From the collection of the

7 n z_ m

o Prelinger v Jjibrary

t p

San Francisco, California 2007

7OL. XVI .NO. 1

JOURNAL

OF THE SOCIETY OF

vTOTION PICTURE ENGINEERS

r i

E

JANUARY, 1931

The Society of Motion Picture Engineers

Its Aims and Accomplishments

The Society was founded in 1916, its purpose as expressed in its constitution being "advancement in the theory and practice of mo- tion picture engineering and the allied arts and sciences, the standardi- zation of the mechanisms and practices employed therein, and the maintenance of a high professional standing among its members."

The Society is composed of the best technical experts in the various research laboratories and other engineering branches of the industry in the country, as well as executives in the manufacturing and producing ends of the business. The commercial interests also are represented by associate membership in the Society.

The Society holds two conventions a year, one in the spring and one in the fall, the meetings being generally of four days' duration each, and being held at various places. At these meetings papers are pre- sented and discussed on all phases of the industry, theoretical, tech- nical, and practical. Demonstrations of new equipment and methods are often given. A wide range of subjects is covered, and many of the authors are the highest authorities in their distinctive lines.

Papers presented at conventions, together with discussions, contributed articles, translations and reprints, abstracts and abridge- ments, and other material of interest to the motion picture engineer are published in the Journal of the Society.

The publications of the Society constitute the most complete ex- isting technical library for the motion picture industry.

JOURNAL =

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

SYLVAN HARRIS, EDITOR Volume XVI JANUARY, 1931 Number 1

CONTENTS

Page

Condenser and Carbon Microphones — Their Construction and

Use W. C. JONES 3

Microphone Concentrators in Picture Production

CARL DREHER 23

Tests of Motion Picture Screens WILLIAM F. LITTLE 31

Dubbing and Its Relation to Sound Picture Production

GEORGE LEWIN 38

Three Color Sub tractive Cinematography

P. D. BREWSTER AND PALMER MILLER 49

Double Toning of Motion Picture Films

J. I. CRABTREE AND W. MARSH 57 Some Causes for Variations in the Light and Steadiness of High

Intensity Carbons D. B. JOY AND A. C. DOWNES 61

An Estimate of the Present Status and Future Development of

the Home Talkies

J. B. CARRIGAN AND RUSSELL C. HOLSLAG 67

Methods of Securing a Large Screen Picture 80

Report of Secretary 86

Committee Reports 90

Abstracts 114

Patent Abstracts 117

Officers 120

Committees 121

Contributors to This Issue 123

Society Notes 125

JOURNAL

OF THE SOCIETY OF

MOTION PICTURE ENGINEERS

SYLVAN HARRIS, EDITOR

Associate Editors

R. E. FARNHAM C. E. K. MEES E. I. SPONABLE

O. M. GLUNT W. B. RAYTON L. T. TROLAND

P. E. SABINE

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers

Publication Office, 20th & Northampton Sts., Easton, Pa. General Office, 33 West 42nd St., New York, N. Y.

Copyrighted, 1931, by the Society of Motion Picture Engineers, Inc.

Subscription to non-members $12.00 per year; to members $9.00 per year; single copies $1.60. Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton Sts., Easton, Pa., or 33 W. 42nd St., New York, N. Y.

Papers or abstracts may be reprinted if credit is given to the Journal of the Society of Motion Picture Engineers.

The Society is not responsible for statements made by authors.

Entered as second class matter January 15, 1930, at the Post Office at Easton, Pa., under the Act of March 3, 1879.

CONDENSER AND CARBON MICROPHONES— THEIR CONSTRUCTION AND USE*

W. C. JONES**

Summary. — Of the numerous microphones which have been developed since Bell's original work on the telephone, only two are used extensively in sound recording for motion pictures, namely, the condenser microphone and the carbon microphone.

The condenser microphone was first proposed in 1881 but owing to its low- sensi- tivity was limited in its field of usefulness until the development of suitable amplifiers. In 1917 E. C. Wente published an account of the work which he had done on a con- denser microphone having a stretched diaphragm and a back plate so designed as to introduce an appreciable amount of air damping. The major portion of the condenser microphones used today in sound recording embody the essential features of the Wente microphone. Marked progress has, however, been made in the design and construction of these instruments with the result that they are not only more sensitive but also more stable. The factors which contribute to this improvement are described in detail in this paper. Recently a number of articles have appeared in the technical press calling attention to certain discrepancies between the conditions under which the thermophone calibration of the condenser microphone is made and those which exist in the studio. The nature of these discrepancies and their bearing on the use of the microphone are discussed.

Microphones in which the sound pressure on the diaphragm produces changes in the electrical resistance of a mass of carbon granules interposed between two electrode surfaces have been used commercially since the early days of the telephone. In recent years the faithfulness of the reproduction obtained with the carbon micro- phone has been materially improved by the introduction of an air damped, stretched diaphragm and a push-pull arrangement of two carbon elements. This instrument is finding extensive use in sound recording and reproduction fields where carbon noise is not an important factor. The outstanding design features of the push-pull carbon microphone are described in this paper and suggestions made as to the pre- cautions to be taken in its use if the best quality, maximum life, etc., are to be obtained.

Of the numerous microphones which have been developed since Bell's original work on the telephone, only two are used extensively in sound recording for motion pictures, namely, the condenser micro- phone and the carbon microphone. It has, therefore, been sug- gested that it would be fitting to review at this time the construction of these instruments and consider some of their transmission char-

* Presented at the Fall 1930 Meeting, New York, N. Y. ** Bell Telephone Laboratories, Inc., New York, N. Y.

4 W. C. JONES [J. S. M. P. E.

acteristics and the precautions which should be exercised in their use.

Condenser Microphone. — In 1881 A. E. Dolbear1 proposed a tele- phone instrument which could be used either as an electrostatic microphone or receiver. This instrument consisted of two plates insulated from one another and clamped together at the periphery. The back plate was held in a fixed position whereas the front was free to vibrate and served as a diaphragm. It is obvious that if the diaphragm were set in vibration by sound pressure, the electrical capacitance between the two plates would be changed in response to the sound waves, and if a source of electrical potential were con- nected in series with the instrument a charging current would flow which would be a fairly faithful copy of the pressure due to the sound wave. Apparently Dolbear realized that the current developed in this way would be minute, for in the telephone system which he proposed as a substitute for the one using Bell's magnetic instru- ments he employed the electrostatic instrument only as a receiver and adopted the loose contact type of microphone. At approxi- mately the same time an article appeared in the French press2 calling attention to the use of a condenser as a microphone and commenting on the fact that this type of microphone had been found to be less sensitive than the loose contact type.

Owing to the low sensitivity of the condenser microphone, the field of usefulness of this instrument was extremely limited for a number of years and it did not assume a position of importance among the instruments used in acoustic measurements and sound reproduction until suitable amplifiers had been developed. The development of the vacuum tube amplifier, however, filled this need. In 1917 E. C. Wente3 published an account of the work which he had done on an improved condenser microphone having a stretched diaphragm and a back plate so located relative to the diaphragm that, in addition to serving as one plate of the condenser, it added sufficient air damping to reduce the effect of diaphragm resonance to a minimum4. The response of this instrument was sufficiently uniform over a wide range of frequencies to make it not only useful in high quality sound reproduction but a valuable tool in acoustic measurements in general.

The major portion of the condenser microphones used today in sound recording embody the essential features of the Wente micro- phone. Marked progress has, however, been made in the design

Jan., 1931]

CONDENSER AND CARBON MICROPHONES

and construction of these instruments since the initial disclosure and it will no doubt be of interest to many to consider briefly the nature of this advance.

In the early microphones employing air damping the diaphragm was composed of a thin sheet of steel which was stretched to give it a relatively high stiffness. When assembled in the microphone the stiffness was further increased by that of the air film between diaphragm and the damping plate with the result that the resonant frequency was well above the frequencies which it was desired to transmit and the diaphragm vibrated in its normal mode over a wide frequency range. In such a structure the mechanical impedance for

FIG. 1. Western Electric Company's 394 type condenser microphone.

frequencies below resonance is due almost entirely to stiffness reac- tance. Hence a constant sound pressure produces substantially the same displacement of the diaphragm at all frequencies within this range and uniform response results except at the very low fre- quencies where an appreciable reduction in the stiffness of the air film occurs. The effective mass of a steel diaphragm is, however, relatively large and necessitates a comparatively high stiffness to secure the desired resonant frequency. From the standpoint of securing maximum sensitivity of the microphone, i. e., displacement of the diaphragm per unit force, it is of course important to make the stiffness as low as possible and employ as small a value of mechani- cal resistance as is consistent with the degree of damping required.

W. C. JONES

[J. S. M. P. E.

An improvement in both respects can be effected by decreasing the mass of the diaphragm, for with a reduced mass a given resonant frequency can be obtained with lower values of stiffness and the desired damping constant secured with less mechanical resistance.

The aluminum alloys have therefore replaced steel in the dia- phragms of most of the condenser microphones in use today. A typical example of such a microphone is the Western Electric Com- pany's instrument (394 type) shown in the photograph, Fig. 1, and the cross-sectional view, Fig. 2. The diaphragm of this instrument is made from aluminum alloy sheet 0.0011 inch in thickness. The

COMPENSATING DIAPHRAGM

DIAPHRAGM

GAS

FILLING

TUBE

DAMPING PLATE GROOVE

ACOUSTIC VALVE

FIG. 2. Cross-sectional view of the 394 type condenser microphone.

edges are clamped securely between threaded rings, gaskets of softer aluminum being provided to prevent damage at the clamping surfaces. The requisite stiffness is obtained by advancing the stretching ring until a resonant frequency of 5000 cycles is obtained. The method of determining the resonant frequency of the diaphragm is as follows. The diaphragm assembly to be tested is coupled to a condenser micro- phone which is provided with a suitable circuit for measuring its output. A special telephone receiver is placed in contact with the diaphragm on the side opposite to the coupler. Current from a vacuum tube oscillator is then passed through the winding of the receiver, setting up eddy currents in the diaphragm under test.

Jan., 1931 ] CONDENSER AND CARBON MICROPHONES 7

The forces which are developed as a result of the reaction of the mag- netic field produced by the eddy currents and that of the permanent magnet of the receiver set the test diaphragm in motion. The reso- nant frequency is determined by noting the frequency at which the output from the condenser microphone is a maximum.

In the early Wente microphone the damping plate was a continuous surface. Subsequent work by I. B. Crandall5 showed that the re- quired amount of damping at the resonant frequency could be ob- tained without adding unduly to the impedance at other frequencies by cutting grooves in the plate. This reduced the stiffness introduced by the air film and decreased the irregularity in response at low fre- quencies previously mentioned. The grooves in the damping plate of the Western Electric Company's 394 type microphone are cut at right angles. Holes, tapered at the outer end to reduce resonant effects, are bored through the plate at the intersection of the grooves to form connecting passages between the air film at the front and the cavity at the back. In order to prevent the resonance which would result if the grooves extended into the portion of the chamber surrounding the damping plate, the outer ends are closed by an annular ring which is pressed over a shoulder on the plate. The surface of the damping plate is plane within 8 X 10 ~5 inch. The departure from a plane in any individual case is determined com- mercially by the interference pattern developed when an optically flat plate is placed over the damping plate under test.

A duralumin spacing ring 0.001 inch in thickness separates the damping plate from the diaphragm. It is essential that all dust and dirt be excluded from this space. To prevent foreign material from entering through the holes in the plate a piece of silk is fastened over the outer surface. The assembly of the diaphragm and damp- ing plate is made in a dust-proof glass cabinet.

If the back wall of the condenser microphone were rigid, changes in the separation between the damping plate and the diaphragm of sufficient magnitude to affect not only the sensitivity of the instru- ment but also its frequency response characteristic would result from variations in barometric pressure. Complete compensation for these changes in pressure can only be obtained by permitting free interchange of air between both sides of the microphone dia- phragm. This is, however, objectionable owing to the fact that sufficient moisture is likely to be introduced to start corrosion and affect the insulation between the damping plate and the diaphragm.

s

W. C. JONES

[J. S. M. P. E.

A compensating diaphragm of organic material has therefore been introduced which prevents this undesirable effect of humidity but is sufficiently low in stiffness to equalize the changes in pressure encountered in the normal use of the microphone.

In order to prevent transmission losses at voice frequencies due to the presence of the compensating diaphragm, an acoustic valve is inserted between the damping plate and this diaphragm. This valve consists of a disk of silk clamped between two aluminum plates of unequal diameters. Gas in passing from the damping plate to the compensating diaphragm moves laterally from the edges of the smaller plate through the silk to a hole in the center of the larger plate. The impedance of this path is high at voice frequencies but low enough

CONDENSER TRANSMITTED

TO VACUUM TUBE VOLTMETER

THERMOPHONE OF GOLD FOIL

TO VOLTAGE SUPPLY FOR RESISTOR

INLET OUTLET HYDROGEN

FIG. 3. Cross-sectional view of the thermophone and the condenser microphone.

for steadily applied pressure differences to permit compensation for changes in barometric pressure.

After the damping plate and diaphragm are assembled the space between the clamping rings is filled with beeswax to make the joints gas-tight and exclude moisture. A hole is, however, provided for filling the microphone with nitrogen. The purpose of the nitrogen is to prevent corrosion of the damping plate and diaphragm surfaces and eliminate any reduction in pressure due to oxidation of the sealing compound.

It has been customary for some time to determine the response characteristics of a condenser microphone by the thermophone method.6 In making this measurement the diaphragm of the micro-

Jan., 1931] CONDENSER AND CARBON MICROPHONES

9

phone is coupled acoustically to the thermophone in the manner shown in Fig. 3. The thermophone consists of two strips of gold foil which are mounted on a plate and fit into the recess in the front of the microphone. Capillary tubes are provided for filling the space enclosed between the plate and the microphone diaphragm with hydrogen. This is done in order to make the wave-length of the sound developed in the recess as large as possible compared with dimensions of the chamber. If this were not the case the sound pressure at different positions in the chamber would not be in phase and the conditions on which the computations of the magnitude of the sound pressure are based would not be met. A direct current of known value is passed through the foil. Super-imposed upon the direct current is an alternating current of the desired frequency which causes fluctuations in the temperature of the foil and in the gas immediately surrounding it. These fluctuations in temperature in

-45 - 55

— «:

— —

O db = 1 POLARIZ •

VOLT ING V

fOPE /OL1

IN AG

C E

R(

;L 1C

IT 0

) PER BAF VOLTS . —

? — — '

=»— '

•— -

-s

\

-

-60 2

\ \

0 100 1000 IOOC

FIG. 4.

FREQUENCY IN CYCLES PER SECOND

Pressure calibration of the 394 type condenser microphone.

turn cause changes in the pressure on the microphone diaphragm. The magnitude of the pressure developed on the diaphragm can be computed from the constants of the thermophone and the coupling cavity, and the voltage developed by the microphone for a given pres- sure determined with suitable measuring circuits.7 Obviously, such a calibration affords a measure of the response of the microphone in terms of the actual pressure developed on the diaphragm and is independent of the external dimensions of the instrument. Hence, it does not take into account any effect which the microphone may have on the sound field when used as a pickup instrument for re- cording or broadcasting purposes. The thermophone calibration is often referred to as a "pressure" calibration and the response ob- tained by placing the instrument in a sound field of constant pressure, a "field" calibration. A thermophone calibration of a representative Western Electric 394 type condenser microphone is shown in Fig. 4.

10

W. C. JONES

[J. S. M. P. E.

For many of the uses to which the condenser microphone is put, for example, the calibration of head type telephone receivers, the conditions under which it operates agree with those under which the thermophone calibration is made. There are, however, cases where this agreement does not exist, for when a microphone is inserted in a sound field of uniform intensity the pressure on the diaphragm may depart rather widely from a constant value in certain frequency ranges. Several articles8 have recently appeared calling attention to this discrepancy between the pressure and field calibrations and pointing out that a pressure calibration of a microphone may not be entirely representative of its performance under the conditions which exist in a studio.

The difference between the pressure and field calibrations is due to several factors. In the first place the sound is diffracted around

RESPONSE IN db k (k * £ t f 6 cf 0 o

O db - 1 0 VOLT (OPEN POLARIZING VOLTAGE =

CIRCUIT) PER 200 VOLTS

BAR

j

?

^

\

-

«.

/

\

-

s

-60

20

10000

100 1000

FREQUENCY IN CYCLES PER SECOND

FIG. 5. Field calibration of the 394 type condenser microphone for a direction of approach of sound normal to the diaphragm.

the microphone differently at different frequencies. At frequencies where the wave-length is large as compared with its external dimen- sions the pressure is the same as that of the undisturbed wave. At the higher frequencies where the microphone is large in comparison with the wave-length of the sound, the pressure is twice that developed at the lower frequencies. In the 394 type microphone the effect of diffraction first becomes noticeable in the region of 1200 cycles and reaches a maximum of 6 db. at approximately 2200 cycles. The second factor which causes a difference between the pressure and field calibrations is acoustic resonance in the shallow cavity in front of the microphone. This causes the pressure actuating the dia- phragm to be higher than that of the incident sound wave in the fre- quency region of 1500 to 5500 cycles. The maximum increase in pressure occurs at approximately 3500 cycles. If the sound source

Jan., 1931]

CONDENSER AND CARBON MICROPHONES

11

is so located relative to the microphone that the waves approach from a direction normal to the diaphragm and reflection from sur- rounding walls and objects is negligible, the combined effect of diffrac- tion and resonance is to produce a maximum departure from flatness of approximately 12 db. as is shown by the field calibration, Fig. 5.9 If the sound wave travels along the diaphragm the effective pressure is reduced at the higher frequencies due to difference in phase. Hence, if the direction of approach of the sound wave is parallel to the plane of the diaphragm, the departure from flatness is materially reduced. This is brought out quite clearly by the field calibration for sound approaching from a direction parallel to the diaphragm, Fig. 6.9

The discrepancy between the pressure and field calibrations of the

-4O -45 -50 rr

-

p - rv> * o» w MILLIVOLTS

— —

•— —

==

* ^

-— ,

•*•*,

—ZL

—

5

\

S

\

o db= i VOLT (OPEN CIRCUIT)PER BAR POLARIZING VOLTAGE = 200 VOLTS

s

\

-70

\

20

100

IOOO FREQUENCY IN CYCLES PER SECOND

JO.OOO

FIG. 6. Field calibration of the 394 type condenser microphone for a direction of approach of sound parallel to the diaphragm.

condenser microphone involves two important assumptions, namely, a plane sound wave and no reflection from walls or surrounding objects. When the microphone is used in a studio much of the sound reaches the diaphragm by way of reflection from the walls of the room. The requirement of no reflection is therefore not met and the influence of the acoustic properties of the reflecting surface is added to the characteristics of the microphone. The effect of the diffusion of the sound field and the tendency for most materials to be more absorbent for sounds of high frequency appears to cause the response under studio conditions to be more nearly like that obtained when the sound approaches in a direction parallel to the diaphragm and makes the departures from the pressure calibration less marked than the field calibration for a direction normal to the diaphragm would indicate. This perhaps accounts in part at least for the instances

12 W. C. JONES [J. S. M. P. E.

in which a corrective network designed to compensate for the field calibration normal to the diaphragm failed to effect a material improvement in quality.

