink (Reviewed by Mclntosh & Inglis)
Apr. p. 359

Motion Pictures, 1912-1939, Library of
Congress Catalog Mar. p. 268

The Television Program — Its Writing, Direc-
tion, and Production, by Edward Stasheff
and Rudy Bretz (Reviewed by Dik
Darley) Mar. p. 268

Einfilhrung in die wissenschajtliche Kinemato-
graphie, by Dr. Werner Faasch (Reviewed
by Walter Clark) Feb. p. 176

Fundamental Mechanisms of Photographic Sensi-
tivity, Edited by J. W. Mitchell (Reviewed
by Herman H. Duerr) Feb. p. 176

Acoustical Terminology, sponsored by ASA
and IRE Jan. p. 78

The Little Fellow, the Life and Work of Charlie
Chaplin, by Peter Cotes and Thelma
Niklaus Jan. p. 78

Charlie Chaplin, by Theodore Huff

Jan. p. 77

The Film Industry in Six European Countries,
by Film Centre, London Jan. p. 77

The Indian Film, by Panna Shah (Reviewed
by Raymond Spottiswoode) Jan. p. 77

Three-Dimensional Photography: The Principles
oj Stereoscopy, by Herbert C. McKay
(Reviewed by J. A. Norling)

Jan. p. 76

CHEMICAL CORNER

Mar. p. 272

CINEMATOGRAPHY (see also HIGH-
SPEED PHOTOGRAPHY)

Resolution Test Chart of the Motion Pic-
ture Research Council, Armin J. Hill

June pp. 529-530

The Nature and Evaluation of the Sharp-
ness of Photographic Images, G. C.
Higgins and L. A. Jones

Apr. pp. 277-290

COLOR

Printer Control in Color Printing, C. A.

Horton Mar. pp. 239-244

Color Negative and Color Positive Film

for Motion Picture Use, W. T. Hanson,

Jr. -Mar. pp. 223-238

Cinecolor Multilayer Color Developing

Machine, J. W. Kaylor and A. V. Pesek

Jan. pp. 53-60

Color Television Reproducers, Harry R.

Lubcke Jan. pp. 22-27

June 1952 Journal of the SMPTE Vol. 58

557

CURRENT LITERATURE

June p. 550 Mar. p. 269

FILM
General

The Ansco Color Negative-Positive Process,
Herman H. Duerr June pp. 465-479

The Nature and Evaluation of the Sharp-
ness of Photographic Images, G. C.
Higgins and L. A. Jones

Apr. pp. 277-290

Color Negative and Color Positive Film for

Motion Picture Use, W. T. Hanson, Jr.

Mar. pp. 223-238

Image Gradation, Graininess and Sharp-
ness in Television and Motion Picture
Systems — Part II: The Grain Struc-
ture of Motion Picture Images — An
Analysis of Deviations and Fluctuations
of the Sample Number, Otto H. Schade
Mar. pp. 181-222

Educational, Documentary and Training

University Film Producers Association

May p. 454

Audio-Visual Instruction Conference, D.
F. Lyman May pp. 445-449

Film Production Principles — The Subject
of Research, Ken Kendall

May pp. 428-444

A Scientific Approach to Informational-
Instructional Film Production and Utili-
zation, C. R. Carpenter and L. P.
Greenhill May pp. 415-427

GENERAL

University Film Producers Association

May p. 454

Desirable Characteristics of 16mm Enter-
tainment Film for Naval Use, Lowell O.
Orr and Philip M. Cowett

Mar. pp. 245-258

Film-Spool Drive With Torque Motors,
A. L. Holcomb Jan- PP- 28-35

HIGH-SPEED PHOTOGRAPHY
Applications

Multiple-Image Silhouette Photography
for the NOTS Aeroballistics Laboratory,
Ernest C. Barkofsky June pp. 480-486

Techniques for Effective High-Speed
Photography and Analysis, Richard O.
Painter May pp. 373-384