The acoustic conditions under which a microphone is used cover a wide range. It would therefore be difficult if not impossible to adopt a set of conditions for use in connection with a field calibration of the condenser microphone, which would be known to be represen- tative of those encountered in practice. The pressure method of calibration on the other hand is definite, simple, and is capable of being accurately duplicated in different laboratories. In view of this situation it would seem advisable to retain, at least for the present, the thermophone or pressure method of calibration for general use. In cases where precise quantitative measurements are required a field calibration of the microphone should of course be secured under the conditions of actual use. Various methods of making such a calibration have been proposed. The Rayleigh disk has been used extensively in this work thus far but there are certain very definite limitations to the extent to which it can be applied. An interesting discussion of the use of the Rayleigh disk may be found in papers by E. J. Barnes and W. West,10 and L. J. Sivian.11

It would seem reasonable to expect that future design work would be directed toward reducing transition, resonance, and phase differ- ence effects to a minimum. The results of work along this line have been reported by S. Ballantine12 and D. A. Oliver.13 In both instances the mechanical design is such that the resonant cavity in front of the diaphragm is eliminated and the housing is spherical or stream line to reduce the diffraction effect. There has as yet been little opportunity to determine the extent of the practical improve- ment effected by these changes in design and the whole discussion continues to be somewhat academic in character.

Carbon Microphone. — Bell's original microphone was essentially a generator and hence was limited in its output to the maximum speech power available at its diaphragm. The demand for telephonic communication over longer distances led to the early introduction of a carbon microphone. In this instrument the resistance of the carbon element is caused to vary in response to the sound pressure on the diaphragm and produces changes in the current supplied from an external source of electrical potential, which are fairly faithful copies of the pressure changes which constitute the sound wave. The carbon microphone is therefore in general an amplifier in which

Jan., 1931] CONDENSER AND CARBON MICROPHONES 13

a local source of power is controlled by the acoustic power of the sound wave.

The carbon element or "button" of the first microphones (Edison — 1877) was made from plumbago compressed into cylindrical form. This type of button was relatively insensitive and shortly after its introduction the suggestion (Runnings — 1878) was made that the space between the diaphragm and the fixed electrode be "partially filled with pulverized engine coke,"14 in order to increase the number of contact points and render them more susceptible to the forces developed by the motion of the diaphragm. When at its best the Runnings transmitter was fairly efficient but at times was erratic in its performance due in part to the nature of the microphonic

FIG. 7. Western Electric Company's 387 type carbon microphone.

material. In 1886 Edison15 proposed the use of granules of hard coal which had been heat treated. This was an important advance for carbon made from anthracite coal is not only used in the micro- phones which are being considered in this paper but in commercial telephone transmitters as well.

As in the case of the condenser microphone, the displacement of the diaphragm of the carbon microphone must be substantially constant at all frequencies if uniform response is to be obtained. In the early microphones of the carbon type diaphragm resonance introduced rather prominent irregularities in response. Air damped stretched diaphragms offered one solution of this problem. During the World War instruments of this type were developed and applied to the problem of locating airplanes. In 1921 double button stretched

14

W. C. JONES

[J. S. M. P. E.

diaphragm microphones were made available for use with the public address equipment installed for the inaugural address of President Harding and the exercises at Arlington on Armistice Day.16 The carbon microphones employed in sound picture recording are of the stretched diaphragm double button type. The electrical output from this type of microphone is not only of substantially uniform intensity over a wide frequency range but due to the "push-pull" arrangement of the buttons is comparatively free from harmonics; A typical example of the present day carbon microphone is shown

DAMPING _AT£ GROOVE

FIG. 8. Cross-sectional view of the 387 type carbon microphone.

in the photograph, Fig. 7. Fig. 8 is a cross-sectional view of the same type of microphone.

The diaphragm is made from duralumin 0.0017 inch in thickness and is clamped securely at its outer edge. The clamping surfaces are corrugated and emery cloth gaskets are provided to prevent slipping. The stretching of the diaphragm is done in two steps. The initial stretching ring is first advanced by means of six equally spaced screws until the diaphragm is smooth and free from irregular- ities. The inner or final stretching ring is then adjusted to a position which gives the diaphragm a resonant frequency of 5700 cycles per

Jan., 1931] CONDENSER AND CARBON MICROPHONES

15

second. The method employed in making the determination of the resonant frequency is substantially the same as that used in con- nection with the assembly of the condenser microphone, with the exception that the frequency at which the maximum output occurs is usually determined by ear rather than by the coupler method previously described. In order to insure a uniformly low contact resistance the portions of the diaphragm which are in contact with the granular carbon are covered with a film of gold deposited by cathode sputtering.

A spacing washer 0.001 inch in thickness separates the diaphragm from the damping plate. A single concentric groove is provided in the damping plate.

The buttons are of the conventional cylindrical type but are pro- vided with a novel form of closure to prevent carbon leakage at the

-0 $-50 z 0 £-55 lu cr -60 5

/

•-%

ss

— .

-—

1

—

=»«

— «_

^— — — —

^

±x

s^

^

_ J

0 db=

VOLT (OPEN CIRCUIT) PER BAR

-

0 100 1000 I0.00(

FIG. 9.

FREQUENCY IN CYCLES PER SECOND Pressure calibration of the 387 type carbon microphone.

point where they make contact with the diaphragm. The closure consists of twenty-seven rings of 0.0004 inch paper clamped firmly together at the outer edge and spreading apart at the inner edge to form a structure which effectively seals the junction between the diaphragm and the buttons without adding materially to the mechani- cal impedance.

As has already been pointed out the granular carbon is made from selected anthracite coal. The size of the granules is such that they will pass through a screen having 60 meshes per inch but will be retained on a screen having 80 meshes per inch. Before heat treat- ment the raw material is treated with hydrofluoric and hydrochloric acids to reduce the ash content. Each button contains 0.060 cc. of carbon, i. e., about 3000 granules.

The bridge which supports the button on the front of the diaphragm partially closes the acoustic cavity on that side. It is essential,

16 W. C. JONES [J. S. M. P. E.

therefore, that it be so proportioned as to have a minimum reaction on the response of the microphone and yet provide the required degree of rigidity. It was this consideration that led to the smooth stream line contour now employed.

Referring to Fig. 9 it will be observed that the adoption of an air damped stretched duralumin diaphragm for the carbon micro- phone has resulted in an instrument having a substantially uniform response over a wide range of frequencies. The arrangement of the

FIG. 10. Apparatus employed in calibrating the 387 type carbon microphone.

apparatus employed in securing the data from which this curve was plotted is shown in the photograph Fig. 10. The microphone under test was mounted in a highly damped room at a distance of six to eight feet from a source of sound which consisted of two loud speaking receivers. One of the receivers was the conventional form of moving coil direct radiator and was used to provide sound in the lower fre- quency range. The other was a special moving coil receiver with a short horn so designed as to serve as an efficient source of sound up to 10,000 cycles. l7 To reduce the effect of standing waves the mounting

Jan., 1931] CONDENSER AND CARBON MICROPHONES

17

for the receivers was so constructed that they could be rotated through a circle approximately five feet in diameter and always face the microphone under test. Before starting the test of the carbon micro- phone the receivers were calibrated by placing a calibrated condenser microphone at the point where the test instrument was to be located and determining the receiver current required to produce a pressure of one bar (one dyne per square centimeter) on the microphone dia- phragm. The condenser microphone was then removed and the test microphone substituted. The open circuit voltage developed by the microphone when supplied with a direct current of 0.025

METER

FIG. 11. Circuit employed in calibrating the 387 type carbon microphone.

ampere per button was then measured. The data obtained in this way are essentially a "pressure calibration" of the microphone and in interpreting them in terms of "field" performance the same factors must be taken into account which have been discussed in considerable detail in connection with the condenser microphone.

The circuit employed in measuring the response of the carbon microphone is shown on Fig. 11. Two steps are involved in the cali- bration of the sound source. With the output terminals of the micro- phone circuit and the sound source short circuited and the polarizing voltage for the condenser microphone removed, the attenuator is

18 W. C. JONES [J. S. M. p. E.

adjusted until the voltage applied to the measuring circuit is that developed by the condenser microphone when a sound pressure of one bar is impressed on its diaphragm. A record is made of the reading of the output meter in the measuring circuit. The polarizing voltage is then applied to the condenser microphone. After the output terminals of the attenuator have been short circuited an alternating current of a known frequency is supplied to the sound source and the magnitude of this current adjusted until the meter reading is the same as that previously obtained with the attenuator. This completes the calibration of the sound source for that frequency. After the carbon microphone has been placed in the position pre- viously occupied by the condenser microphone, the polarizing voltage is once more removed from the condenser microphone and the output from the carbon microphone circuit impressed on the measuring cir- cuit. The reading of "the output meter is recorded. The sound source and carbon microphone circuit are then short circuited and the output from the attenuator again applied to the measuring circuit. The attenuator is adjusted until the reading of the output meter is the same as was previously obtained with the carbon microphone in circuit. In this way the voltage applied to the measuring circuit when the carbon microphone is in operation is determined. The open circuit voltage developed by the carbon microphone may then be computed from the voltage and the constants of the microphone circuit. At the locations where these measurements were made a certain amount of interference from 60 cycle circuits and low fre- quency acoustic disturbances was encountered. The high pass filter in the measuring circuit was introduced to facilitate the measure- ments under these conditions. The adjustable low pass filter was used to confine the measurements to the fundamental frequency. Only that portion of the apparatus to the left of the dotted line was mounted in the damped room.

The two buttons of the carbon microphone are identical in their dimensions and if the granular carbon is in the same mechanical state have substantially the same electrical characteristics. They are also practically free from the cyclic variations in resistance known as "breathing" which result from the temperature changes caused by the power dissipated in the granular carbon. It is, however, a matter of every-day experience that a given mass of granular material will occupy different volumes depending upon the configuration of the particles. In the case of microphone carbon this change in con-

Jan., 1931] CONDENSER AND CARBON MICROPHONES 19

figuration of the granules results in changes in the contact forces of sufficient magnitude to affect the resistance and sensitivity. If these changes occur in unequal amounts in the buttons electrical unbalance results. When complete balance exists the electrical output is free from all harmonics introduced by the circuit. Hence, in using the microphone care should be taken to see that a fair degree of balance between the buttons is maintained.

The performance of a carbon microphone may be affected adversely by cohering of the granules. Severe cohering is accompanied by a serious reduction in resistance and sensitivity which persists for an extended period unless the instrument is tapped or agitated mechani- cally. One of the common causes of cohering is breaking the circuit when current is flowing through the microphone. Experiment has shown that the insertion of a simple filter consisting of two 0.02 m^t. condensers and three coupled retardation coils each having a self- inductance 0.0014 henry, will effectively protect the microphone button from cohering influences without introducing an appreciable transmission loss. This filter may be located in the base of the mounting or in a container fastened to the back of the microphone.

Aging of granular carbon may result from changes in the contact surface caused either by mechanical abrasion or overheating due to excessive contact potentials. Aging is usually accompanied by an increase in resistance and loss in sensitivity. Care should therefore be exercised in the use of the carbon microphone that it is not subjected to unnecessary vibration which would cause the granules to move relative to one another and abrade the surfaces. The use of ab- normally high voltages should also be avoided.

The quality of transmission obtained with the double button carbon microphone compares favorably with that secured with a condenser microphone. The carbon microphone also requires less amplification. There is, however, one characteristic which limits its use, namely, carbon noise. The level of the noise is much higher than that due to thermal agitation within the carbon granules18 and appears to be caused by heating at the contacts between the granules. A certain amount of gas is contained in the pores in the contact surfaces. When current passes through the button, a sufficient increase in contact temperature takes place to cause a portion of this gas to be driven off and produce the non-periodic changes in resistance which give rise to carbon noise.

In conclusion, it may be stated that the condenser and carbon

20 W. C. JONES [J. S. M. P. E.

types of microphones have been developed to a point where there is little to choose between them from the standpoint of quality of trans- mission. The design from a mechanical standpoint has also been carried to a point where little difficulty should be experienced in their use if reasonable precautions are exercised. Although requiring less amplification than the condenser microphone the extent to which the carbon microphone is used at present is limited by the higher noise level obtained. The condenser type of microphone has therefore been adopted for most of the recording work in the sound picture field.

REFERENCES

1 DOLBEAR, A. E. : "A New System of Telephony," Scientific American (June 18, 1881), p. 388.

2 LaLumiere Electrique, 1881, p. 286.

8 WENTE, E. C.: "A Condenser Transmitter as a Uniformly Sensitive In- strument for the Absolute Measurement of Sound Intensity," Physical Review, (July, 1917), pp. 39-63; "Electrostatic Transmitter," Physical Review (May, 1922), pp. 498-503.

4 CRANDALL, I. B.: "Theory of Vibrating Systems and Sound," pp. 28-39. Van Nostrand Co 1927.

5 CRANDALL, I. B.: "The Air Damped Vibratory System: Theoretical Cali- bration of the Condenser Transmitter," Physical Review (June, 1918), pp. 449- 460.

6 ARNOLD, H. D., AND CRANDALL, I. B.: "The Thermophone as a Precision Source of Sound," Physical Review (July, 1917), pp. 22-38; WENTE, E. C.: "The Thermophone," Physical Review (April, 1922), pp. 333-345; FLETCHER, H.: "Speech and Hearing," 1929, Appendix A.

7 MARTIN, W. H., AND GRAY, C. H. G.: "Master Reference System for Tele- phone Transmission," Bell System Technical Journal (July, 1929), pp. 556-559.

8 ALDRIDGE, A. J.: "The Use of a Wente Condenser Transmitter to Measure Sound Pressures in Absolute Terms," P. 0. E. E. Journal (Oct., 1928), pp. 223- 225; BALLANTINE, S.: "Effect of the Diffraction around the Microphone in Sound Measurements," Physical Review (Dec., 1928), pp. 988-992; WEST, W. : "Measurements of Sound Pressure on an Obstacle," Inst. Elec. Eng. Journal (1929), pp. 1137-1142.

• These curves are taken from unpublished work of P. B. Flanders of the Bell Telephone Laboratories, Inc.

10 BARNES, E. J., AND WEST, W.: "The Calibration and Performance of the Rayleigh Disc," Inst. of Elec. Eng. Journal, 65 (1927), pp. 871-880.

"SiviAN, L. J.: "Rayleigh Disc Method for Measuring Sound Intensities," Philosophical Magazine (March, 1928), pp. 615-620.

11 BALLANTINE, S. "Contributions from the Radio Frequency Laboratories," No. 18 (April 15, 1930).

"OLIVER, D. A.: "An Improved Microphone for Sound Pressure Measure- ments," Journal of Scientific Instruments (April, 1930), pp. 113-119.

f

Jan., 1931] CONDENSER AND CARBON MICROPHONES 21

14 RHODES, F. L.: "Beginnings of Telephony," (1929), p. 79. 16 U. S. Patent No. 406,567, 1889.

16 GREEN, I. W., AND MAXFIELD, J. P.: "Public Address Systems," A. I.E. E. Journal (April, 1923), pp. 347-358.

17 BOSTWICK, L. G.: "An Efficient Loud Speaker at the Higher Audible Fre- quencies," Journal of the Acoustical Society (October, 1930), p. 242-250.

18 JOHNSON, J. B.: "Thermal Agitation of Electricity in Conductors," Physical Review (July, 1928), pp. 97-109.

DISCUSSION

MR. COOK: I believe it has been stated in some of the published literature that the main contribution to the mechanical impedance of the diaphragm from the air chamber back of this diaphragm is from an increase in stiffness alone. Is the advantage of that thin slice of air in stiffness or is it in resistive load as far as the diaphragm alone is concerned?

MR. JONES: It is true that in the early microphones, such as the Wente micro- phone shown in Fig. 2, the air film introduced considerable stiffness as well as resistance. In the later designs where an effort has been made to secure maximum sensitivity, the stiffness of the air film has been materially reduced by the use of a grooved damping plate. The resistance introduced by the air film is now the more important factor and in microphones of the type described in this paper is relied upon to give the desired degree of flatness of response at resonance.

MR. LEWIN: I should like to inquire, regarding the condenser microphone, how constant does the tuning point maintain itself and how often should it be checked? In our studio we have a number of microphones which have been in use for over two years, and I wonder if it is necessary to re-check the tuning point after such a long period.

MR. JONES: The tuning of a microphone diaphragm involves two factors, namely, its effective mass and stiffness. If either or both change a different resonant frequency may be obtained. There is no evidence that the diaphragm changes its mode of vibration. Inasmuch as it is the predominant mass no change in the resonant frequency is introduced from this source. Changes in stiffness might arise from slipping at the diaphragm clamping surfaces or a drop in tension if the stresses in the diaphragm material are too near the yield point. All the available data point to stable clamping and the characteristics of the diaphragm material are such that the stresses are well below the yield point.

MR. KELLOGG : One wonders on hearing the statement in the paper in regard to the large contribution which the air behind the diaphragm makes to the net stiffness of the diaphragm, whether the air film might be relied on rather than the tension on the diaphragm to give the necessary high natural frequency. I assume that the tension must be sufficient to keep the diaphragm flat, but I should be interested in what Mr. Jones has to say about this possibility. I wonder if it would be asking too much to have the author give further explana- tion of the reasons why and the conditions under which the free-space calibra- tion seems to give a result which is no more representative of what you actually hear than the pressure calibration.

22 W. C. JONES

MR. JONES: In regard to the use of air stiffness to secure a high resonant frequency, my suggestion is this: It is common knowledge that the stiffness of an air film is a function of frequency. The stiffness is small at the low fre- quencies where the air is readily displaced when the diaphragm vibrates. At the higher frequencies there is a tendency to imprison the air and compress it and a higher stiffness results. If air stiffness were relied upon to control the resonant frequency, these changes in stiffness with frequency would make it difficult to secure uniform response. In addition it should be borne in mind that a small separation between the diaphragm and damping plate is necessary if the desired degree of sensitivity is to be obtained. A high diaphragm stiff- ness is required to maintain this separation.

Perhaps I did not make my views in regard to the relative merits of the pres- sure and field calibrations entirely clear. I didn't mean to convey the impres- sion that the pressure calibration is a better measure of what goes on under free space conditions. In fact, if quantitative acoustic measurements of high pre- cision are to be made with a condenser microphone, it is essential to secure a field calibration under the actual conditions of use. The "field" conditions which exist in different studios vary to such an extent that it is difficult, if not impossible, to arrive at a set of acoustic conditions which are representative of all conditions of use. The pressure calibration on the other hand is precise and can be duplicated in any laboratory. It would therefore appear advisable to retain for the present the pressure calibration for checking and comparing instru- ments and make such corrections as are necessary to fit each set of studio con- ditions.

MICROPHONE CONCENTRATORS IN PICTURE PRODUCTION*

CARL DREHER**

Summary. — By means of microphone concentrators high quality sound pickup is rendered possible at distances of the order of 20-40 feet. One such device utilizes a metal horn with, the microphone placed at the throat. In another form, applied commercially by RKO Studios, sound is picked up by an ellipsoidal or parabolic reflector and focused on a microphone, with the sensitive face of the transmitter turned away from the action. The advantages of this type of concentrator are relatively high gain, ability to record against wind or noise interference, and suitable acoustic characteristics for high quality pickup at a distance. The importance of these factors in lowering moving picture production costs is described.