Cameras

Optical Problems in High-Speed Camera
Design, John C. Kudar

June pp. 487-490

High-Speed Motion Picture Cameras From
France, Paul M. Gunzbourg

Mar. pp. 259-265

High-Constant-Speed Rotating Mirror, J.
W. Beams, E. C. Smith and J. M.
Watkins Feb. pp. 159-168

LABORATORY PRACTICE

General

American Standard Splices for 8mm
Motion Picture Films, PH22.77-1952

June p. 541

American Standard Splices for 16mm

Motion Picture Films for Projection,

PH22.24-1952 June p. 539

Laboratory Practice Committee Report,

John G. Stott, Chairman

June pp. 531-534

The Nature and Evaluation of the Sharp-
ness of Photographic Images, G. C.
Higgins and L. A. Jones

Apr. pp. 277-290

Color Negative and Color Positive Film for
Motion Picture Use, W. T. Hanson, Jr.

Mar. pp. 223-238

Image Gradation, Graininess and Sharp-
ness in Television and Motion Picture
Systems — Part II: The Grain Struc-
ture of Motion Picture Images — An
Analysis of Deviations and Fluctuations
of the Sample Number, Otto H. Schade
Mar. pp. 181-222

Factors Affecting the Quality of Kine-
recording, P. J. Herbst, R. O. Drew
and J. M. Brumbaugh

Feb. pp. 85-104

Cinecolor Multilayer Color Developing

Machine, J. W. Kaylor and A. V. Pesek

Jan. pp. 53-60

Film-Spool Drive With Torque Motors,
A. L. Holcomb Jan. pp. 28-35

Printing

Proposed American Standard Enlarge-
ment Ratio for 16mm to 35mm Optical
Printing, PH22.92 Jan. p. 71

Printer Control in Color Printing, C. A.
Horton Mar. pp. 239-244

Prints From 16mm Originals, R. L.
Sutton, K. B. Curtis and Lloyd Thomp-
son Feb. pp. 145-158

LIGHTING (see also ARCS and HIGH-
SPEED PHOTOGRAPHY)

General

The Ansco Color Negative-Positive Process,
Herman H. Duerr June pp. 465-479

Factors Affecting the Quality of Kine-
recording, P. J. Herbst, R. O. Drew
and J. M. Brumbaugh

Feb. pp. 85-104

Projection

Heat-Transmitting Mirror, G. L. Dimmick
and M. E. Widdop Jan. pp. 36-42

558

June 1952 Journal of the SMPTE Vol. 58

Studio

Recent Improvements in Silencing Engine-
Driven Generators, L. D. Grignon

Jan. pp. 43-52

MOTOR-DRIVE SYSTEMS

Film-Spool Drive With Torque Motors,
A. L. Holcomb Jan. pp. 28-35

NEW PRODUCTS

Lantern Slides and How to Make Them,
Bausch & Lomb Optical Co.

June p. 553

1 6mm Lens Series, Kinoptik Company and
Victor Kayfetz June p. 552

Spectra Brightness Spot Meter, Photo
Research Corp. June p. 552

Fundamentals of Magnetic Recording, Audio
Devices, Inc. May p. 464

The Tener, Mole-Richardson Co.

May p. 464

Common Causes of Damage to 35mm Release
Prints (Revised Edition), Eastman Kodak
Company Apr. p. 365

Silent Magnetic Splicer, Unusual Films

Apr. p. 365

Ultra-High-Speed Camera, Battelle In-
stitute Apr. p. 364

The 1952 Catalog of Films From Britain

Mar. p. 276

The Utility Television Monitor Model
CA16, Conrac, Inc. Mar. p. 276

The Bell & Howell Filmosound 202 16mm

Optical-Magnetic Recording Projector

Mar. p. 275

A Light-Weight Sound-Proof Blimp for
the Arriflex Camera, [correction from
May p. 464 — Kling Photo Supply
Corp.] Mar. p. 274

Developmental Transistor, RCA-Victor
Div. Mar. p. 274

The Berkshire Labstrobe, Model 18,
Berkshire Laboratories Mar. p. 274

Training Aid Guides, published by Film
Research Associates Feb. p. 180

"Wagner-16" MicroDisc Recorder, Model
PI 6-450, Audio and Video Products
Corp. Feb. p. 180

The 1100 Series Portable Magnetic System,
Westrex Corp. Jan. p. 84

TV Camera Car, The Camera Mart, Inc.