When speech is picked up electrically with a microphone it is usually possible to secure high quality only by placing the pickup device relatively close to the source of sound. It is well known that the higher speech frequencies, on which intelligibility largely de- pends, are projected from the mouth in a beam. With the pickup close to the speaker a considerable portion of the energy in this beam is utilized. Moreover, when the microphone is close to the speaker, the direct output of sound predominates, whereas with the pickup at a greater distance from the source, sound reflected from the walls of the enclosure becomes a factor, and the acoustic characteristics of the room or stage enter into the equation. The microphone does not discriminate, as does, the human ear, between the direct emission of speech sounds and the heterogeneous reflections which are set up in an acoustic enclosure.

In talking picture production it is always necessary to compromise between the demands of the camera and those of the microphone. If the microphone were always placed in the most favorable position for sound pickup, it would frequently come within the field of the camera. If the microphone is kept outside the camera lines, as is required in photographing the action, it may be so far from the source of sound that the quality deteriorates. Moreover, the camera has an

* Presented at the Fall 1930 Meeting, New York, N. Y. ** RKO Radio Pictures, Hollywood, Cal.

23

24 CARL DREHER [J. S. M. P. E.

important advantage in production in that by using proper lenses it is often feasible with a given stage set-up to photograph close-up and long shots at the same time. The unaided microphone, on the contrary, will pick up high quality sound corresponding to the close-up picture only when it is near the speaker, say, within three or four feet. Thus it is frequently necessary to make additional takes merely in order to get correct sound pickup corresponding to the two views. This conflict between sound and picture requirements is shown graphically in Fig. 1.

If it were possible to modify a microphone in such a way that high quality speech could be picked up at a distance corresponding to a medium camera shot, takes for sound only could in many instances be

LOHQ SHOT

****.

FIG. 1. Showing simultaneous photographing of long shot and close-up and consequent limitations on high quality sound pickup.

eliminated, with a resulting saving in time and cost. The procedure would then be to shoot close-up sound only, modifying the quality in re-recording when necessary to simulate more distant pickup for the long shot picture. In the attainment of this objective, promising results are being secured by the use of microphone concentrating devices. One such development has been described by Olson and Wolff. l Substantially this is a metal horn in the shape of a parabola of revolution, with the microphone placed at the small end, as shown in Fig. 2. At the higher speech frequencies the horn functions as a parabolic reflector, focusing sound on the microphone. At a lower frequency there is a transition from the reflector action to one in which the horn reenforces the lower notes. Peaks of response in this

Jan., 1931]

MICROPHONE CONCENTRATORS

25

range are smoothed out by the addition of Helmholtz resonators coupled to the horn at the throat. The increase in pressure at the focus over pressure in free space is of the order of five times. The device is directional for frequencies above 200 cycles. The Olson- Wolff device is in successful operation at various points in the sound picture and radio broadcasting fields.

Fig. 3 shows another type of microphone sound concentrator which is in use at RKO Studios in Hollywood in sound picture pro- duction. The microphone is turned away from the source of sound. Sound issues from the source, strikes the reflector, and is concentrated to a focus, at or near which the microphone is placed, with its dia- phragm or sensitive element facing the reflector. Various types of reflector curves may be used, such as parabolas and ellipses. In most

&E5OMATO&S

FIG. 2. Horn sound concentrator.

cases the curve is an open one, corresponding roughly to a parabola of the equation y2 = I2x. The diameter of the surface varies, 40 to 60 inches, being commercial shapes. Means for sighting and orienting the bowl are provided, so that the action may be conveniently followed.

The technic of picking up sound with such a device is not as simple as the theory would indicate. The horn must be kept pointed at the source of sound for good quality pickup. With the above moderate dimensions the concentrator loses low frequencies (below about 500 cycles) to some extent, and contributes a marked increase in pressure, corresponding roughly to a stage of amplification, at high frequencies. In order to obviate excessive reinforcement of higher frequencies it is necessary to modify the shape of the curve or

26

CARL DREHER

[J. S. M. P. E.

to throw the microphone somewhat out of focus. In general the effort is to make the device a soft focus acoustic element, rather than to utilize maximum sharpness of focus. While there are certain voices which benefit decidedly by the dropping out of the lower tones, others become harsh and strident, and care must be used in fitting the characteristics of the concentrator to the type of pickup. How- ever, with experience it is not difficult to arrive at a proper setting for use under average conditions of pickup.

The directional properties of reflectors of the type described are often valuable in discriminating against interfering noises. For example, in RKO's Danger Lights it was necessary to shoot some

H AMPLE FOR OR/ENT/tff

FIG. 3. Open-bowl sound concentrator.

scenes in a railroad roundhouse. With an open microphone it was impossible to pick up intelligible dialog at a distance of three feet from the speaker, but by utilizing the directional properties of the parabolic concentrators, good dialog pickup was possible at much greater distances, the noises being reduced to a realistic background. Another advantage of the device is in outdoor shooting, when wind interference is encountered. When the concentrator can be turned with its concave face down-wind, it acts as a shield, the microphone being in the lee of the reflector. Wind interference is a frequent cause of delay in outdoor shooting, so that this property is valuable from a production standpoint.

Promising experimental results have also been secured with the

Jan., 1931]

MICROPHONE CONCENTRATORS

27

device indoors. On a properly constructed set it is at times difficult to discriminate by ear between the pickup of an open microphone three feet from the speaker and concentrator pickup at a distance of twenty feet. In general it is advisable to place the concentrators no further from the action than the cameras. The concentrator may be mounted on a parallel under which a camera is set up. Sets in which reverberation is kept down to a minimum are required, since concen- trators in sizes which will not interfere with photography have little directional effect at the lower frequencies. However, the loss of lows in the device may compensate roughly for the tendency toward low frequency reverberation commonly found in moving picture sets.

FIG. 4.

To date RKO Radio Pictures have used this form of microphone concentrator for almost all outdoor pickup in connection with such pictures as Danger Lights, Silver Horde, and Cimarron. The absence of reverberation outdoors makes it relatively easy to apply the device without sacrificing quality. A sizeable increase in speed of production and facility in photographing is realized. A ten or fifteen per cent reduction in over-all production cost may be realized by the skillful use of sound concentrators instead of open microphones. Not only is the microphone taken out of the zone of action, but by reason of the directional properties of the reflector

28

CARL DREHER

[J. S. M. P. E.

camera blimps can usually be dispensed with outdoors, thus further freeing the cameramen from the limitations imposed by sound. The photographs show the RKO concentrators on location (Fig. 4 and Fig. 5).

The horn type of concentrator was developed by Harry F. Olson and Irving Wolff of RCA Photophone, Inc., under the supervision of Julius Weinberger. Major credit for the open-bowl type of concen- trator is due to C. W. Horn, now General Engineer of the National Broadcasting Company, who applied it in the broadcast studios of the

FIG. 5.

Westinghouse Electric & Manufacturing Company at Pittsburgh as early as 1923, and was granted a patent, No. 1,732,722, on October 22, 1929, the application being dated January 2, 1924. Parabolic reflectors were used for airplane location work during the World War, but their use for broadcast pickup appears to have originated with Mr. Horn. The application to motion picture sound recording was carried on by Ralph K. Spotts, Philip J. Faulkner, Jr., and other members of the Sound Department of RKO Studios under the direc- tion of the writer.

REFERENCE

1 OLSON, HARRY F., AND WOLFF, IRVING: "Sound Concentrators for Micro- phones," Journal of the Acoustical Society of America, 1 (April, 1930), No. 3.

Jan., 1931] MICROPHONE CONCENTRATORS 29

DISCUSSION

MR. TAYLOR: Wouldn't it add to the interest if we were told one or two spots where the reflector was used and why it was desirable at that spot?

MR. BATSEL: It is my understanding that the reflector was used in all exteriors in the entire feature.

MR. READ: What material was used in the concentrator? A metal?

MR. BATSEL: I don't know what was used. I know that in experimenting they used different materials — plasters having various absorption characteristics. I believe this one was hard plaster but I am not sure.

PRESIDENT CRABTREE: Has this method been used in the East?

MR. BATSEL: The first one shown, the horn type, has been used in the East on short subjects and in broadcasting. The larger reflector is a development of the West Coast. We have not seen one here as yet.

MR. KELLOGG: I was told a number of years ago that large parabolic reflectors had been tried in broadcasting work and had been discarded. Can anyone tell what the faults were?

MR. KNUDSON: There must be difficulty in getting fidelity for frequencies having wave-lengths comparable with the dimensions of the reflector. From the photographs it would appear that the reflector is five or six feet in diameter; difficulties, therefore, would be introduced at frequencies below about 300 cycles.

MR. BATSEL: Mr. Dreher has already covered that subject pretty well in his paper. He doesn't claim that the device gives equal quality to a 3 foot pickup, and the diameter of the reflector is a serious consideration in picking up low frequencies. If a small reflector is used, a considerable range of frequencies is lost.

MR. HILL: I have made forty or fifty measurements on it under a variety of conditions and what Dr. Knudson said holds true. The cut-off at the low frequency end is serious excepting with reflectors 15 or 16 feet across. There is an advantage in that a gain of 10 or 12 db. may be obtained on some frequencies, but not at the low frequencies.

MR. KNUDSON: I should not like these remarks to be construed that we are disapproving of the directive microphone. I think the demonstration we have listened to today is ample justification for its use. The quality of reproduction resulting from the combined use of reflector and microphone was strikingly exhibited in Mr. Dreher's demonstration. Listening to the low puffs of the locomotive, you are conscious of distortion in the low frequency sounds. The high frequency hissing of the steam, I think, was recorded perfectly. The results attained were certainly worth while in spite of the limitations of the combined use of reflector and microphone.

CARL DREHER (Communicated) : Mr. Batsel has already answered most of the questions. Mr. Read asked what materials were used in the construction of the reflector. In some of the early types, we used plaster on burlap and similar materials, but in the forms which have been applied commercially, the usual material for the reflector has been aluminum sheeting about one thirty-second inch thick, backed by wooden cross-pieces to add rigidity. When the reflectors are manufactured in quantity, an all-metal form would naturally be adopted, but, up to now, the number built at RKO has not justified the manufacture of a mold.

As pointed out by Dr. Knudson and as I stated in the paper, there is a loss

30 CARL DREHER

of bass with reflectors of commercial diameters (40-60*). This is more pro- nounced with the original sharply focused parabolic concentrators used in Danger Lights. As was stated in the paper, we have obviated this to some degree by using the microphone out of focus and by developing other forms of reflectors, notably the ellipsoidal type shown in the photographs.

Frequency characteristics and measurements do not always give a true indica- tion of the commercial value of an acoustic device. Such measurements are of the utmost value in indicating methods of improvement and lines of future development, but the actual conditions of motion picture recording are so com- plex that what appears to be a serious fault in theory may at times be a benefit in practice. There are situations in motion picture recording when it is de- cidedly beneficial to drop out some of the energy in the lower part of the frequency band.

Again, we must consider recording as only one part of the process of making a picture, and we must also consider cost. For example, if a given form of pickup yielded good acoustic quality and allowed extreme mobility in camera work, we might be justified in using that form of recording in preference to another which afforded slightly better, sound at a great sacrifice of other picture elements. With the reflector concentrators we have been able to make some shots which added greatly to the interest of the pictures as a whole, and which would have been quite impossible Ajfith the more orthodox methods of pickup.

Moreover, if we can reduce production costs appreciably by a slight or moderate sacrifice of sound quality, there are times when we may be justified in taking that course. Many pictures are made with much greater restrictions in casting photography, set construction, etc., because of commercial limitations on ex- penditures. I do not wish to be understood as advocating deliberate sacrifices in quality of sound recording. No engineer would be justified in recording poor sound, no matter how cheaply he did it, or how much he facilitated the work of the other crafts thereby. My position is merely that, as compromises must often be made, devices like the microphone concentrators described in the paper may find a very useful field of application and that in the present state of the art it is necessary to follow up such developments in the field as well as in the laboratory.

TESTS OF MOTION PICTURE SCREENS WILLIAM F. LITTLE**

Summary. — Inasmuch as the common methods oj measuring motion picture screen reflection factors fail to bring out the essential differences between metallic and beaded screens, and fail to correctly describe the diffusing screen, a modified test procedure is suggested. This involves brightness measurements in two planes mutually perpendicular and perpendicular to the screen on which the incident beam is inclined at some suitable angle above the screen axis. Such a modification of test procedure permits the accurate analysis of a screen in terms of its suitability for a particular theater.

Tests of screen color are also recommended whereby the screen surface color and the color of the light source may be brought into closer coordination.

A superficial examination of the surfaces of a number of motion picture screens suggests their division into three classes. A plain matte surface such as white plaster gives a screen of an extremely diffusing type. A surface prepared with a metallic paint gives a surface which provides somewhat less diffusion and more specular reflection. Covering the screen surface with fine glass beads gives it specular characteristics strikingly different from those of the metallic surface screen. Two kinds of tests for the evaluation of quality in motion picture screens have been in use up to the present time. The first test evaluates the total reflection factor. The second test indi- cates the brightness of the screen surface, illuminated at normal incidence, when viewed from different angles.

The customary procedure used at present for testing motion picture screens closely follows that described by Mr. L. A. Jones and his associates in papers read before the society in 19201 and 19272 and published in the Transactions. In these tests the light was incident on the screen perpendicular to the surface and the brightness was observed at various angles in the plane perpendicular to the screen and passing through the incident beam. It was assumed that for a screen to be satisfactory, the ratio of the greatest to the least bright- ness should be no greater than 4:1. The greatest brightness would be

* Presented at the Fall 1930 Meeting, New York, N. Y. ** Electrical Testing Laboratories, New York, N. Y.

31

32

WILLIAM F. LITTLE

[J. S. M. P. E.

TEST ON BEADED SCREEN

60

50

40

30

INCIDEMT BE/>M

RMAL

TO SCF

EEN

80* 60* 40* 20*

20* 40* 60* 80*

Jan., 1931]

TESTS OF MOTION PICTURE SCREENS

33

observed on the axis of the screen, that is, at 0 degree; and the least brightness would be observed at as great an angle as the width of the house permitted.

TEST ON METALLIC SCREEN

50

30

2°

10

30

20

INCH

NORMA

)ENT B _ TO S

IAM

:REEN

80* 60° 40° 20* 0* 20* 40* 60* 80*

In actual practice the incident beam is inclined downward onto the screen, making an angle with the horizontal of from 5° to 25°, de- pending on the design of the house. The need for testing screens at such an angle is clearly shown in Figs. 1,2, and 3. The upper curves

34

WILLIAM F. LITTLE

[J. S. M. p. E.

show screen brightness data with the light incident normally. The lower curves show the light incident 12° off the normal. In this case, separate determinations must be made in the horizontal and vertical planes. These are shown by full and dash lines, respectively. It will be noted from these curves that the diffusing screen shows substan- tially the same results irrespective of the angle of light incidence, whereas the beaded and metallic surface screens appear substantially the same when tested at normal incidence, but diametrically opposed when the incident light is 12° off the axis.

TEST ON DIFFUSING SCREEN

to 10 0 1 20 10 0 1								^4
/		1 NOI	NCIDEN RMAL T	T BEAI 0 SCRE	fl EN		\
0* 60* 40* 20* 0* 20* 40* 60* 6
						-•
5^-H	>RIZONT/ PLANE	'L 1	JCIDEN ELEVA-	F BEAK ED 12*	1	/ERTICAL PLANE	\
»0* 60* 40* 20* 0* 20* 40* 60* 8 FIG. 3.

Another point in which the test at normal incidence fails to show the screen as it really is in practice can be seen by comparing the maximum brightness of the three screens under the two conditions of the test. This is shown in Table I.

TABLE I Maximum Screen Brightness

Screen Normal Incidence 12 ° off Normal

Beaded 58 ml. 62 ml.

Metallic 48 ml. 52 ml.

Diffusing 14 ml. 14 ml.

This suggested change in the test procedure makes possible the more accurate use of another criterion in the comparison of screens.

Jan., 1931]

TESTS OF MOTION PICTURE SCREENS

35

The fact that one observer in a house might see the screen four times as bright as another observer would seem to be of less consequence than for one observer to see one edge of the screen two or three times as bright as the other edge. For the comparison of screens without regard to a particular theater, suitable angles such as 12° or 15°

146 1.43 100

EVENNESS OF SCREEN BRIGHTNESS

1.79 1.42 1.04

2.16 1.75 1.08

ELEVATION

1.79 1.76 100

192 1.92 1.00

1.91 1.82 1.00

1.68 1.69 104

I 47 3.00 1.08

1.32 132

100

130 1.30 1.00

1.16 1.21 104

1.12 I 12 104

HALF PLAN

I 10 1. 10 1.04

I 05 I 05 104

I Oo 1.05 104

1.06 I 05 1.04

105 105

1.00

106 1.05 100

1.06 I 12 1.00

FIG 4

may be arbitrarily chosen or experience may indicate the advisability of making complete tests at two angles, say, 12° and 25°.

The chart, Fig. 4, shows a typical theater in elevation and one- half of it in plan, on which are recorded screen brightness ratios

36 WILLIAM F. LITTLE [J. S. M. P. E.

taken from edge to edge of the screen. In each group, which are assumed to be 20 feet apart, the upper number refers to the beaded screen, the center number to the metallic screen, and the lower one to the diffuse screen. From the viewpoint of axial to marginal brightness a person viewing the diffusing screen from one of the back corners of the house would see it at its worst. In the case of the metallic and beaded screens the variation of brightness over the screen from a single observation point becomes of increased im- portance. The location which, according to this criterion, presents the poorest view for the metallic screen would be that nearest the screen in the angle of specular reflection from the lower edge. Assum- ing the front row of seats to be at about twice the screen width from the screen, an observer sitting in a central front seat would be viewing the lower edge at approximately 12° to the normal, the upper edge at approximately 35° to the normal and there would be a varia- tion in brightness on the metallic screen of about 3 to 1, assuming the illuminated field was actually uniform as viewed from the projector. The poorest view for the beaded screen, judged by the same criterion, would be near the screen and just below the incident beam. In this position the observer would see the bottom edge of the screen at ap- proximately 12° and the top at about 17° or a variation of screen brightness of about 5 to 1.

Since a change in test procedure is being advocated, it might be we-11 to consider the advisability of including a test of the variation of re- flection factor with wave-length. There are a few tinted screens on the market for which special claims are advanced and among the so- called "white" screens the degree of whiteness is by no means con- stant. Just as the angular brightness tests involve the shape of the house, so the spectral brightness tests involve the spectral com- position of the projector light source for the complete interpretation of results. It is a fact which can be substantiated by test that since high intensity arcs, low intensity arcs, and Mazda lamps differ slightly in color, screens can be chosen to match any of these particular sources.

In view of the fact that our present test procedure shows, first, the metallic and beaded types of screen to have the same brightness characteristics, which is not the case in actual practice, second, that the data are not sufficiently complete to make screen brightness com- parisons possible from any part of the house, third, that the projection angle is practically never at 0° incidence in a motion picture house, and fourth, that the importance of color or variations in whiteness

Jan., 1931] TESTS OF MOTION PICTURE SCREENS 37

is at present ignored, it is suggested that the Society review the situa- tion with the thought that some modification in test methods might

be recommended.

It is a pleasure to acknowledge obligation to E. I. Du Pont de Nemours & Company who through the kindness of their Dr. Mc- Burnie provided the screens used in getting data for this paper.

REFERENCES

1 JONES, L. A., AND FILLIUS, MILTON F.: "Reflection Characteristics of Projection Screens," Trans. Soc. Mot. Pict. Eng., No. 11 (1920), p. 59.

2 JONES, L. A., AND TUTTLE, CLIFTON: "Reflection Characteristics of Pro- jection Screens," Trans. Soc. Mot. Pict. Eng., X, No. 28 (Feb., 1927), p. 183.

DISCUSSION

MR. RAVEN: I heartily agree with Mr. Little in his comment that some thorough work should be done in this matter with regard to reading screen brightnesses from angles and positions that approximate the actual conditions in the theater rather than by projecting the light normally and having the angle readings taken from that. As Mr. Little has said, little worthwhile information can be obtained by a screen purchaser from a reading taken when the light is normal. I hope the Projection Committee will take up the matter thoroughly so that the exhibitor will soon have available some real information that will enable him to decide what screen is best suited for his particular theater.

MR. TUTTLE: In the paper referred to by Mr. Little, of which Mr. Jones and I are the authors, we discussed the effect of varying the angle of incidence. In none of the large number of screens which we tested did the magnitude of the angle of incidence affect the measured value of reflection factor by any con- siderable amount.

From our experience, I should say that the method of test which we used is just as applicable as that advocated by Mr. Little. I am inclined to think that the differences between the results of the two methods are not real, but are due to misinterpretation of the original data.

In our data, the value of the zero degree reflection factor should, of course, be taken as the factor for regular reflection — i. e., angle of incidence equal to angle of reflection — and the other values given represent factors at various angles from the angle of regular reflection. When properly interpreted such data are applicable to conditions existing in any theater.

DUBBING AND ITS RELATION TO SOUND PICTURE PRODUCTION*

GEORGE LEWIN**

Summary. — Dubbing is essentially a re-recording process and has three important applications. The first is the re-recording of a completed feature from one form to another, as from film to disk, for release purposes. The second is the re-recording of the dialog, for the purpose of mixing in with it, sound effects or incidental music which, for technical or economic reasons, could not have been put in during the original recording. The third application is the synchronizing of foreign voices to a picture which was originally recorded in English. This last is a "doubling" rather than a dubbing process.

The original meaning of the term "dubbing" as applied to sound pictures was simply the process of re-recording a sound record. The object of re-recording is usually to transform the record from one form to another, as from film to disk, or vice versa, or else simply re-record in the same form, for the purpose of changing the recorded level or frequency characteristic. But with the rapid development of sound pictures the meaning of the term "dubbing" broadened more and more, until at the present time, it is used rather loosely to describe any process whereby the original recording is modified in any way. It is also used to describe the process whereby foreign versions of domestic pictures are made by synchronizing foreign voices to the lip movements of the original version. This latter process is essentially a "faking" process, since, when viewing such a picture, the voices we hear are not those of the original cast, but of an entirely different group of people. The same principle was used in some of the earlier domestic talking pictures in an effort to maintain the popularity of certain actors and actresses, whose speaking accents or singing voices would have been a disappointment to the film fans. So-called "voice doubles" were used to actually speak or sing while the player himself simply went through the lip motions. This form of faking, however, has been completely aban-

* Presented at the Fall 1930 Meeting, New York, N. Y. ** Paramount Publix Corp., Long Island City, N. Y.

DUBBING 39

doned now and the public may rest assured that they are actually hearing the voices of their favorites, in all domestic releases.

For the purpose of discussion, dubbing may be classified into three broad groups:

(1) Straight Dubbing

(2) Combined Dubbing and Synchronizing

(3) Dialog Dubbing

To make the discussion complete it might be well to also add a fourth group which may be called "indirect recording." This is not a form of dubbing at all, but since it is one purpose of this paper to dispel the illusion which many people seem to have, that most sound pictures are full of artificial and faked effects, it would be well to say some words later on this subject also.

STRAIGHT DUBBING

Straight dubbing is the process of re-recording a sound record by reconverting the recorded vibrations into electrical vibrations and using these reproduced vibrations to make a new record, straight dubbing may be subdivided into four groups:

(1) Film to Disk

(2) Film to Film

(3) Disk to Film

(4) Disk to Disk

In a studio, such as Paramount's, where all recording is originally done on film, film to disk dubbing is the most common form of straight dubbing. It is used only on completed features and short subjects, after they are ready for release, and is done for the purpose of making the product available to houses which are equipped for disk reproduction only.

Straight dubbing from film to film is used only for the purpose of level and quality correction. It sometimes is found, when editing a film, that various sequences which were recorded at widely sepa- rated times, or by different monitor men, or which were subjected to different laboratory processings, do not match each other in level or quality. In such cases the faulty sequences can be re- recorded and the level changed or the quality corrected by the use of suitable equalizers. This form of dubbing becomes less and less necessary as the personnel of the studio and laboratory become more expert in their respective duties, but occasionally instances do arise where expensive retakes can be avoided by suitable dubbing from

40 GEORGE LEWIN [J. S. M. P. E.

film to film. Level correction is made by simply raising or lowering the recording level of the dubbed record to what is considered the correct value. The level can be reduced to any desired point without difficulty, but in raising the level we are limited by the surface noise which is inherent in any form of recording. Quality correction is made by inserting suitable equalizers into the dubbing circuit. These will be discussed more fully later on. Film to film dubbing has its most important application in combined dubbing and synchronizing and will be discussed more fully under that heading.

Disk to film dubbing is comparatively rare ; however, on one or two special occasions it has proved quite useful in this studio. Dubbing from ordinary pressings is not entirely satisfactory as the surface noise is somewhat high. Better results are obtained by dubbing from a metal mould, which has been chromium plated for the purpose. The surface noise from a chromium plated dis1; is about 6 db. lower than that of a regular pressing, for the same signal output, and there is also a noticeable improvement in the reproduction of high fre- quencies.

Straight dubbing from disk to film is done for release purposes by studios which record originally on disk. One or two special cases of disk to film dubbing are worthy of mention. One of these was where a certain musical selection which had been recorded on disk for the scoring of a feature picture was desired for a new feature on film. Rather than go to the expense of bringing in a full orchestra to make a new sound track, the selection was dubbed from the disk and served the same purpose. Another instance was where a silent picture, The Silent Enemy, had been scored entirely on disk. The first reel of this picture had a spoken prologue which had been origi- nally recorded on film and later dubbed into the disk. On releasing this picture for foreign countries the problem was encountered of making a new first reel disk which would contain the prologue in the language of the respective countries to which it would be released. The different languages had already been recorded on film. The problem was solved by dubbing the entire first reel from the disk onto film. This film was in turn dubbed back to disk and as the spoken prologue started, the English version was turned off and the foreign version turned on. After the prologue finished we turned back to the music and completed the reel. It will be noted that the music on the completed first reel was therefore dubbed twice, from disk to film and then from film back to disk.

Jan., 1931] DUBBING 41

Disk to disk dubbing has practically no application in a studio which does all its original work on film. In a studio which records on disk, however, this form of dubbing is undoubtedly just as im- portant as film to film dubbing in this studio.

COMBINED DUBBING AND SYNCHRONIZING

Combined dubbing and synchronizing is by far the most important application of the re-recording principle. After a picture has been completed and is cut into its final form as regards action and dialog, we find that much still remains to be done before it is ready for release. We find, for instance, many dialog scenes which are supposedly occurring in places where we would expect various forms of background noise to be heard. For instance, the dialog may be taking place in a street, and we would naturally expect to hear the characteristic street noises in the background. Actually, of course, such scenes are, as a rule, recorded in the studio, without the background noises and it becomes necessary to put these sounds in after the picture is complete. This is accomplished by combined dubbing and synchronizing.

The question might be raised as to why such scenes are not recorded in their actual location, with the real background noises taking place during the actual shooting of the dialog. There are several answers to this question. In the first place, there are many locations which are often called for in stories, where it would be practically impossible to do combined recording and photographing.

For example, if we were walking along a crowded and noisy street, and were at the same time trying to hear the conversation of two people walking in back of us, we could probably do so without much difficulty, because our ears would automatically concentrate on what we were trying to hear and would reject all extraneous noises. A microphone, unfortunately is not capable of differentiating between what we are trying to record and the background noises, for it will pick up the latter with discouraging fidelity. In addition to this there is the difficulty of controlling crowds of curious onlookers and of placing cables and other sound equipment in locations where traffic is heavy. For similar reasons it would not be practical to do any recording on an actual train. It would be found that while the noise of the wheels striking the rails would seem natural enough to a person actually sitting in a train, it would sound all out of pro- portion to the dialog when heard in the theater.

42 GEORGE LEWIN [J. S. M. P. E.

All such scenes must therefore be recorded in the studio, using an artificial set, and any background noises which may be necessary are easily put in later by dubbing. They can then be controlled and made to sound just as we want them to. Other examples of sound effects best put in by dubbing, which are worthy of mention, are thunder and wind noise for storm scenes, the roar of cannon or gun shots for battle and fight scenes, the noise of passing trains and automobiles for indoor scenes where it is desired to convey the effect that outdoor noises are being heard.

The argument might be raised by those who advocate natural sound effects as opposed to artificial ones, that, granted it is im- possible to successfully record natural sound effects together with the dialog in the actual location, we might at least record them in the studio while the actual dialog takes place. It should be pointed out in this connection that dubbing of all such characteristic noises rather than recording them together with the action, has an advantage not only as regards tone fidelity, but also from an eco- nomic standpoint. It is of great importance that a feature be com- pleted in the shortest possible time. If production is delayed while the monitor man experiments with the balance between voices and sound effects, the cost of production mounts up rapidly. The work- ing crew during the shooting of a feature picture usually consists of about forty people, and is composed of directors, assistants, sound men, cameramen, electricians, and so forth, in addition to the players and extras, of whom there may be hundreds during some scenes. A dubbing crew for sound effects, on the other hand, con- sists at the most of ten or twelve men and they can in one or two working days synchronize a complete feature picture. By putting in the incidental effects after the picture is completed, considerably more time can be devoted and more pains taken to obtain the desired effects, at but a fraction of the cost.

Another important advantage of dubbing in sound effects is that stock sound tracks of these effects can be dubbed whenever necessary. This studio has a record of a thunder storm which has stood in good stead in the dubbing of several pictures.

Incidental music is almost always dubbed in after the picture is completed. In many pictures there are sequences which can be rendered more effective by the addition of a background of appro- priate music, which can either be played by an orchestra while the dialog is being re-recorded, or can be dubbed from previously

Jan., 1931] DUBBING 43

recorded sound track. The present tendency is to avoid the use of music during the shooting of the picture wherever possible, as the presence of music in the sound track hampers the editing of a picture. Without music under the dialog it is possible to rearrange se- quences, and make additions or omissions wherever desired when cutting the picture. This would be impossible, of course, if there were music in the track.

Straight musical sequences, however, such as songs or dance scenes, are usually recorded with an orchestra on the set. Attempts have been made in the past to economize on the use of orchestras during the shooting of such sequences, by having the artist sing or dance only to the accompaniment of a piano and drum, and then later dub a full orchestral accompaniment over this. This has not proved very successful, as it has been found difficult to keep the orchestra in exact time with the original track in dubbing, and even more important than this, it has been found that the artist usually does not perform with as much enthusiasm accompanied by piano and drum as he does with the aid of a full orchestra.

DIALOG DUBBING

Dialog dubbing is the expression used to describe the synchro- nizing of words to the lip movements of a picture which was shot silent, or with sound in some other language. This is not really a dubbing process at all, as it does not involve the re-recording prin- ciple, and is mentioned only for the sake of completeness. Dialog dubbing is used principally in adapting domestic pictures for foreign release. The foreign market has always been very important in motion pictures, from an economic standpoint. In the days of silent pictures, there was no particular problem involved, as it was only necessary to translate the English titles into the foreign language. The advent of sound pictures introduced a new problem, and three different solutions have been attempted. One is to record the pic- ture in the foreign language while the English version is being made. This is done by having two casts, one for each version, and so record each scene in both languages. This plan was used in mak- ing the French version of The Big Pond. A disadvantage of this method is that the success of a picture usually depends upon the popularity of its star, and unless he can also appear in the foreign version, the picture is considerably handicapped from the box-office standpoint. It is also difficult to obtain a full cast in America to

44 GEORGE LEWIN [J. S. M. P. E.

speak a foreign language without traces of American accent. The second method seeks to overcome this problem by making the foreign version in the foreign country itself. This method is quite expensive, however, and the finished product still lacks the box-office attraction of the original star.

The third method is the so-called "dubbed" version. In this method the original picture is preserved, but a foreign cast is as- sembled and after much painstaking preparation, a foreign dialog script is prepared which matches the lip movements of the original version. This script is then recorded in synchronism with the original picture. The dubbed version has the advantage of preserving the original cast in the picture, but is very difficult to synchronize, and there are many places where it is undoubtedly apparent that the voices have been faked.

In making the dubbed version, only the dialog is recorded. After this has been completed, the picture must be scored and syn- chronized, just as an original version is. If several foreign versions of the same picture are to be made, as is often the case, it is a good plan to record all synchronized music and sound effects on a separate track. All the foreign versions can then be synchronized by dubbing this sound track.

TECHNICAL PROBLEMS IN DUBBING

We come now to a short discussion of the technical problems in- volved in dubbing. An ordinary recording channel is used, and the output of a projection machine is fed into one of the mixer positions on the monitor table. In the case of straight dubbing, this is all that is necessary, except for the addition of suitable equalizers, if they are required. In the case of combined dubbing and synchronizing, several projection machines or sound dubbing heads are fed into as many mixer positions, in order to combine several sound tracks. At the same time microphones and non-synchronous records can also be mixed in. In some elaborate cases of combined dubbing and syn- chronizing, as many as seven or eight mixer positions may be in use simultaneously. These might include the original dialog, a sound track of street noises, a synchronized track of background music, a non-synchronous record of characteristic crowd noise, one or two microphones for direct pickup of special sound effects, and so forth. All of these are under control of the monitor man and can be faded in or out in any desired combination. The combined output is

Jan., 1931] DUBBING 45

recorded in the usual fashion on film to produce a new negative which is finally cut into the finished picture.

Obviously, the quality of the combined product depends to a great degree upon the fidelity with which each separate sound track is reproduced during the dubbing process. There is a certain amount of distortion inherent in any form of reproducing apparatus. In a high-grade projector using a carefully prepared release print this distortion is quite small, and for this reason the sound reproduction in high-class theaters is as a rule quite satisfactory. When re- producing sound track for dubbing purposes, however, it should be remembered that whatever distortion is present, even though it be very slight, it is recorded into the new sound track, and when this track is again reproduced in the theater, the two distortions add up, and the final effect is much more noticeable.

It might be well at this point to go into some detail regarding the inherent distortion present in a sound projection machine so as to make clear why it is negligible in a theater using high-quality release prints, and why it constitutes a serious problem in dubbing work, where we must obtain exceptionally good reproduction and must get it from green film. (Green film is the name given to film which is fresh from the laboratory and has not been run through a pro- jector more than once or twice.) The distortion present in a sound projector may be divided into two types. One is the loss of high frequencies, and the second is the introduction of a mechanical nutter due to lack of absolutely uniform motion of the film past the scanning beam in the sound gate, which results in distortion of the high frequencies. The simple loss of these high frequencies is not a very serious matter in a good projector. By actual measurement of frequency test films, recorded at constant level with our best com- mercial recording set-up, there is no appreciable loss up to 2000 cycles, and from this point upward, the loss increases gradually to about 9 db. at 6000 cycles. This loss includes both the recording and reproducing loss, and is not serious because it can be compensated for by the use of a suitable equalizer.

The introduction of mechanical flutter, however, is a much more serious problem. This flutter is apparently caused chiefly by the friction which is present between the film itself and the pressure pad which holds it in the focal plane of the optical system of the sound head. If the film has become thoroughly dry and the emulsion hardened by several days' aging, and if it has acquired a slight

46 GEORGE LEWIN [J. S. M. P. E

coating of oil as a result of having been played through a projector five or six times, the friction between film and pressure pad is very slight and uniform, and the flutter is quite negligible. If, on the other hand, the film is green, the friction is much greater and less uniform, with the result that considerable flutter is produced which results in reproduction which is popularly described as being fuzzy or raspy. In addition to this, the softness of the emulsion allows some of it to scrape off and pile up on the pressure pad to such an extent that the film sometimes goes considerably out of focus, with resulting loss of volume especially at the high frequencies.

Much work has been done on the development of special equipment which would be capable of high quality reproduction regardless of the mechanical condition of the film. An ordinary film recording machine has recently been modified to enable it to be used as a reproducer, and appears to solve the problem quite well. This machine is capable of reproducing up to 9000 cycles without ap- preciable flutter, and the frequency characteristic is better than that of an ordinary projector to the extent of about 6 db. at 6000 cycles, without equalization of any sort. Another development which has been worked out for the purpose of accelerating the dub- bing and synchronizing of pictures, is a "toe recording" process which enables one to dub directly from the negative of a sound track, without waiting for a print to be made. Toe recording is the process whereby the exposure in recording is held down to a point where we operate on the toe of the negative H & D curve of the film, rather than on the straight line portion. This process has been evolved with the view of making the negative and print interchangeable, so that prints can be made if desired, but the nega- tive itself can be used to save time. As a matter of fact it has been found that the negative gives even better quality than a regular process print. Use of this process is made in cases where the syn- chronizing music is first recorded on separate tracks and these tracks later dubbed with the original dialog.

DUBBING EQUALIZERS

In recording sound tracks which are to be used for dubbing pur- poses, the level is kept as high as possible, so that the ground noise will be relatively low. This is important especially when equalizers are used, as the action of an equalizer usually results in bringing up the ground noise. Two forms of simple equalizers are used. In

Jan., 1931] DUBBING 47

dubbing from film to wax it is, of course, necessary to reduce the energy of the low frequencies ; this is done by shunting an inductance coil of proper value across the projector output. In this way there is obtained a gradual cut-off of low frequencies from 500 cycles down. In dubbing from film to film use is made of a tuned circuit filter giving a gradual rise at high frequencies beginning at about 2000 cycles and coming to a peak at 6000 cycles. This rise in high fre- quencies approximately compensates for the combined loss which takes place in recording and reproduction.

INDIRECT RECORDING

As is quite well known, it is important to have the microphone reasonably close to the source of sound in order to obtain a good recording. Instances often arise where extremely long shots are necessary and make it impossible to get the microphone closer to the principals than thirty or forty feet. A good example of a case of this sort is in the shooting of large chorus scenes with one or two principals singing out in front. In viewing such a picture the audience would naturally expect the voices of the soloists to be clear and distinct, and to stand out from the voices of the chorus, and yet it would obviously be impossible to get a microphone close to the soloists and at the same time keep it out of the camera angle, es- pecially if the principals move back and forth during the rendition of the number. In instances such as these we resort to what is known as "indirect recording" or more popularly, the "synchronous play- back." In this method the sound is recorded first, without the picture, so that the singers may be placed in any way desired. After a good take is obtained, it is printed and then played back on the stage through large horns. The cast then take up their regular formation on the stage and go through their actions in synchronism with the sound coming from the horns, while the cameras grind. The picture is then printed together with the original sound track, and the final effect gives the illusion that both sound and action took place at the same time. In this way it is seen that full scope is give to both the sound men and cameramen to do the most justice to their respective tasks without handicapping each other. It should be understood, however, that this is not a faking process in the or- dinary sense of the word, because the voices we hear are actually those of the people we see, except that they were not recorded at the same time that the action was photographed. It cannot be called a

48 GEORGE LEWIN

dubbing process either, since the original sound track is used. It is mentioned in this paper simply for the sake of completeness in order to cover all forms of recording other than simple, direct recording.