Jan. p. 83

The Aminco Photomultiplier Microphotom-
eter, American Instrument Company,
Inc. Jan. p. 82

OBITUARIES

Pacent, Louis Gerard May p. 455

OPTICS

Resolution Test Chart of the Motion
Picture Research Council, Armin J.
Hill June pp. 529-530

The Nature and Evaluation of the Sharp-
ness of Photographic Images, G. C.
Higgins and L. A. Jones

Apr. pp. 277-290

Image Gradation, Graininess and Sharp-
ness in Television and Motion Picture
Systems — Part II: The Grain Struc-
ture of Motion Picture Images — An
Analysis of Deviations and Fluctuations
of the Sample Number, Otto H. Schade
Mar. pp. 181-222

Continuous Motion Picture Projector for

Use in Television Film Scanning, A. G.

Jensen, R. E. Graham and C. F. Mattke

Jan. pp. 1-21

PRODUCTION

TelePrompter — A New Production Tool,
Fred Barton and H. J. Schlafly

June pp. 515-521

Film Production Principles — The Subject
of Research, Ken Kendall

May pp. 428-444

A Scientific Approach to Informational-
Instructional Film Production and Utili-
zation, C. R. Carpenter and L. P.
Greenhill May pp. 415-427

A Technical Solution of Magnetic Record-
ing Cost Reduction, Kurt Singer and
H. Connell Ward Apr. pp. 329-340

Desirable Characteristics of 16mm Enter-
tainment Film for Naval Use, Lowell
O. Orr and Philip M. Cowett

Mar. pp. 245-258

PROGRESS COMMITTEE REPORT

By Charles W. Handley, Chairman

May pp. 397-409

PROJECTION

16mm and 8mm

American Standard for 16mm Motion

Picture Projection Reels, PH22.11-1952

June p. 535

Optical-Magnetic Sound 16mm Projector,
G. A. del Valle and F. L. Putzrath

Apr. pp. 312-322
35mm

Proposed American Standard Screen
Brightness for 35mm Motion Pictures,
PH22.39 May p. 452

The Cash Customers at the Festival of
Britain Telecinema, Norman Jenkins

Apr. pp. 304-311

Heat-Transmitting Mirror, G. L. Dimmick
and M. E. Widdop Jar. pp. 36-42

PULL-DOWN MECHANISMS

Continuous Motion Picture Projector for

Use in Television Film Scanning, A. G.

Jensen, R. E. Graham ano C. F. Mattke

Jan. pp. 1-21

Index to Subjects

559

SCREEN BRIGHTNESS

Proposed American Standard Screen
Brightness for 35mm Motion Pictures,
PH22.39 May p. 452

Heat-Transmitting Mirror, G. L. Dimmick
and M. E. Widdop Jan. pp. 36-42

SCREENS

The Synchro-screen as a Stage Setting for
Motion Picture Presentation, B. Schlan-
ger, W. A. Hoffberg and C. R. Under-
hill, Jr. June pp. 522-528

SOCIETY ACTIVITIES
General

Designer of New Society Symbol

Jan. p. 81
Discussions in the Journal Jan. p. 74

Awards and Citations

Honorary Members Apr. p. 358

SMPTE Honor Roll Apr. p. 358

Regulations and Former Recipients (of

four awards) : Journal, Progress Medal,

Samuel L. Warner Memorial, and

David Sarnoff Gold Medal

Apr. pp. 355-358

Board Meeting Feb. p. 174

Committees

Listing and Personnel: Apr. pp. 366-372

Reports:

Laboratory Practice, John G. Stott,

Chairman June pp. 531-534

Television Studio Lighting, Richard

Blount, Chairman May pp. 450-451

Progress, Charles W. Handley, Chairman

May pp. 397-409
SMPTE Standards Committee, Frank E.

Carlson, Chairman Feb. pp. 169-172

Constitution and Bylaws

Apr. pp. 341-348
Convention

71st, Chicago, 111., Papers Presented,

June pp. 554-556

Report June pp. 543-549

Announcement: Mar. p. 267; Feb. p.
173; Jan. pp. 73-74

Engineering Activities (News and Brief
Reports)

June p. 549; May p. 453; Feb. p. 175;
Jan. p. 75

Financial Reports Apr. pp. 352-354

Membersiip and Subscriptions

Membership Directory, Part II, May
New Members:

June p. 5M; May p. 456; Apr. p. 362;

Mar. p. 27); Feb. p. 177; Jan. p. 79

New Membership Directory (plans)

Jan. p. 81
Reports Apr. p. 354

Nominations

1952 Nominations

Apr. p. 359

Officers and Governors of the Society

Apr. pp. 349-351

SOUND RECORDING
General

New Principle for Electronic Volume
Compression, Harold E. Haynes

Feb. pp. 137-144

Magnetic

Magnetic Print-Through — Its Measure-
ment and Reduction, Lyman J. Wiggin
May pp. 410-414

A Technical Solution of Magnetic Re-
cording Cost Reduction, Kurt Singer
and H. Connell Ward

Apr. pp. 329-340
Filte

Twin-Drum Film-Drive Filter System for
Magnetic Recorder-Reproducer, Carl
E. Hittle Apr. pp. 323-328

Optical-Magnetic Sound 16mm Projector,
G. A. del Valle and F. L. Putzrath

Apr. pp. 312-322

Magnetic Sound Track Placement, Loren
L. Ryder and Bruce H. Denney

Feb. pp. 119-136

Multichannel Magnetic Film Recording
and Reproducing Unit, C. C. Davis,
J. G. Frayne and E. W. Templin

Feb. pp. 105-118

New Magnetic-Recording Head, Marvin
Camras Jan. pp. 61-66

Photographic

Push-Pull Direct-Positive Recording — An
Auxiliary to Magnetic Recording, L. I.
Carey and Frank Moran

Jan. pp. 67-70
Re-recording

Optical-Magnetic Sound 16mm Projector,
G. A. del Valle and F. L. Putzrath

Apr. pp. 312-322

SOUND REPRODUCTION

General

Twin-Drum Film-Drive Filter System for
Magnetic Recorder-Reproducer, Carl
E. Hittle Apr. pp. 323-328

Multichannel Magnetic Film Recording
and Reproducing Unit, C. C. Davis,
J. G. Frayne and E. W. Templin

Feb. pp. 105-118

Loudspeakers

Optical-Magnetic Sound 16mm Projector,
G. A. del Valle and F. L. Putzrath

Apr. pp. 312-322

560

June 1952 Journal of the SMPTE Vol. 58

STANDARDS and RECOMMENDA-
TIONS: See the listing on p. 562 or
the specific subject heading.

Report of SMPTE Standards Committee,
Frank E. Carlson, Chairman

Feb. pp. 169-172

STEREOSCOPY

The Cash Customers at the Festival of
Britain Telecinema, Norman Jenkins

Apr. pp. 304-311

Progress in Three-Dimensional Films at
the Festival of Britain, Raymond Spottis-
woode Apr. pp. 291-303

TELEVISION (see also LIGHTING —
Studio, and THEATER TELE-
VISION)

General

The Image Iconoscope — A Camera Tube
for Television (Abstracted by Pierre
Mertz), P. Schagen, H. Bruining and
J. C. Francken June pp. 501-514

Continuous Motion Picture Projector for

Use in Television Film Scanning, A. G.