In closing this paper I would like to emphasize the fact that or- dinary dubbing is not a form of faking, since, regardless of how many times a voice may be re-recorded for the purpose of adding sound effects, it still remains the actual voice of the person who is seen speaking in the picture. The only time voices are really faked is in the preparation of foreign versions in which case it is done only to bring to foreign countries at least the face and personality, if not the actual voice, of a popular star. The old practice of using "voice doubles" to fake the speech of actors whose own voices were not suited for recording has been completely abandoned, and only those players who can record as well as act have survived the complete transformation which the microphone has wrought in the motion picture industry.

THREE COLOR SUBTRACTIVE CINEMATOGRAPHY* P. D. BREWSTER AND PALMER MILLER**

Summary. — It is suggested that the most promising line of development of the three-color camera will involve use of three films sensitized primarily for light of dif- ferent colors, and that a lens of 50 mm. focus and f/2 speed will be used in connection with twin revolving bladed mirrors for splitting the light from the lens. The require- ment of the positive print will be met by means of a transparent dye mordant that will at least retain the size and outlines of the negative grain to produce the necessary definition.

It seemed to the writers that a general outline of the problems con- fronting those engaged in trying to improve three color subtractive pictures might be of interest to the members of the Society.

Up to the present only two color subtractive pictures have been shown, and while great improvements have been made in two color subtractive cinematography, these pictures only seem to stress more greatly the need for a three-color process. It is apparent that color cinematography will never be generally demanded by the public until it can portray colors with a reasonable degree of accuracy.

The problem is divided into two parts: first, the design of the camera, and second, the chemistry and the development of the mechanisms necessary to produce a three-color film adapted for use in any theater without changes in the projection apparatus.

It is generally conceded that any practical color camera must make its color separations simultaneously to avoid intolerable flashes or fringes of color around moving objects and that all three separations must be made from the same viewpoint; otherwise, it would be im- possible to register or superimpose the several component color images in the positive.

Accepting the limitations of a camera for making simultaneous separations from the same viewpoint, the next step is to inquire into the requirements of lenses with regard to focal length and speed. Under sound studio conditions where tungsten light is very largely in

* Presented at the Fall Meeting, October, 1930, New York, N. Y. ** Brewster Color Film Corp., Newark, N. J.

49

50 P. D. BREWSTER AND PALMER MILLER [J. S. M. P. E.

use, and where an excessive amount of light cannot be used on account of the incident heat and strain on the actor's eyes, it is necessary to use the fastest possible lens having good color correction. The limit- ing aperture at the present time is//2.

The great size of some of the sets used in the studios, and the limited floor space of sound stages, make it essential that the color camera be adapted to use a wide angle lens of not over 50 mm. focus, though 40 mm. would be still better. At the same time the beam splitting system must permit the use of lenses of from 100 mm. to 150 mm. focal length for making simultaneous close-ups and semi-close-ups in connection with a 50 mm. camera shooting long shots.

This is a very difficult requirement for both the 50 mm. and 150 mm. lenses for several reasons. In a 50 mm. camera it is very difficult to get a double beam splitter (adapted to reflect two images and transmit one) in the small lengths of 33 or 35 mm. between the rear vertex of the lens and the focal plane; while in the case of the 150 mm. f/2 lens the cone of light leaving the rear vertex is nearly 75 mm. in diameter, which very greatly increases the size of the beam splitter if no light is to be lost.

Where two or three matched lenses are used, it is necessary to have a beam splitter in front of these lenses to reflect the light rays re- ceived from one point into the separate lenses, and where one lens is employed the splitter must be behind to divide the light rays pro- jected from the single lens into three groups. We believe this can be done only in two ways ; either by a series of glass prisms, or by means of a highly polished mirror revolving at an angle to the lens and in the path of light rays. This mirror consists of a disk having a number of slots in it so that one portion of the light rays is transmitted through these slots or openings, and after passing through a suitable filter, is recorded as one of the separations; the portion of the light rays which strikes the polished surface of the blades is reflected through another filter to form the second separation; a second mirror revolv- ing at right angles to the first is used for making the third separation. The mirror usually has three blades and makes at least two revolu- tions for each exposure so that each frame is exposed three or four times. These repeated exposures have proven to give exactly the same effect on the screen as simultaneous exposure of the different color separations.1

1 U. S. Patent No. 1,752,477.

Jan., 1931] SUBTRACTIVE CINEMATOGRAPHY 51

The glass prism system has the advantage of extending, in effect, the extremely important distance between the rear vertex of the lens and the focal plane in proportion to the index of refraction of the glass used. It also has the advantage of cheapness when compared with the revolving mirrors, while the size of the driving mechanism of the camera is reduced thereby preventing noise and reducing the size of the camera.

The revolving mirror system has the advantage of not having to transmit the light through glass, which results in a loss of light, but what is more important, a possible loss of definition near the edges of the picture if the glass path is too long. Most important of all, it is possible with a revolving mirror system to make three color separa- tions on three separate films from a 50 mm. f/2 lens, without adding any lenses to the standard objective to increase the light path be- tween the rear vertex and the focal plane.

The decision as to whether to use one, two, or three films for re- cording the color separations depends not only on the camera design, but also on the study of the relative efficiency of panchromatic film ex- posed through three filters in comparison with that of two or three separate films sensitized for the region in the spectrum which they are to record.

Color separations are usually made on panchromatic emulsions by photographing through the Wratten filter No. 25 for the red, No. 57A or 58 for the green, and 49A, 49, and 49B for the blue. Transmission curves for these filters taken from the Eastman filter chart and illustrated in Fig. 1 show that No. 25 is nearly an ideal filter for the red. It transmits light of its own color, red, with high efficiency and then cuts off the other colors abruptly. None of the green filters are nearly as perfect — they transmit blue-green and green fairly well, but cut well into the orange by a long slope, with a possible average efficiency in the very important yellow green region of 30% or 40%. This critical region which largely controls the true color rendering of flesh and foliage is also harmed by the low sensitivity of panchromatic film at this point.

The blue filters 49A, 49, and 49B are even less efficient; their total over-all efficiency being only 0.7%, 0.5%, and 0.3%, respectively, and of their most favorable colors they transmit only 42%, 26%, and 15%. They cut off practically all exposure in the violet and record solely in

Ithe true blue region, while the sloping cut transmits some of the blue- green which should not be recorded by the blue separation.

52

P. D. BREWSTER AND PALMER MILLER [J. S. M. P. E.

The lack of efficiency of these filters is due to inherent qualities common to all dyes of these colors and cannot be improved. In fact, we have found Wratten filters to be of very high efficiency, and were it possible to have filters in the blue and green as good as the red No. 25, which hypothetical filters are represented by the dotted lines, they would be satisfactory.

By using three separate films for the color separation, it is possible to use an old type of non-color sensitive negative for the blue separa- tion. The sensitiveness of this type of emulsion stops almost exactly at

400

z. -^ fc — ^

UJ ^ O Ul O

O BLUE £ YELLOW ^ 3 -J

>, BLUE GREEN.O, GREEN ,> § > ,

ORANGE, RED

0

\0

420 440 460 480 500 520 540 5<30 580 600 620 640 660 680 70O

WAVE LENGTH OF LIGHT FIG. 1.

the ideal point, naturally recording the violet as well as all of the blue. Not having to use a filter, its speed is many times greater than if it were necessary to use an inefficient type of blue filter with panchro- matic film. Advantage can be taken of this fact by reflecting only a small portion (possibly 10% to 15%) of the light rays received from the lens to form the blue separation.

In case of the green separation, the use of separately sensitized films is even more important, for we then are able to obtain an emulsion which records the green and yellow-green very evenly,

Jan., 1931]

SUBTRACTIVE CINEMATOGRAPHY

53

almost to the D line, and then abruptly cuts off. Of course this emulsion is sensitive to blue, but this blue is cut off by the use of a filter of high efficiency such as an Eastman K2. By this means we get a much higher over-all efficiency and are able to record the yellow-green region with much greater fidelity.

The red separation can be made on a red sensitive emulsion, but the present panchromatic emulsion is excellent for this purpose. In either case it is necessary to use the No. 25 filter which cuts in exactly the correct place and which has a very high efficiency.

UJ

y

100 90 80 70 60 50 40 30 20 •10 0 420

0 BLUE £ VELLOW£ o Hj >, BLUE |GREEN,U| GREEN ,>-, e» , ORANGE| RED

X

r~-

~. —

— — •

\

\

A

/ /

^

-

\

\

\

/

y

7

\

\

\

/

t

/

\ \

\

N

\

2

\ i

\

\B

1 \ cl

\

s \

\

1 \

\

>

—

— •

^

\

_L

440 460 480 500 5'zc' 540 560 5$ WAVE LENGTH OF FIG. 2.

JO 6od 6EO 640 ^80 680 7i LIGHT

There is a second point in favor of separate films. It is well known that if different portions of a negative emulsion are exposed to light of different colors, they will develop to different contrasts for the same time in the developer; or these different portions of the films acted upon by lights of different colors will have different gammas. This would result in an incorrect contrast scale of the color positive, and make it difficult, if not impossible, to get a true rendering of high- lights and shades; though it would be perfectly possible to reproduce

54

P. D. BREWSTER AND PALMER MILLER [J. S. M. P. E.

middle tones in the picture substantially correctly. The film exposed to the red light will develop the highest contrast or gamma, and the blue the lowest, for a given time in the developer. For example, if the middle tones were correct one might have red highlights and blue deep shadows.

By determining in advance the gamma curves of the separate films for light of the three primary colors, it is possible to time the develop- ment of these films so that they will produce three negatives of equal

lU

uo Q-J

O

100 90 80 70 60 50 40 30 20 0

o BLUE rf YELLOW uiOUJ >, BLUE ,GREEN,0, GREEN ,^,CD>- , ORANGE, RED

.— —

.

-^^

7

^

\

/

^

1 — ^

"~^

v^

x

/,

/

N

X

\

/

X

\

\

/

\

\

\

\

\

\

V.

/

\

\

\

\

y

\

.

X

WAVE LENGTH OF LIGHT

FIG. 3.

gamma, or contrast range, from which correct positive prints can be made.

In our opinion, the requirements in the positive for each of the component images of a three color film are: definition, transparency, gradation, and hue.

Definition, especially for the blue-green and magenta images, is a matter of extreme importance. In our experience, it is necessary to retain the outlines and size of the negative image grain on the screen in order to maintain proper sharpness. Anything less than this pro-

Jan., 1931]

SUBTRACTIVE CINEMATOGRAPHY

55

duces a soft effect which, although very desirable for certain effects, is objectionable for long shots.

Transparency throughout the entire color range is absolutely es- sential. Three color cinematography requires the exact blending of all colors, and frequently needs a small percentage of one primary mixed with the other two to obtain the exact shade. It is essential that each of these primaries, whether in heavy or light shades, shall be absolutely transparent and not have the heavy tones blocked up

100 90 Z80 3 70

2 BLUE OC YELLOW uj OuJ >, BLUE |GREEN,0| GREEN i*-,^^, ORAMGE, RED

. —

^^-

-^^

:

/

7

X

^ 60 £ 50 z 40 85 30 20 -10 0

i

B/

c/

I

7

/

/

/ j

-

/

i

/

/

1

7

i

i

/

/

/

440 460 480 500 520 540 560 580 600 620 640 660 680 700 WAVE LENGTH OF LIGHT FIG. 4.

by a residual silver or mordanted image. The ideal component image would be like a color filter, pure color imbedded in the gelatin.

Finally we come to the hue and gradation of the color images. We again have the same difficulty in securing dyes that approximate the ideal as noted in the case of the filters.

The ideal requirements of the three color dyes are that each should transmit as nearly as possible 100% of the light of two of the three primary colors and in its heavier densities absorb entirely light of the other primary color.

56 P. D. BREWSTER AND PALMER MILLER

In Fig. 2 Curve C shows the transmission of a heavy step in a magenta "H & D" strip and curves A and B the lighter steps. This dye passes nearly all the blue and red but no green. Figs. 3 and 4 show the blue-green and yellow curves for the same densities.

In order to obtain a good black it is necessary that each of the three colors absorb practically all light of one of the other primaries, and it is equally important that each in their lighter gradations pass prac- tically equal quantities of the corresponding primary in order to obtain good greys, as is seen by the opening of the filter in the lighter steps. With the three dyes shown equal densities of the three super- imposed yield a grey.

DOUBLE TONING OF MOTION PICTURE FILMS J. I. CRABTREE AND W. MARSH**

Summary. — A motion picture image with green shadows and blue halftones can be prepared by toning the image blue in the usual toning solution, fixing in hypo, washing, and then re-toning followed by immersion in a solution of a basic dye. In this way the first toning bath converts the silver image to a mixture of Prussian blue and silver ferrocyanide, the reaction going to completion in the halftones but incompletely in the shadows so that some of the silver is unaffected. The silver ferrocyanide is removed in the hypo solution leaving an image consisting of pure Prussian blue in the halftones and a mixture of this substance and silver in the shad- ows. On re-immersion in the blue toning bath, the silver in the shadows is again converted to a mixture of Prussian blue and silver ferrocyanide which latter substance is a mordant for basic dyes, so that on immersion in a dye solution the dye is absorbed only to the shadows.

Commencing with a black and white image on positive motion picture film, it is possible to color this differentially by purely chemi- cal means so that the hue of the shadows is different from that of the halftones while the highlights remain perfectly clear.

One method of accomplishing this worked out by one of the authors and described previously1 consists in toning the positive image in the usual single solution iron toning bath from which the potassium alum has been omitted, washing, and then immersing in a solution of a basic dye. The omission of the potassium alum from the for- mula causes the bath to convert the halftones to white silver ferro- cyanide while only the shadows are toned blue. On immersing the film in a basic dye, the halftones assume the color of the dye while the color of the shadows is a combination of blue and that of the dye employed. For example, safranine gives pink halftones and purple shadows while auramine gives yellow halftones and green shadows.

A new method of double toning recently devised produces blue halftones and differently colored shadows. The procedure consists of four operations as follows :

* Presented at the Fall 1930 Meeting, New York, N. Y. ** Communication No. 455 from the Kodak Research Laboratories.

57

58 J. I. CRABTREE AND W. MARSH [J. S. M. P. E.

(1) Tone the positive print of normal quality in the following:

Avoirdupois Metric

Ammonium persulfate 31/* ounces 100 grams

Ferric alum (ferric ammonium sulfate) 8x/4 ounces 250 grams

Oxalic acid I1 A pounds 600 grams

Potassium ferricyanide 8l/< ounces 250 grams

Ammonium alum 1 pound 10 ounces 800 grams

Hydrochloric acid (10%) 61/* ounces 200 cc.

Water to make 50 gallons 200 liters

The method of compounding this bath is very important. Each of the solid chemicals should be dissolved separately in a small quantity of warm water, the solutions allowed to cool, filtered into the tank strictly in the order given, and the whole diluted to the required volume. If these instructions are followed, the bath will be a pale yellow color and perfectly clear.

Time of Toning— Tone fully at 70°F. (21°C.). The color of the toned image varies from a light bluish gray for short time toning (about 3 minutes) to a deep blue for long time toning (10 minutes).

Time of Washing. — Wash for 10 to 15 minutes until the high- lights are clear. A very slight permanent yellow coloration of the clear gelatin will usually occur, but should be only just perceptible. If the highlights are stained blue, then either the film was fogged during development or the bath was not compounded correctly. Washing should not be carried out for too long a period, especially with water inclined to be alkaline, because the toned image is soluble in alkali.

Life of Bath. — If the acid is renewed to the extent of the original amount after toning each 5000 feet, the bath is capable of toning 15,000 feet per 50 gallons of solution.

If even after revival the tone remains flat, the bath is exhausted and should be thrown away.

After continued use, a slight bluish sludge will collect in the bath, but this is not harmful. Should this form, to any appreciable extent, it is a result of incorrect mixing, the action of light, contact with metallic surfaces, or the presence of hypo in the bath.

(2) Immerse in a 10 per cent solution of hypo for 2 to 3 minutes and wash for 5 to 1 0 minutes.

(3) Re-immerse in the above iron toning bath for 5 minutes and wash for 10 to 15 minutes.

(4) Immerse in the solution of the basic dye for 5 to 15 minutes

Jan., 1931] DOUBLE TONING 59

until the desired depth of color in the halftones is obtained. The formula for the dye solution is as follows :

Dye 3.2 grams

Acetic acid (glacial) 2 cc. Water to make 4 liters

Dissolve the dye thoroughly in hot water, filter, add the acid, and dilute with cold water. After toning, wash the film in water until the highlights are clear or the halftones are blue.

Suitable dyes are Safranine A (pink), Chrysoidine 3R (yellowish brown) , and Auramine (yellow) supplied by the National Aniline & Chemical Company, New York, N. Y. They produce purple, dark green, and green shadows, respectively.

Theory of Process. — (a) The iron toning bath consists essentially of a solution of ferric ferricyanide in oxalic acid. This reacts with the silver image forming silver ferrocyanide and ferric ferrocyanide according to the following equation :

4Fe3(FeCN6)3 + 12Ag = 3Ag4FeCN6 + 3Fe4(FeCN6)3 Ferric Silver Silver Ferric Ferrocyanide

Ferricyanide Ferrocyanide (Prussian blue)

The reaction goes to completion in the highlights but not in the shadows so that after toning the composition of the shadows and halftones may be represented as follows:

Halftones — Silver Ferrocyanide + Prussian blue Shadows — Silver + silver ferrocyanide + Prussian blue

(b) Treatment with hypo removes the silver ferrocyanide from the halftones and shadows leaving Prussian blue in the halftones and a mixture of silver and Prussian blue in the shadows.

(c) Further treatment in the blue toning bath does not affect the halftones but the silver in the shadows is converted to a mixture of silver ferrocyanide and Prussian blue as explained above. The composition of the shadows and halftones is now as follows:

Halftones — Prussian blue

Shadows — Silver Ferrocyanide + Prussian blue

(d) Silver ferrocyanide is a mordant for basic dyes and on im- mersion in the dye bath the blue color of the shadows is therefore modified by virtue of the addition of the dye.

60 J. I. CRABTREE AND W. MARSH

Effect of Toning on Sound Track. — Tests with both the variable area and variable density types of sound records indicated that toning by the above method had little or no effect on sound quality. It is therefore possible to apply this method to sound prints.

Equipment. — Suitable materials for the construction of processing apparatus have been described.2 Allegheny metal is fairly resistant to toning baths but hard rubber is the most satisfactory material for constructing sprockets or moving parts which come into contact with the toning solution.

REFERENCES

1 "Toning and Tinting of Eastman Positive Motion Picture Film," pub- lished by Eastman Kodak Co., Rochester, New York.