Jensen, R. E. Graham and C. F. Mattke

Jan. pp. 1-21

Color

Color Television Reproducers, Harry R.
Lubcke Jan. pp. 22-27

Film Recording

Factors Affecting the Quality of Kine-
recording, P. J. Herbst, R. O. Drew
and J. M. Brumbaugh

Feb. pp. 85-104

Picture Quality

Effective Sum of Multiple Echoes in Tele-
vision, A. D. Fowler and H. N. Christo-
pher June pp. 491-500

The Nature and Evaluation of the Sharp-
ness of Photographic Images, G. C.
Higgins and L. A. Jones

Apr. pp. 277-290

Image Gradation, Graininess and Sharp-
ness in Television and Motion Picture
Systems — Part II: The Grain Struc-
ture of Motion Picture Images — An
Analysis of Deviations and Fluctuations
of the Sample Number, Otto H. Schade
Mar. pp. 181-222

Studio Production

TelePrompter — A New Production Tool,
Fred Barton and H. J. Schlafly

June pp. 515-521

THEATER TELEVISION

A Direct-Projection System for Theater
Television, F. N. Gillette

May pp. 385-396

The Cash Customers at the Festival of
Britain Telecinema, Norman Jenkins

Apr. pp. 304-311

Index to Subjects

561

American Standards — by numbers

JVo. Title Page, issue

PH22.1 1-1952 16mm Motion Picture Projection Reels 535, June

PH22.24-1952 Splices for 16mm Motion Picture Films for Projection 539, June

PH22.39 Proposed, Screen Brightness for 35mm Motion Pictures 452, May

PH22.77-1952 Splices for 8mm Motion Picture Films 541, June

PH22.92 Proposed, Enlargement Ratio for 16mm and 35mm 71, Jan.
Optical Printing

562 June 1952 Journal of the SMPTE Vol.58

INDEX TO AUTHORS

January — June 1952 • Volume 58

Barkofsky, Ernest C., Multiple-Image
Silhouette Photography for the NOTS
Aeroballistics Laboratory

June pp. 480-486

Barton, Fred, and Schlafly, H. J., Tele-
Prompter — A New Production Tool

June pp. 515-521

Beams, J. W., Smith, E. C., and Watkins,
J. M., High-Constant-Speed Rotating
Mirror Feb. pp. 159-168

Blount, Richard, Chairman, Television
Studio Lighting Committee Report

May pp. 450-451

Bruining, H., Francken, J. C., and
Schagen, P., The Image Iconoscope —
A Camera Tube for Television (Ab-
stracted by Pierre Mertz)

June pp. 501-514

Brumbaugh, J. M., Herbst, P. J., and
Drew, R. O., Factors Affecting the
Quality of Kinerecording

Feb. pp. 85-104

Camras, Marvin, New Magnetic-Record-
ing Head Jan. pp. 61-66

Carey, L. I., and Moran, Frank, Push-Pull
Direct-Positive Recording — An Auxil-
iary to Magnetic Recording

Jan. pp. 67-70

Carlson, Frank E., Chairman, Report of
SMPTE Standards Committee

Feb. pp. 169-172

Carpenter, C. R., and Greenhill, L. P.,
A Scientific Approach to Informational-
Instructional Film Production and Utili-
zation May pp. 415-427

Christopher, H. N., and Fowler, A. D.,
Effective Sum of Multiple Echoes in
Television June pp. 491-500

Cowett, Philip M., and Orr, Lowell O.,
Desirable Characteristics of 1 6mm Enter-
tainment Film for Naval Use

Mar. pp. 245-258

Curtis, K. B., Thompson, Lloyd, and
Sutton, R. L., Prints From 16mm
Originals Feb. pp. 145-158

Davis, C. C., Frayne, J. G., and Templin,
E. W., Multichannel Magnetic Film
Recording and Reproducing Unit

Feb. pp. 105-118

del Valle, G. A., and |Putzrath, F. L.,
Optical-Magnetic Sound 16mm Pro-
jector Apr. pp. 312-322

Denney, Bruce H., and Ryder, Loren
L., Magnetic Sound Track Placement
Feb. pp. 119-136

Dimmick, G. L., and Widdop, M. E.,

Heat-Transmitting Mirror

Jan. pp. 36-42

Drew, R. O., Brumbaugh, J. M., and
Herbst, P. J., Factors Affecting the
Quality of Kinerecording

Feb. pp. 85-104

Duerr, Herman H., The Ansco Color
Negative-Positive Process

June pp. 465-479

Fowler, A. D., and Christopher, H. N.,
Effective Sum of Multiple Echoes in
Television June PP- 491—500

Francken, J. C., Schagen, P., and Bruin-
ing, H., The Image Iconoscope — A
Camera Tube for Television (Abstracted
by Pierre Mertz) June pp. 501-514

Frayne, J. G., Templin, E. W., and
Davis, C. C., Multichannel Magnetic
Film Recording and Reproducing Unit
Feb. pp. 105-118

Gillette, F. N., A Direct-Projecticn System
for Theater Television

May pp. 385-396

Graham, R. E., Mattke, C. F., and
Jensen, A. G.,- Continuous Motion
Picture Projector for Use in Television
Film Scanning Jan- PP- 1~21

Greenhill, L. P., and Carpenter, C. R.,
A Scientific Aoproach to Informational-
Instructional Film Production and Utili-
zation May pp. 415-427

Grignon, L. D., Recent Improvements in
Silencing Engine-Driven Generators

Jan. pp. 43-52

Gunzbourg, Paul M., High-Speed Motion
Picture Cameras From France

Mar. pp. 259-265

Handley, C. W., Chairman^ Progress Com-
mittee Report May pp. 397-409

Hanson, W. T., Jr., Color Negative and
Color Positive Film for Motion Picture
Use Mar. pp. 223-238

Haynes, Harold E., New Principle for
Electronic Volume Compression

Feb. pp. 137-144

Herbst, P. J., Drew, R. O., and Brum-
baugh, J. M., Factors Affecting the
Quality of Kinerecording

Feb. pp. 85-104

Higgins, G. C., and Jones, L. A., The
Nature and Evaluation of the Sharpness
of Photographic Images

Apr. pp. 277-290

June 1952 Journal of the SMPTE Vol. 58

563

Hill, Armin J., Resolution Test Chart of
the Motion Picture Research Council

June pp. 529-530

Hittle, Carl E., Twin-Drum Film-Drive
Filter System for Magnetic Recorder-
Reproducer Apr. pp. 323-328

Hoffberg, W. A., Underbill, C. R., Jr.,
and Schlanger, Ben, The Synchro-
screen as a Stage Setting for Motion
Picture Presentation

June pp. 522-528

Holcomb, A. L., Film-Spool Drive With
Torque Motors Jan- PP- 28-35

Horton, C. A., Printer Control in Color
Printing Mar. pp. 239-244

Jenkins, Norman, The Cash Customers at
the Festival of Britain Telecinema

Apr. pp. 304-311

Jensen, A. G., Graham, R. E., and
Mattke, C. F., Continuous Motion
Picture Projector for Use in Television
Film Scanning Jan. pp. 1-21

Jones, L. A., and Higgins, G. C., The
Nature and Evaluation of the Sharpness
of Photographic Images

Apr. pp. 277-290

Kaylor, J. W., and Pesek, A. V., Cinecolor
Multilayer Color Developing Machine
Jan. pp. 53-60