2 CRABTREE, J. I., MATTHEWS, G. E., AND Ross, J. F.: "Materials for the Construction of Photographic Processing Apparatus," published by the Eastman Kodak Company, Rochester, New York.

DISCUSSION

MR. TEITEL: I would like to point out, in regard to multi-toning, that these colors have been successfully produced in the laboratories of the Multicolor Improving Co., Inc., as far back as 1914. When projected, the colors will show up properly only when the subjects portrayed are still objects. If the sub- ject were in motion, as a moving person, vehicle or fast moving clouds, the effect would be that of a mass of uneven color spots, quite unpleasant to view.

MR. CRABTREE: I agree, of course. The applications of the process are limited.

SOME CAUSES FOR VARIATIONS IN THE LIGHT AND STEADINESS OF HIGH INTENSITY CARBONS*

D. B. JOY AND A. C. DOWNES**

Summary. — The arc length-arc voltage relations of the high intensity arc depend very largely upon the relative positions of the positive and negative carbons. There is a very definite point at which the light is a maximum and the point of maximum light is not the point of maximum steadiness.

It has been shown 1>2>3>4 that the relative positions of the positive and negative carbons in the high intensity arc affect its behavior. This paper deals with the variations in the relative positions possible in commercial lamps where the angle formed by the axes of the two carbons is fixed. It will be shown that rather minor variations have an unexpectedly large effect on the amount of light and the steadiness of the arc.

The results of these variations are common to all types of high intensity lamps and carbons but the greater part of the work de- scribed here was done on 13.6 mm. positives with 3/sin. copper coated cored negative carbons at 120 amperes unless otherwise specified. The angle of the axes of the carbons was 45 degrees.

It has been the practice for carbon manufacturers to specify the current at which high intensity carbons of various sizes should be burned, but they have been reluctant to specify the voltage. A glance at Fig. 1 will explain the reason for this reticence. This figure gives graphic representations of three 70 volt arcs, but the arc lengths, measured as shown in the figure, vary from ll/6 in. to 6/s in. In X the negative flame does not touch the lower part of the positive carbon, in Y it just touches it, and in Z it overlaps it considerably. These arcs give entirely different results in quantity and quality of crater light and the projectionist would only be con- fused by any voltage specification without qualification as to the relative position of positive and negative carbon and this latter relationship is probably more important than the arc voltage.

*Presented at the Fall 1930 Meeting, New York, N. Y.

*Research Laboratories, National Carbon Company, Inc., Cleveland, Ohio.

61

62

D. B. JOY AND A. C. DOWNES

[J. S. M. P. E.

The most practical means of studying the results of the movement of the positive carbon with respect to the negative is to hold the negative carbon tip in one position and move the positive carbon along its axis. Graphic representations of the arcs obtained at 120 amperes by moving the positive carbon successive steps of l/* in. along its axis are given in Fig. 2. The arc voltage for this particular set varies from 86 in position A to 55 volts in position F. The negative flame in position A in Fig. 2 is considerably ahead of the positive so that the positive flame actually rolls out of the bottom of the positive crater before the negative flame strikes it and diverts it upward. As the positive carbon is moved ahead this condition is altered so that at D the edge of the negative flame just touches the lower edge of the positive carbon and practically the whole negative

y

FIG. 1.

flame is sweeping across the crater opening as though compressing the positive flame. Finally at F a good portion of the negative flame plays against the bottom of the positive carbon and again only a part sweeps across the positive crater. The values of relative light and arc voltage for these different arcs at 120 amperes are shown in Fig. 3. The maximum useful light is obtained at position D (as would be expected from the above description of the action of the negative flame against the positive crater opening). The light diminishes as the positive is moved in either direction from posi- tion D.

Unfortunately the position of maximum light is not the position of maximum steadiness. With the arc in position A, the direction of the positive flame from the crater is not stable, resulting in many

Jan., 1931]

CAUSES FOR LIGHT VARIATIONS

63

large flickers in the crater or useful light. This condition is im- proved as the positive carbon is moved forward so that the large flickers decrease and are practically eliminated at positions C and D. In these positions small flickers of rather short duration are evident. The negative flame is either just hitting or just clearing the lower side of the positive carbon in these positions and tends to oscillate on and off the edge of the positive shell in a rapid movement causing mediiun size flickers of short duration.

When the positive has been advanced to position E in Fig. 2 the edge of the negative flame is permanently on the bottom side of the

FIG. 2.

positive shell and the negative flame drives against the positive arc stream with a steady force of uniform direction and magnitude so that there is practically no flicker in the useful light from the arc. With the arc in this position, a high intensity spot lamp has been observed for half an hour at a time without detecting any noticeable flicker in the spot.

The light is lower as shown in, Fig. 3 for this position than in position D where some flickering is obtained. The light is still lower in position F without any change in steadiness so that the optimum condition position is that in which the edge of the negative flame

64

D. B. JOY AND A. C. DOWNES

[J. S. M. P. E.

impinges on the positive carbon as close to the end as possible without noticeable flickering on the screen or in the spotlight.

If the positive carbon is changed from C or D to that of E without changing the position of the negative or the ballast resistance, as is often done in the projection booth, there might actually be an in- crease in light with the elimination of practically all of the noticeable

• o—o RELATIVE: LIGHT AT 120 AMPERES X X ARC VOLTAGE RELATIVE LIGHT WITH VARYING CURRENT AND CONSTANT BALLAST RESISTANCE 110
1C	Q					/	s
10	Q					/ / /
|_ g	4|	•\f\			/I	\
O -1 ft	ft Q	n		/	7	\
	0 (/") 0	r\	yj			\	\
-i	8 b° e>7		/	^	x^		\
7	0 6	o				\	\
	f	r\					N
		k	i.	3 <	; i	) E	; F

FIG. 3.

flicker. If the positive is moved in the opposite direction toward position A the decrease of light will be much greater than that shown in the solid line of Fig. 3. This change is shown by the dotted line in Fig. 3 for a 115 volt power line. The reason for this is obvious when the decrease of voltage in going from A to F is noted. If the ballast resistance and line voltage were kept constant, a movement

Jan., 1931] CAUSES FOR LIGHT VARIATIONS 65

of the positive from position A to any of the other positions would necessarily give an increase in current.

The distance from the projected axis of the positive carbon to the tip of the negative carbon for the arc illustrated in Fig. 2 is 6/s in. Similar experiments were made with distances of Y2 in. and 3/4 in., with exactly the same results. Within these limits and with the same relative position of the positive and negative flames the arc length had no noticeable effect on the useful light.

In the high intensity arc burning 16 mm. positives and 11 mm. plain cored negatives with an angle of 28 degrees between the carbon axes, it was found similarly that the position of maximum light was not that of maximum steadiness and that the edge of the negative flame definitely bathed the lower edge of the positive carbon when the light was most free from flickers.

REFERENCES

1 JOY AND DOWNES: "Characteristics of High Intensity Arcs." /. Soc. Mot. Pict. Eng., XTV (March, 1930), No. 3.

2 BENFORD: "The High Intensity Arc," Trans. Soc. Mot. Pict. Eng., No. 24 (March, 1926).

3 BASSETT: "The High Power Arc in Motion Pictures," Trans. Soc. Mot. Pict. Eng., No. 11 (1920).

4 BASSETT: "Electrochemistry of the High Intensity Arc," Trans. Amer. Electrochem. Soc., 44 (1923).

DISCUSSION

MR. BASSETT: I should like to congratulate Mr. Downes on this short paper with a lot of meat in it because it is the first time that one of the mysteries of the high intensity arc has been brought down to a concise explanation. Some operators can always get the best out of a high intensity arc, and this was con- sidered a special knack. Any operator who will study this paper can acquire the knack and improve his projection.

MR. BENFORD: I think there is one point about that paper that might be stressed a little more and that is that it is not always wise to increase the current in order to get more light. When the electrode is over-loaded it is likely to smoke and the gas becomes extremely unstable. I have known of several cases where there is an actual decrease in light after the current had been increased some 10 per cent over its rated value.

PRESIDENT CRABTREE: What are the probabilities of our getting a light source of greater brightness ; also what is the temperature of the brightest source that you have been able to obtain as compared with the sun?

MR. BENFORD: The temperature of the high intensity current as measured by its color is some 5600 °K., a brightness temperature which is comparable with that of the sun.

PRESIDENT CRABTREE: This is of importance in connection with large screen

66 D. B. JOY AND A. C. DOWNES

pictures. With the present 35 mm. film with a very small aperture, we cannot get enough light through it to give a large screen having sufficient brightness. That is one of the unfortunate limitations of the use of 35 mm. film for the very large theaters.

MR. DOWNES: In the paper we presented last year, I think at Toronto, there were values given for the average intrinsic brilliancy of several high intensity arcs. The most efficient one is the 13.6 mm. arc at about 125 amperes when looked at from the point of view of high average intrinsic brilliancy. That par- ticular one, as I remember it, is of the order of 820 candle-power per square millimeter of crater opening area. That is the highest of all the ordinary high intensity arcs. The super high intensity arc at about 250 amperes has a higher intrinsic brilliancy, say from 850 to 1200 candle-power per square millimeter with the sun at about 900. Attempts have been made to use this arc for motion picture projection but so far this seems impracticable as this arc tends to be un- stable and is very difficult to handle. There is work going on in our laboratories in efforts to improve the figures, and we hope that we may be able to get some- thing satisfactory for the large size pictures.

PRESIDENT CRABTREE: Yes, but what percentage increase of brightness over the present source are you hopeful of getting?

MR. DOWNES: To increase the intrinsic brilliancy and at the same time retain the necessary steadiness of operation is very difficult and efforts to do both have not been very successful so far. Probably slightly larger light sources of about the same intrinsic brilliancy as the present arcs can be used.

AN ESTIMATE OF THE PRESENT STATUS AND FUTURE DEVELOPMENT OF THE HOME TALKIES*

J. B. CARRIGAN** AND RUSSELL C. HOLSLAGf

Summary. — In this paper the 16 mm. home talkie situation is considered from the viewpoint of the amateur. The nature and interests of present users of 16 mm. apparatus are discussed. It is concluded that only a modest distribution of sound equipment among the amateurs will be realized, and that widespread use of this equipment will be found in a new group looking more for a source of entertainment than for a hobby. Available 16 mm. sound apparatus and subject matter are de- scribed. An estimate of possible developments in apparatus and appropriate sub- jects is given. There is also a discussion of the amateur's requirements in regard to sound apparatus from the technical viewpoint.

In approaching the problem of the 16 mm. home talkie, it will be the purpose of this paper to examine the subject primarily from the viewpoint of the users and prospective users of home talkie equip- ment, considering its many angles chiefly as the consumer sees them and touching upon the questions of its nature, design, production, and distribution as they affect or will affect this great potential market. The conclusions reached are based on the data which it was possible to secure from the industry and upon personal contact or correspondence with the thousands of present home movie users who are members of the Amateur Cinema League, their international organization, or readers of its publication, Movie Makers.

Prior to the comparatively recent widespread adoption of sound motion pictures in the commercial theaters, the home, or amateur movie field, was concerned solely with the making or projection of silent pictures. It is a major fact in the situation that this is still practically the case. One reason for this lies in the difference between the interest of those who have so far embraced amateur movies, approximately 200,000 people, and the millions who are the patrons of commercial movie theaters. Both are seeking entertainment, of

* Presented at the Fall 1930 Meeting, New York, N. Y.

** Editor of Movie Makers, magazine of the Amateur Cinema League, Inc. t Technical Editor of Movie Makers and Technical Consultant of the Amateur Cinema League, Inc.

67

68 J. B. CARRIGAN AND R. C. HOLSLAG [J. S. M. P. E.

course, but the latter find satisfaction in entertainment in which they take no part, and of which they are merely spectators. The amateur movie enthusiasts, on the other hand, find in their avocation, entertainment of an active nature, a recreation in which they are producers, exhibitors, and spectators combined. All of these func- tions are a present actuality with the silent film, whereas the making of home talkies is at present attended with such difficulties that the results are nearly always of an indifferent sort. Consequently only a part of the enjoyment of home movies to which they are accustomed is provided by the present home talkie. One may rent or buy pro- fessional 16 mm. talkie productions synchronized with disk records for showing on any one of several machines now available for this pur- pose. One may be an exhibitor and spectator but not be a producer. This, with the vast majority of the present group of amateurs, would not seem to be wholly satisfying. Hence, we find a very practical psychological reason for the modest distribution to date of sound apparatus among the present home movie consumers.

But, it might be asked, are there not thousands of amateurs who are interested only in projection, who have bought projectors in order to be able to have their own home movie shows and who are not interested in making their own films? Undoubtedly there are some who answer to this description, but the limited number would be astonishing to anyone examining the situation unless one were more or less acquainted with the nature of the home movie enthusiast. While it may be obvious that there are practically no camera owners who do not have projectors, the converse is also true, that there are very few projector owners who do not also own cameras. That more projectors are sold than cameras might point to a different conclusion but examination of the facts shows that this disparity comes chiefly from the wide purchase of projectors by industry for use in selling, by schools for the development of visual education programs, and so forth.

Therefore, it would seem reasonable that we may not look to the present type of amateur for a wide adoption of home talkies until such time as the amateur can make his own. There will be a steady con- version of large numbers of the present group, of course, since the distinctions which have been drawn are purely relative and vary in intensity with the individual. The availability of synchronized films, on both a sale and rental basis, is a vital factor. At present, the home talkie offerings are distinctly limited, for the combined sale

Jan., 1931]

DEVELOPMENT OF HOME TALKIES

price of film and disk is considerable. Rental libraries for inexpensive distribution are just coming into being but undoubtedly these fa- cilities will rapidly be improved and an increasing amount of talkie equipment be gradually absorbed by a certain percentage of the present silent film users.

But, if this would not seem to promise a wide growth for home talkie exhibition, wherein lies the future of this development? Having described the present home movie user as somewhat similar to the radio fan, who in the early days of radio was chiefly con- cerned with the making of things, we must not forget that these radio set construc- tors were decidedly limited in number when compared with the millions who today enjoy commercially built radio re- ceivers. Nor should we over- look the similarity between this latter group, enjoying the sedentary amusement of radio reception, and the millions, possibly the very same, who patronize the commercial talkies.

That the users of home talkies should ever approach in number those enjoying radio reception seems doubtful. While the first cost does not seem to be prohibitive, as

home talkie equipment can even today be bought as moderately as a good radio and will undoubtedly be cheaper in the future, upkeep, however, is a different matter. The program for a radio set costs its owner nothing, at least directly, while home talkie programs mean a regular and not inconsiderable outlay. Furthermore, the radio requires only the turn of a knob in order to operate it, while the showing of a film and synchronized record requires more effort and intelligence. However, this problem, which will be discussed more fully later on, is not insurmountable.

FIG. 1. Victor Animatophone with unique vertical turntable for 16 mm. talkies.

70 J. B. CARRIGAN AND R. C. HOLSLAG [J. S. M. p. E.

Let us now consider the 16 mm. talkie equipment commercially available. As mentioned before, only projection equipment has to date been marketed and all of these machines have provided only for sound-on-disk. No sound-on-film apparatus has yet been offered commercially, although many companies are said to be working on such equipment. Several of the sound-on-disk machines first ad- vertised for the home have been withdrawn by their makers because of technical obstacles encountered in their satisfactory operation under home conditions. At the moment, there are three 16 mm. units being offered specifically for home use, the Cine- Voice, produced by the Hollywood Film Enterprises, Inc., of Hollywood, California, the Tone-O-Graph, manufactured by the North American Sound Pic-

FIG. 2. Cine-Voice, attachable with flexible shaft to any make of 16 mm. projector.

tures Corporation of New York City, and the Filmophone-radio, of the Bell & Howell Company of Chicago. Path£ Films, Inc., is also offering a 9.5 mm. machine. Two other units are being distributed specifically for use in industry and education, although they may be used in the home as well. They are the Project-O-Phone, manu- factured by the Bell & Howell Company of Chicago, and the Cinetone of the QRS-DeVry Corporation, also of Chicago. Two other units which will be appropriate for home use will shortly be announced for distribution. They are the Animatophone of the Victor Animato- graph Corporation of Davenport, Iowa, and the Visionola of the Visionola Manufacturing Company, New York City.

The Cine- Voice may be attached to any of the 16 mm. projectors now in use. It is a separate twelve or sixteen inch turntable unit

Jan., 1931] DEVELOPMENT OF HOME TALKIES 71

which is operated by the projector motor through a flexible shaft attachment. It will play either 33 Vs rpm. theater records or 78 rpm. home phonograph records. When using the former, a film speed of 24 frames per second must of course be used, the regular sixteen frames per second for the latter. It can be played either through the home radio set or through a standard amplifier and dynamic loud sp>eaker which are available as a separate unit. It sells from $105 to $129 plus $80 for the amplifying unit.

The Tone-O-Graph, consisting of motor, projector, and turntable for 16 inch records, is a compact unit incorporated in a single carrying

FIG. 3. Pathe cabinet model for 9.5 mm. sound pictures.

case. The separate motor unit drives both projector and turntable in synchronism. It can be operated through the home radio or a separate amplifier unit. It can be adapted for either 33 Ys or 78 rpm., and film speed of either 24 or 16 frames per second. Its price is $175.00, amplification system extra.

The Pathe" 9.5 mm. machine is cabinet housed and one motor operates both turntable and projector. The cabinet may be closed during projection, a port being provided in one of the doors for the light ray. It sells at $195, plus amplification system.

The Bell & Howell Project-O-Phone is provided in three carrying cases, one for projector, one for dynamic speaker, and one for a turn-

72 J. B. CARRIGAN AND R. C. HOLSLAG [J. S. M. p. E.

table and amplifier. The turntable is operated by an induction motor independently of the projector motor, excepting in so far as they are connected by a flexible shaft, insuring synchronous motor action but relieving the projector motor of the turntable load. Its 16 inch turntable revolves at 33 Vs rpm., with film speed at 24 frames per second. It weighs sixty-nine pounds and sells complete at $761.

The Filmophone-radio, also manufactured by this company, is a combination home talkie and radio placed in a handsome cabinet with synchronized turntable for either 78 or 33 Vs rpm.

The QRS-DeVry Cinetone uses an independent synchronous motor which controls both projector and turntable. A specially designed governor insures fixed operating speed. Projector, motor, sixteen inch turntable, and pickup are contained in one case, the amplifier and speaker are packed in another, being separated when in use. It operates at 33L/3 rpm., film speed of 24 frames per second, weighs ninety pounds, and is priced at $500 plus tubes.

The Animatophone is unique in construction, varying from the other units described in that the turntable operates in a vertical position, perpendicular to the projector base, instead of in the cus- tomary horizontal plane. In this instrument, the shaft of the turn- table is intimately connected with the projector mechanism, being operated through an extension of one of the projector gear shafts, thus eliminating the necessity for auxiliary flexible shafts or gear trains. The customary electrical pickup and arm are used, counter- balanced so that the needle comes in contact with the vertical record with the correct pressure for reproduction. It runs either at 33 Vs rpm., film speed of 24 frames per second or, by shifting the turntable to a secondary geared shaft, at 78 rpm., film speed of 16 frames per second. Thus either 16 inch records or ordinary home phonograph records may be reproduced. A special "air vane" governor has been incorporated in a revised model of the Victor Projector which must be used in connection with this unit. The blast from a cooling fan on this governor, impinging against a vane which causes a break in the circuit when the speed is too high momentarily slows down the motor to the proper speed, whereupon the contact is reestablished. Ampli- fication may be provided either by the home radio or by means of a unit and speaker provided separately. The device will sell at ap- proximately $100, not including projector, amplifier, and speaker.