Kendall, Ken, Film Production Prin-
ciples — The Subject of Research

May pp. 428-444

Kudar, John C., Optical Problems in
High-Speed Camera ( Design

June pp. 487-490

Lubcke, Harry R., Color Television Re-
producers Jan. PP- 22-27

Lyman, D. F., Audio-Visual Instruction
Conference May pp. 445-449

Mattke, C. F., Jensen, A. G., and Gra-
ham, R. E., Continuous Motion Picture
Projector for Use in Television Film
Scanning Jan. pp. 1-21

Mertz, Pierre, Abstractor of The Image
Iconoscope — A Camera Tube for Tele-
vision by P. Schagen, H. Bruining
and J. C. Francken June pp. 501-514

Moran, Frank, and Carey, L. I., Push-Pull
Direct-Positive Recording — An Auxil-
iary to Magnetic Recording

Jan. pp. 67-70

Orr, Lowell O., and Cowett, Philip M.,
Desirable Characteristics of 16mm En-
tertainment Film for Naval Use

Mar. pp. 245-258

Painter, Richard O., Techniques for
Effective High-Speed Photography and
Analysis May pp. 373-384

Pesek, A. V., and Kaylor, J. W.,. Cine-
color Multilayer Color Developing

Jan. pp. 53-60

Putzrath, F. L., and del Valle, G. A.,

Optical-Magnetic Sound 16mm Pro-
jector Apr. pp. 312-322
Ryder, Loren L., and Denney, Bruce H.,
Magnetic Sound Track Placement

Feb. pp. 119-136

Schade, Otto H., Image Gradation,
Graininess and Sharpness in Television
and Motion Picture Systems — Part
II: The Grain Structure of Motion
Picture Images — An Analysis of Devia-
tions and Fluctuations of the Sample
Number Mar. pp. 181-222

Schagen, P., Bruining, H., and Francken,
J. C., The Image Iconoscope — A
Camera Tube for Television (Abstracted
by Pierre Mertz) June pp. 501-514

Schlafly, H. J., and Barton, Fred, Tele-
Prompter — A New Production Tool

June pp. 515-521

Schlanger, Ben, Hoffberg, W. A., and
Underbill, C. R., Jr., The Synchro-
screen as a Stage Setting for Motion
Picture Presentation

June pp. 522-528

Singer, Kurt, and Ward, H. Connell,
A Technical Solution of Magnetic
Recording Cost Reduction

Apr. pp. 329-340

Smith, E. C., Watkins, J. M., and Beams,
J. W., High-Constant-Speed Rotating
Mirror Feb. pp. 159-168

Spottiswoode, Raymond, Progress in
Three-Dimension al Films at the Festival
of Britain Apr. pp. 29 1 -303

Stott, John, G., Chairman, Laboratory Prac-
tice Committee June pp. 531-534
Sutton, R. L., Curtis, K. B., and Thomp-
son, Lloyd, Prints From 1 6mm Originals
Feb. pp. 145-158

Templin, E. W., Davis, C. C., and Frayne,
J. G., Multichannel Magnetic Film
Recording and Reproducing Unit

Feb. pp. 105-118

Thompson, Lloyd, Sutton, R. L., and
Curtis, K. B., Prints From 16mm
Originals Feb. pp, 145-158

Underbill, C. R., Jr., Schlanger, Ben,
and Hoffberg, W. A., The Synchro-
screen as a Stage Setting for Motion
Picture Presentation

June pp. 522-528

Ward, H. Connell, and Singer, Kurt, A
Technical Solution of Magnetic Re-
cording Cost Reduction

Apr. pp. 329-340

Watkins, J. M., Beams, J. W., and Smith,

E. C., High-Constant-Speed Rotating

Mirror Feb. pp. 159-168

Widdop, M. E., and Dimmick, G. L.,

Heat-Transmitting Mirror

Jan. pp. 36-42

Wiggin, Lyman J., Magnetic Print-
Through — Its Measurement and Re-

duction

May pp. 410-414

564

June 1952 Journal of the SMPTE Vol. 58