The Visionola will be the most elaborate unit yet offered, com- bining an electric phonograph, projector, radio, and screen, all in a

Jan., 1931]

DEVELOPMENT OF HOME TALKIES

73

cabinet of the more elaborate console type. The screen is con- structed on the underside of the cabinet cover. When raised it assumes the proper angle to receive the screen image. A small mirror, carried on a collapsible arm drawn from the front of the cabinet, reflects the film image from the projector back to the screen. A unique arrangement of the film feed and takeup reels on a panel in the front of the cabinet allows easy threading. The radio unit is placed below the projector unit. The turntable, operated by the same motor which operates the projector, is in the upper part of the

FIG. 4. Bell & Howell Filmophone radio, a 16 mm. talkie cabinet model.

cabinet above the projector. It can be operated at 33 Vs or 78 rpm. with appropriate film speeds. This unit will retail at $500.00.

These, then, are the chief instruments at present available. Films synchronized with disks are now being offered in limited numbers by various companies, including Bell & Howell, Hollywood Film Enter- prises, Q.R.S.-DeVry, Fowler Studios, and, most recently, Pathe". Among the present professional producing companies which are re- leasing theater sound subjects on 16 mm. through the companies mentioned are Ufa, Amkino, Educational Pictures, Inc., and Pathe.

But what other developments may be looked for in the near future

74 J. B. CARRIGAN AND R. C. HOLSLAG [J. S. M. p. E.

on equipment for sound-on-disk picture projection? What does the future hold for sound-on-film projection apparatus? Is there any prospect of taking-apparatus either for sound-on-disk or sound-on- film ? What steps are being taken to provide a larger and finer supply of sound film subjects?

There will certainly be several more sound-on-disk projection machines offered in the near future. In the field of amateur record- ing, 16 mm. sound-on-disk recording cameras will probably be avail- able in 1931. Sound-on-film projection machines will come still later, possibly in 1931. Almost certainly the last development will be 16 mm. sound-on-film recording cameras.

In regard to increased offerings of sound films, within six months or a year, there should be greatly increased facilities for home talkie programs of the highest quality, provided from the professional production field.

Let us now consider some of the practical problems arising in the use of home sound pictures by the amateur and some of the advan- tages and disadvantages of both sound-on-disk and the sound-on-film methods. We have seen from the general development of non- professional, or home projection, apparatus that the sound-on-disk synchronizer has so far led the field. The reasons for this are logical. First of all, the turntable and pickup furnished with the home talking picture unit are similar in operation to that of the familiar phonograph and usually little difficulty is experienced in making it work properly. For the rest, since the turntable is connected to the projector by mechanical means, it is only necessary to thread and operate the projector in the usual way. The only added points of difficulty, therefore, are the careful starting of the pickup needle at an indicated spot on the record groove and the placing of a marked film frame in the projector gate, a simplified facsimile of the professional procedure in a theater projection booth when synchronized records are em- ployed. But, whereas the machine in the theater booth is provided with specially built pickup and amplifier systems, the electrical and acoustical characteristics of pickup, amplifier, loudspeakers, and horns being carefully coordinated, the home projectionist usually turns to an unclassified selection of electrical apparatus in order to reproduce the sound vibrations recorded on the disk. The pickup is always furnished with the sound attachment but there is no guarantee that good results in amplification will be obtained when the pickup output is amplified and reproduced through a radio receiver. Such amplify-

Jan., 1931] DEVELOPMENT OF HOME TALKIES 75

ing systems are notorious for their widely varying characteristics and it is only by chance that the best results are obtained, since the im- pedance of the pickup should be taken into consideration when designing amplifier transformers for use in conjunction with it. Loud speakers also are of widely varying types although in most modern sets some form of the dynamic cone is employed.

The home sound projectionist usually makes no effort to place his loud speaker in such a position relative to the screen that the illusion of sound actually emanating from the picture is produced. He is usually content to leave his loud speaker in a fixed position with rela- tion to the radio set — many times the loud speaker is incorporated in the set — and to erect his screen on a wall or table. The sound volume simply fills the room, with no directional effect whatever.

The deplorable tendency to judge a radio set by the amount of noise it will emit seems unfortunately to be carried over by the amateur to his motion picture sound projection. No matter what the size of the screen picture, and it is sometimes as small as 30 by 40 inches, the tendency is to produce a great volume of sound, simply because the amplifier will permit it. Not realizing that this does more to destroy the illusion than to create it, this type of amateur soon tires of home talkies and wonders why they seem unnatural.

In general, therefore, it would seem that with the present home sound synchronizing equipment now in use, there is little chance of even approaching the almost perfect illusion afforded by the specially coordinated apparatus used in the better theaters. Until home talkie outfits are commercially introduced that are entirely self-contained — turntable, projector, pickup, amplifier, and loud speaker self-contained and technically coordinated — the satisfactory reproduction of home talkies is uncertain.

It appears also, that it will be practically impossible to duplicate the perfection of theater installations even on a miniature scale, since remodeling of rooms to improve acoustical properties, the installation of large exponential horns, control boards, and other aids are out of the question for the amateur. In most homes, the motion picture projector is regarded as a piece of portable equipment to be packed up and stowed away in the closet when not in use, and, although several manufacturers have introduced the permanent cabinet idea — the use of a projector as a piece of furniture — it has not met with as much success as predictions would indicate. One reason probably is the already crowded condition of the living room of the average American

76 J. B. CARRIGAN AND R. C. HOLSLAG [J. S. M. P. E.

home which boasts its console or cabinet radio set and overstuffed furniture. The introduction of another cabinet to take up floor-space is frowned upon, and the fact that a talking motion picture cabinet with self-contained screen must of necessity be large is a definite factor for its sales resistance except in those cases where the home is large.

Refinements in the mechanics of home talking picture apparatus have in a general way followed the early development of professional synchronized disk work. Independent designers found that a direct connection of turntable to projector without the intervention of adequate mechanical filters was unsatisfactory for the reasons that the tendency to "flutter" was produced by the projector gears and that the projector usually had no electrical or mechanical governor for maintaining a uniform speed — a requirement absolutely necessary to prevent a periodic variation of the pitch of the reproduced sound.

Many of the familiar professional objections to the sound-on-disk system have also been advanced by the advocates of other systems. Sound film libraries must store, classify, and combine two commodi- ties, the film and the disk. Amateur users must do the same. If the film should become torn, synchronism between film and record would be destroyed. Black leader or blank film would then have to be spliced in, carefully, frame for frame. The proper disk may become separated from the film and misplaced. These and a number of other objections give rise to the question, "Why not sound-on-film for the amateur?"

The problem, of course, is not easy. Lacking definite experience with such apparatus for the amateur, it might be appropriate to discuss the difficulties that will have to be overcome to make the apparatus desirable.

In the first place, expense would have to be considered. That such an apparatus would be costly, there is little doubt. In order to secure results better than mediocre, the 16 mm. or amateur sound head would have to be as well designed as that of the professional projector. Film speed would have to be just as carefully governed and regulated in the small projector as in the large one. Yet the price of the apparatus would have to be on an amateur basis, not a professional one, if such apparatus is to be other than an extreme luxury. Other difficulties are mechanical ones. The width of the customary 35 mm. sound track, 0.1 inch or 2.5 mm., would have to be reduced considerably — half this width or less — to be accommodated in the present picture area of the 16 mm. film and still preserve an

;

Jan., 1931] DEVELOPMENT OF HOME TALKIES 77

image of satisfactory dimensions for home projection. Various plans have been proposed to overcome this difficulty. One of these is to omit the perforations from one edge of the film, leaving this band for the sound track. A film moving mechanism can be designed to function satisfactorily in this way. However, this location of the sound track is such that it would be subject to the extra wear imposed on the outer edges of the film. Plans have also been proposed for film wider than 16 mm. DeForest has recently announced sound-on-film plans involving 20 mm. film. Split 35 mm. film, giving a 17.5 mm. width, is also being tried out. These variations from the accepted home standard might be practical if controlled by a firm of exceptional resources, otherwise they would require the complete redesigning and reequipment of the market for this size. Such a step would involve so many difficulties, considering the present foothold of 16 mm. in the home, that a very drastic series of changes would have to be instituted to accommodate it.

Another potential difficulty lies in the fact that speed of 16 mm. film in passing through the projector is but 38 feet per minute (even at the rate of 24 frames per second), whereas practically all previous sound recording has been done at a film speed of 90 feet per minute. The problem of recording the higher frequencies at a speed almost one-third that of standard practice is a very definite one.

Even if the recording is done on the standard sound track and reduced to small film proportions by optical printing, the compressing of the high frequency record into a smaller space may be prevented by difficulties in resolution caused by the greater magnification of emul- sion "grain."

It is said, however, that these problems may be and are being over- come. If this is the case, there remains but one problem peculiar to amateur use. That is the care and operation of the sound-on-film device. Such systems involve light-sensitive cells and exciter lamps which, with their attendant electrical adjustments, are extremely sensitive. Such apparatus attached to the open type of amateur projector would in all probability be constantly subjected to abuse by a variety of amateur operators who are not particularly trained in such use. A home sound-on-film system would therefore have to be simplified to the utmost in the matter of control and would have to be carefully housed and protected.

So far, no steps have been taken to provide the amateur with means for recording sound. From the number of inquiries received

78 J. C. CARRIGAN AND R. C. HOLSLAG [J. S. M. P. E.

through the technical service department of the Amateur Cinema League, it is evident that the amateur requires such equipment, so that it will be possible to make an audible record with the same ease and refinement as a visual one can be made. However, sound recording is an unfamiliar subject, even in its simplest form, full of technical difficulties. At this juncture, therefore, it may be better that the amateur is not provided with apparatus that will record sound.

Most proposals for sound-on-disk synchronized recording for the amateur have involved the engraving of a soft aluminum disk, from which a play-back may be obtained immediately by means of a pickup actuated by a fiber or cactus needle. A record so engraved and re- produced will last for repeated playings and is quite satisfactory as far as reproduction qualities are concerned if properly amplified. Unsynchronized sound pickup has generally been accomplished through the agency of a carbon microphone and amplifier. Record- ings have been made at 78 rpm. with much success, although the problem of recording at a slower 33 Vs still demands mechanical re- finements in most cases. However, these are already beginning to appear, so that an entire 400 foot reel may be synchronized on a 16 inch aluminum disk.

A number of central sources for the recording of sound have been instituted for meeting the slowly increasing amateur demand. To a small, properly equipped, sound studio, the amateur may bring his films, which have already been developed, have them run off on a synchronized projector, and have his sound recorded to match on the spot. Such a procedure is entirely feasible and will undoubtedly become more prominent in the amateur field as sound reproduction becomes more familiar.

DISCUSSION

MR. ENGSTROM: From the amateur's standpoint in making his own sound picture, what priced apparatus would he be most interested in, what degree of apparatus complication would be acceptable, and what standards of sound quality would he set up?

MR. CARRIGAN: The answer to the first question is that the amateur, being familiar with the present equipment prices, would be willing to pay not exceeding $250.00; generally, he would pay that but would hesitate to pay more except in the case of wealthy individuals. It would have to be as simple as possible. There are three types of amateurs: the one who knows nothing, the one who knows something, and the one who knows a lot. The first and second groups form the great majority and would want something very simple.

Jan., 1931] DEVELOPMENT OF HOME TALKIES 79

With regard to the quality of reproduction, the amateur would not be overly critical; he is not so of his film at present. Since he made the picture, he will swallow a good deal, and the same would be true of sound.

PRESIDENT CRABTREE: On the one hand, the amateur has his radio playing every day, and occasionally the sound movie; doesn't he have a measure there? Except for recordings of himself and his friends, he would probably be just as critical as he is of the radio.

MR. CARRIGAN: I think you are correct; I was thinking of personally made films. , -

MR. TOWNSEND: As I remember, it was mentioned that acoustics in the home would need to be corrected. I don't think that is a strong factor. The paper also stated that air column horns were necessary. It seems to me that the acoustics in the average home are so far ahead of those in the average theater that this would not be a difficulty. The average speaker used in the home should be acceptable for reproduction of sound pictures.

MR. CARRIGAN: I was going more deeply into possible refinements. In order to get perfection, precautions would probably have to be taken.

MR. COOK: What size picture would the amateur expect? How long will he tolerate this "breadboard" collection of apparatus, or how long will his family tolerate it? This is a question that interests the manufacturer. The question of price which was brought up by Mr. Engstrom applies to the cabinet machine. In cruder for a manufacturer to go into it, he must make a profit. Expensive apparatus can be sold, I suppose, but most people won't buy it. The radio and automobile industry offer good proof of this. If a manufacturer can market this apparatus at a price comparable to present-day prices in radio, he can expect a profitable market from those whose earnings are from $1500 to $2500 a year.

MR. CARRIGAN: I think there the measuring stick will be the radio. I think the cabinet outfits could be put forward as cars are. The price will determine the size of the market. At first it will probably be a luxury at a high price. However, I think a firm could put out a line varying in price and touch various groups.

The normal size of picture is about 30 by 40 unless it is Kodacolor. I think the majority would accept a smaller picture if the screen were incorporated, as in the "Visionola," which gives a good effect, and has a smaller screen.

METHODS OF SECURING A LARGE SCREEN PICTURE

OPEN DISCUSSION AT THE OCTOBER, 1930, MEETING AT NEW YORK,

N. Y.

PRESIDENT CRABTREE: In order to give everyone an opportunity to air his views on the possible methods of securing a large screen picture, we reserved a place on this program for an open discussion on the subject. As Professor Hardy pointed out, if the photographic emulsion were absolutely grainless, if it were sufficiently fast, if it were so hard that it could not be scratched, and that it would not accumulate dirt, then wide film would not be necessary. Enough light could then pass through the film to ensure a reasonably large screen picture.

It has been suggested that the 35 mm. film should be run sideways. I think Mr Fear was originally responsible for that suggestion. Please correct me if I am wrong.

MR. FEAR: I believe I was the first.

PRESIDENT CRABTREE: The wide image has been squeezed opti- cally on the 35 mm. film and then stretched out optically in pro- jection. You can see an example of that at the Capitol Theater this week. This picture was produced by reducing an image on 70 mm. film down to 35 mm. film. It has been suggested that the sound be put on a separate film so as to permit of more picture space on the 35 mm. film, and there is the suggestion of the Standards Committee to introduce a film intermediate in size between 70 mm. and 35 mm. There are probably other alternatives. I should like to have your opinions.

MR. STERN: I gave a demonstration at the Paramount Theater on the 30th of September in which standard 35 mm. film was projected with the theater's own Magnoscope projector on the large screen measuring 43 X 31V2 ft. with excellent definition, and without excess granulation. This result was made possible by a special laboratory process of my own which will make feasible the use of large screens with 35 mm. film. I have also an invention for putting the sound track on separate film, saving the whole field for the picture. This 80

LARGE SCREEN PICTURES 81

•

invention consists of printing two sound tracks on 35 mm. film running in opposite directions. The film so printed is processed in the usual way and then slit in half, each half accompanying a reel of picture. It is wound on a special combination reel, of which one side carries the picture and the other side the sound track.

MR. Ross: We strongly believe in maintaining standards when- ever possible. We further believe it would be a mistake to adopt a standard of 50 or 65 mm. film or any size other than 35 and 70 mm. The small house does not have a large enough stage to accommo- date wide screen pictures, whereas the de luxe houses have such stages. The de luxe house with its comparatively larger box-office receipts can easily afford to install 70 mm. apparatus, whereas the cost of such a change would be prohibitive to the small house. We recommend the use of 70 mm. film and apparatus for the de luxe houses and 35 mm. film and apparatus for the smaller houses. Furthermore, we believe that sound will eventually be recorded on a separate film. The sound for Hell's Angels is produced on separate film having two sound tracks. We will have more to say of this during the discussion of the question of sound on separate film. It is our belief that all pictures should be recorded on 70 mm. film; however, we see no reason why pictures dealing exclusively with intermediate and close- up shots should not be recorded on 35 mm. film and optically con- densed laterally for printing wide film 1 to 1.8 release prints. Mr. Fear has modestly refrained from mentioning his system wherein the pictures are recorded longitudinally on 35 mm. film, whereby 70 mm. pictures may be printed directly therefrom. This requires the building of new cameras but so does the use of 70 mm. film. Another method of recording wide film consists of recording on 35 mm. film in the regular cameras, optically condensing the picture laterally during recording, and then optically printing normal pic- tures on 70 mm. film for the de luxe houses as well as optically printing normal pictures on 35 mm. film for the small houses. This can be accomplished by using bi-convex lenses, now standard products, which do not seem any different from ordinary printing lenses. In the final analysis we believe that all pictures will be recorded on 70 mm. film in the 1 to 1.8 ratio suggested, directly printed for the de luxe releases while for the smaller houses the 70 mm. pictures will be optically printed on 35 mm. film in the 3 to 4 regular ratio. This will make the objects appear slightly more slender than normal, an attribute for which all actors longingly crave. Obviously the

82 LARGE SCREEN PICTURES [J. S. M. P. E.

suggestion for using 35 and 70 mm. standards has to do with permit- ting the manufacturers of film to continue the production of 70 mm. raw stock which may be employed for 35 or 70 mm. recording or printing.

MR. FEAR: Gentlemen, it occured to me that you might be quite as interested in what we are doing in Hollywood as in the theoretical discussion of wide film. You have already seen two ex- periments, one of which was Happy Days, and soon you will see The Big Trail — one of the finest picture epics ever made, due to the photog- raphy and wide film. Wide film furnishes a clear background; you will see close-ups and yet miles away there will be clearly defined re- sults. This can only be accomplished on wide film. Big pictures and equipment cannot be installed in all theaters without properly considering the economic side of the question. The producers in Hollywood are trying to find the solution. The wide pictures pro- duced cost too much to show. In one case special cameras had to be bought, but no projectors were available. It was suggested that an optically reduced print be made and shown in 35 mm. projectors. The man who projected it knew something about this and was so in- terested that this method was adopted for release prints. It con- sists of reducing 70 mm. to 35 mm., using the full width of the film and a separate sound film. Two extra sound heads are required for the projection machine.

Other methods have been suggested, such as rebuilding the pres- ent equipment; this is feasible, but may involve considerable cost. In designing projectors there is a definite practical width which limits that of the film ; it is the widest width of film that can possibly pass through the projector without rebuilding the latter. That width is 50 mm. If a film of that width is used the height must be consid- ered. It is impracticable to use a higher picture in the theaters than is used at present, so it resolves itself into widening the pictures. The solution, then, is a 35 mm. film widened out. This is the answer to the controversy on wide film in Hollywood. It applies only to release prints.

PRESIDENT CRABTREE: Is Mr. Powrie here? With regard to reducing down from wide film on to 35 mm., if you look in the Trans- actions of the Society for 1924, you will find a paper by Mr. Powrie on the subject. He demonstrated the process and practically showed the improved graininess obtained by that method.

MR. Ross: If we understand Mr. Fear correctly, he stated that

Jan., 1931] LARGE SCREEN PICTURES 83

wide pictures can be photographed only onto wide film. We again call your attention to the fact that Mr. Fear's system produces wide film pictures recorded on 35 mm. film.

MR. FEAR: The 35 mm. method I suggested last year solved the problem as to projection suitable for general use. The great dif- ficulty lies in the matter of employing untrained men.

In Hollywood, we are producing the 65 mm. and 70 mm. cameras. MGM and Fox are using 70 mm. There is some difficulty which probably will be overcome, due to using four perforations instead of five. It has been suggested that five perforations be used. Warner Brothers, First National, and Universal are using 65 mm. cameras for their photography. Some prefer to make release prints on 65 mm. film. It is desirable to eliminate the human factor as far as possible in all laboratory work. There should be one negative of constant density without light changes due to the inexperience of operators who are likely to miss a light change. This can be done only by making a larger negative — 65 mm. or 70 mm. ; from this a master positive is made and corrected for light change. An expert technician should do this. Then an optically reduced duplicate negative is produced. This optically reduced negative will be su- perior in quality to an original of the same size because of the reduc- tion of grain in such an optical print.

I think that answers your question ; it is an economic problem.

MR. Ross: I merely wish to add that in suitably reconstructed printers no prisms are required. Further, the question of making dupe negatives or fixed density positives seems to be apart from the question of wide film.

MR. GRIFFIN: The suggestion has been made that it is a good plan to reduce optically from a wide negative to 35 mm. film. I have seen pictures projected which were made in this manner and as far as pictorial quality is concerned the result is very good. It must not be forgotten, however, that the problem of projecting this type of picture to a screen 40 ft. wide is highly impracticable because it is impossible to pass the necessary amount of light through the small aperture. The size of the aperture in this case is approxi- mately 0.940 wide by half that in height. Using 135 amperes at high intensity and condensers of the most improved design, it is im- possible to procure more than half the illumination on the screen that is acceptable for the projection of standard film, and it is necessary that the projectionist be on the alert at all times to constantly secure

84 LARGE SCREEN PICTURES [J. S. M. p. E.

even this result. It must also be remembered that this reduction print, running as it does across the film from sprocket hole to sprocket hole, allows for no sound track and it is necessary to record on either disk or separate film, which adds considerably to the cost of ap- paratus necessary for sound reproduction, to say nothing of the errors in synchronizing which may and do arise frequently under this system. I don't think the solution lies in using 35 mm. reduced prints. I feel that the industry should certainly consider going to a film of, perhaps, 50 mm., in which all the excellent quality obtainable in wide negatives can be incorporated and which we know can be satisfactorily projected. By adopting such a dimension all the pro- jectors now in use could be converted to handle this size as well as 35 mm. film at comparatively little cost to the exhibitor com- pared with the cost of equipment for the wide film as we now know it. Such a standard would be economically sound and enable the pro- duction of wide film to go ahead without a great deal of delay. Our corporation has spent a tremendous amount of money on film equip- ment and wide films but I am sure we should be willing to discard this for a standard which is economically sound and which allows the salvaging of the greater part of equipment at present in use.

MR. STERN: I should like to know if Billy the Kid was produced by making a 35 mm. print from the 70 mm., or was it an optical print from a 70 mm. negative?

MR. FEAR: It was produced by original reduction of the negative to the positive. On such a huge reduction, it is almost impossible to utilize the method I outlined before. I have a company in produc- tion on a wide film picture with four more to start in the next 90 days, which will be released to the independent exhibitor. We are making plans for a patent license to rebuild projection machines for a certain size film and are anxious to have a standard to adopt. In our system of conversion of projectors, the 35 mm. sprocket is cut in two so that the film is run throughout in the extensible position. We have added space between the sprockets. By moving levers we can change from one film to the other. The method is extremely inexpensive.

MR. Ross: I wish to call attention to the fact that the frames in Billy the Kid are about one-third smaller in height than in standard 35 mm. film and that, therefore, in the systems in which we have suggested using standard size frames there will be available one- third more light, or aq average of approximately 7 foot candles. Furthermore, whereas the foot candles have been reduced from, say,

Jan., 1931] LARGE SCREEN PICTURES 85

11 to 7, the picture viewed has approximately twice the area and there will be as much illumination at an intensity of 7 foot candles on wide pictures as at 11 foot candles on regular size pictures. We believe that if the wide screen pictures were to be projected with an average screen intensity of 11 foot candles, the amount of light reflected by the screen would be objectionable to the audience, especially to those in the rear portion of the auditorium. Further- more, with quick changes of scene the light and shadows reflected onto the walls of the auditorium would also be objectionable.

PRESIDENT CRABTREE: What technical difficulties have been encountered in the handling of wide film in Hollywood?

MR. FEAR: In Hollywood, everybody is enthusiastic about wide film. The producers are a little anxious about the situation because they want to know what is going to be done about it. No cameras are being sold for 35 mm. film. I will not sell them because I know we are going to a new standard and they will become obsolete in a short time. United Artists, Warner Brothers, First National, Fox, MGM, and Universal are working on wide film at the present time. One of my cameramen, Mr. J. O. Taylor, started on another picture last week. Every producer out there is awaiting your decision. It is highly improbable that every producer will have a different standard. In the majority of cases where we have handled wide film we have not had any technical difficulty. It is handled the same as the other, and the cameras have caused no more trouble than the others. The cameraman shoots a little differently from the 35 mm., but the main difficulty lies in projection.

MR. GRIFFIN: Mr. Fear said that difficulty has been experienced on the Coast in the projection of wide pictures. I know that to be so but I believe it is because the improper condenser combinations were used and improper distances were maintained between the arc and condensers and the condensers and aperture. We in the East are closer to the optical manufacturing organizations and lamp manu- facturers and close cooperation has enabled us to obtain satisfactory results more quickly. All of the data obtained have been forwarded to the Pacific Coast and I have word that they are getting far better results than formerly. I don't think there is any difficulty now with the projection of wide film but certainly a film of approximately 50mm. width would eliminate any slight difficulty which might be experi- enced.

REPORT OF THE SECRETARY* Sept. 30, 1929, to Oct. 1, 1930.

This report covers the term of the fiscal year beginning October 1, 1929, and ending September 30, 1930. During the first four months of the period covered by this report, the affairs of the Secre- tary were conducted by Mr. R. S. Burnap. Changes in the business administration of the company with which he was connected required his resignation from the office of secretary on February 9, 1930. Thereupon, at the invitation of the Board of Governors, the writer assumed the duties of the secretary's office for the remainder of the term.

MEMBERSHIP

The total membership of the Society, as of the last day of the past fiscal year, is 756 members, divided as follows :

Eight Honorary members. These are :

Mr. George Eastman, Rochester, New York.

Mr. Thomas A. Edison, West Orange, New Jersey.

Dr. F. E. Ives, Philadelphia, Pa.

Mr. C. Francis Jenkins, the founder of the Society, Washington, D.C.

Mr. Louis Lumiere, Paris, France.

The Presidency, Societe* Franchise de Photographic, Paris, France.

The Presidency, Die Deutsche Kinotechnische Gesellschaft, Berlin, Germany.

The Presidency, Royal Photographic Society, London, England.

There are 371 Active members, and 377 Associate members. And, in addition, there are 14 sustaining members consisting of various commercial and industrial organizations.

DISTRIBUTION OF MEMBERSHIP

Of the Society's membership 664 are under the jurisdiction of four local sections, with headquarters in New York, N. Y.; Chicago, 111. ; Hollywood, Calif. ; and London, England. The distribution of members among these four sections is as follows :

* Presented at the Fall 1930 Meeting, New York, N. Y. 86

REPORT OF THE SECRETARY 87

New York Section 182 Active, 175 Associate

Chicago Section 29 Active, 48 Associate

Pacific Coast Section 61 Active, 47 Associate London Section 71 Active, 51 Associate

The combined territory of the three American Sections covers the entire United States, exclusive of territorial possessions. The terri- torial limits of the several sections were defined by the Board of Governors as follows:

The United States is divided from East to West into three geo- graphical sections by drawing two north and south parallels. One of these lies fifty miles west of Cleveland, Ohio, and the other fifty miles west of Denver, Colorado. That part of the United States lying east of the first-named parallel comprises the territory of the New York Section; that part of the United States lying west of the second-named parallel comprises the territory of the Pacific Coast Section ; the area intermediate between the two parallels constitutes the territory of the Chicago Section.

The territory of the London Section remains unchanged as con- sisting of the British Isles.

There are 84 members of the Society, residing in 18 foreign coun- tries, not including Great Britain, who do not come under the juris- diction of any local Section. The membership distribution among these countries follows:

Active Associate

Argentina 1

Australia 2

Brazil 1

Burma 1

Canada 5 10

France 7 11

Germany 8 9

Holland 1

India 5 5

Italy 1 2

Japan 1 3

New Zealand 2

Norway 1

Poland 1 1

Russia 2

South Africa 1

Sweden 1

Switzerland - 1

88 REPORT OF THE SECRETARY [J. S. M. p. E.

Territorial possessions of the United States provide one associate member who resides in the Philippine Islands.

In addition to the above, there are pending, 20 applications for Active membership, and 25 applications for Associate membership.

During this term 7 Active members and 4 Associates resigned; 10 Active members and 19 Associates were dropped for non-payment of dues ; 3 Active members were transferred to Associate membership, and 2 Associate members were transferred to Active membership.

NEW MEMBERS

The large increase in membership during the past year was primar- ily due to increased interest in the Society's JOURNAL, now published monthly; a more widespread knowledge of the Society's aims and accomplishments; and the noteworthy activity of the Membership Committee. A total of 199 new members were admitted during this past year to the Society. The sectional distribution of these new members is as follows :

New York Section 77

Chicago Section 21

Pacific Coast Section 23

London Section 38

Foreign countries, not including Great Britain, 40.

JOURNAL SUBSCRIPTIONS

While the number of annual subscriptions to the JOURNAL is not in- creasing as rapidly as was at first expected, satisfactory progress is being made in this direction. Subscriptions total 202, of which 173 are paid, 4 are free to Local Sections, 11 are exchanges with other publications, and 14 are free to sustaining members.

SALE OF JOURNALS AND TRANSACTIONS

Total sales of single copies of the nine issues of the JOURNAL to date, exclusive of the October issue, numbered 52, as contrasted with 1030 copies of back numbers of the Transactions, sold during the past fiscal year.

SECOND CLASS POSTAL PRIVILEGES FOR JOURNAL

After many months of negotiation with the Post Office Department, your Secretary is especially pleased to report that second class postal privileges have at last been granted to our Society in the matter of mailing the monthly JOURNAL. By reason of obtaining this privilege, which was granted only after certain requirements of the Post Office

Jan., 1931] REPORT OF THE SECRETARY 89

Department were met, a considerable saving in the cost of mailing the JOURNAL will be effected in the future, in addition to our obtain- ing a substantial refund on the mailing of past issues.

One of the more important changes required by the Post Office Department was in the matter of the subscription price to members of the Society. The ten dollar allowance for annual subscription to members was changed to nine dollars to make it less than the amount of the annual dues for Associate members. Notice to this effect is to be incorporated in new application blanks which will shortly be printed.

Respectfully submitted,

J. H. KURLANDER, Secretary

IMPORTANT NOTICE!

It is very necessary that all members of the Society and subscrib- ers to the JOURNAL immediately advise the General Office of the Society, when a change in mailing address is made. Otherwise, when literature is returned by the Post Office for this reason, the member's or subscriber's name is removed from the mailing list for the JOURNAL until the proper address is obtained. Future issues of the JOURNAL will contain, from time to time, lists of members or subscribers for whom no address is known. Anyone knowing the whereabouts of those members or subscribers is requested to advise the General Office promptly.

COMMITTEE REPORTS

REPORT OF THE JOURNAL COMMITTEE

October, 1930

In this report an attempt will be made to set forth the manner in which the JOURNAL OF THE SOCIETY OF MOTION PICTURE EN- GINEERS has been conducted since its establishment in January, 1930.

Following the autumn convention of 1929, at which time the publication of a journal was authorized by the Board of Governors, immediate steps were taken to set up the necessary machinery for publishing this journal at regular monthly intervals. The requirements and problems of publishing a technical journal were discussed with several publishing houses and bids on cost of publica- tion were requested from two or three of those which to the committee appeared most capable of handling this work. After careful con- sideration a contract was signed with the Mack Printing Company of Easton, Pa. The decision of the committee to give the contract to this concern was based not only upon its bid on cost, but also upon its proximity to the cities in which are located the editorial office and the offices of the secretary and treasurer.

The question of general style, typography, etc., was discussed at considerable length with the publisher and a style sheet was com- piled to serve as a guide in obtaining uniformity of style throughout the JOURNAL.

An attempt was made to get the machinery of publishing estab- lished sufficiently early so that the first issue could appear on January 1. As a matter of fact the January issue was a few days late but was mailed during the first week of the month. Since that time, with perhaps one exception, the JOURNAL has been mailed from the office of publication prior to the first of each month and in most cases has reached the members and subscribers within the first few days of each month.

The committee has endeavored to keep the contents of the JOURNAL in harmony with the wishes of the Board of Governors as expressed specifically in the resolutions passed at the time the decision was 90

COMMITTEE REPORTS 91

made to publish the JOURNAL. A statement of the material to be incorporated in the JOURNAL will be found in the JOURNAL XIV (Janu- ary, 1930) p. 8, under item 5.

In Table I following will be found an analysis of the contents of the JOURNAL up to and including September, 1930. The first section of the table analyzes the contents in terms of numbers of pages. The totals at the extreme right of the table indicate the way in which the space of the JOURNAL has been utilized. In the first nine issues a total of 1130 pages have been published, of which 895 were devoted to purely technical papers; 36 pages to abstracts of scientific articles which, in the opinion of the editorial office, should be of interest to the membership; 9 pages to book reviews; and 168 pages to society notes, tables of officers, committees, photographs of officers and committees, and material of general interest to the Society but of a non- technical nature. The committee feels that this disposition of the space is fairly well in accord with the wishes of the Society as set forth by its Board of Governors.

TABLE I

Jan. Feb. March April May June July Aug. Sept. Total

Analysis of Contents in Terms of Pages

108 895

4 36

1 9

22 168 3 22 138 1130

11 67 6 6 3

11 80

13 169

1 17

The second part of the table shows an analysis of contents in terms of material. Again in the total column it will be seen that of the total scientific papers published, 67, were obtained from our two semi-annual conventions, while 6 were contributed directly to the editorial office. Six committee reports were published and 3

Technical Papers	127	108	91	73	84	90	122	92
Abstracts	4	3	4	3	4	6	4	4
Book Reviews	2	2	1		1		1	1
Society Notes, Com-
mittees, etc.	13	9	4	27	37	9	14	33
Miscellaneous	2	2	2	3	2	3	3	2
Total pages	148	124	102	106	128	108	144	132
Analysis of	Contents	in Terms of Material
Convention Papers	8	7	11	8	3	5	7	7
Contributed Papers			1		2	1		2
Committee Reports	1	1		2	. .	. .	1	1
Translations	. .	. .			1			2
Total No. Papers	9	8	12	10	. 6	6	8	10
Abstracts	16	13	23	20	18	21	23	22
Book Reviews	4	3	3		2		2	2
Reprints ordered	2500 940	3850	3250	2500	2100	1150	1500

92 COMMITTEE REPORTS [J. S. M. P. E.

translated foreign articles, making a total of 80 scientific papers. The number of abstracts was 169 and book reviews 17.

In the last section of the table are shown the number of reprints which have been ordered from each month's publication. For the nine months covered by this table a total of 23,840 reprints have been ordered.

We now come to the very important question of how much the JOURNAL is costing the Society. In Table II will be found an analysis of the costs involved in publishing the first nine issues (January to September, inclusive) of the JOURNAL. In the first column will be found the number of copies printed for each month and in the second column the total number of pages in each month's issue. In the third column are shown the amounts directly chargeable to "editorial"

TABLE II

Month	Edition	Pages Editorial	Publisher Printing Cuts	Total	Postage	Reprints Cost Postage
Jan.	2500	148	$120.00	$801	.75	$75.85	$877	.60	$99.15	$108.32	$14.12
Feb.	2000	124	84.00	591	.25	74.89	666	.14	46.23	90.91	3.54
March	2000	102	75.50	509	.43	80.00	589	.43	48.85	123 . 10	7.54
April	2000	106	83.50	513	.35	139.74	653	.09	45.55	101.06	6.15
May	2000	128	120.00	619	.98	115.62	735	.60	42.16	116.16	6.88
June	2000	108	132.00	576	.49	98.74	675	.23	55.42	175.97	9.32
July	2000	144	135.00	653	.94	144.02	797	.96	65.03	116.33	10.21
Aug.	2000	132	151.00	632	.87	108.93	741	.80	62.42	95.53	4.27
Sept. Total	2000	138	112.00	633	.92	133.98	767	.90	87.01	130.27	7.14
	1130	1013.00	5532	.98	971.77	6504	.75	551.82	1057.65	69.17
Av.			112.55				722	.75

work. This includes the preliminary preparation of the manuscript for the printer. It should be pointed out that many of the manu- scripts as received by the editorial office require considerable work before they are in shape to be sent to the publisher. In some cases the drawings and illustrations submitted by the author are not satisfactory for reproduction. In some cases the editorial office has assumed the responsibility of having these redrawn, while in other cases they have been returned to the authors for correction. This item also includes all proof-reading, both of the galley and page. It includes all stenographic work and amounts paid out for the trans- lation of foreign articles. The item does not include any charges for postage, telegrams, and long distance telephone messages. These items were absorbed by the office of the editor pro tern and paid by

Jan., 1931] COMMITTEE REPORTS 93

the company by which he is employed. It is estimated that the total amount chargeable to these items up to the present time is not over $75.00. In the last column will be found the monthly cost of printing the JOURNAL. This includes all charges made by the pub- lisher with the exception of cuts and postage. In the next column are shown the monthly costs chargeable to the making of cuts for the illustrations of the JOURNAL, and in the next column a total of these two items which represents the actual cost of printing the JOURNAL. The monthly postage bills chargeable directly to the mail- ing of the JOURNAL are shown in the column so designated. In the last two columns of the table are shown the cost to the Society of the reprints ordered by the various contributors and the postage involved in sending these out.

In arriving at the total cost of the JOURNAL for the first nine months, we must include the following items:

Editorial $1013.00

Publisher 6504.75

Postage 551.82

Reprints 1057.66

Postage on reprint 69 . 17

$9196.40

Since reprints are billed to the author at cost plus 50 per cent, the profit on reprints should be subtracted from the above figure. This profit is $528.83. Subtracting this from the total cost we find that the JOURNAL for the first nine months has actually cost the Society $8667.57.

Let us turn again for a moment to the consideration of Table I in which the analysis of content is shown. It will be noted that of the total number of technical papers published only six may be classed as contributed, all the remaining material of this type being derived from the 1929 autumn and the 1930 spring conventions. The JOURNAL Committee had hoped that as soon as a monthly journal had been established there would be a goodly number of contributions other than papers read at conventions. We feel that in the future there will be more material of this character. There can be no doubt that in many cases the results of experimental and research work going on in various localities mature and are ready for publication at times between our semi-annual meetings. The JOURNAL Com- mittee would like to encourage authors to submit manuscripts

94 COMMITTEE REPORTS [J. S. M. P. E.

at any time. It is probable that greater activity on the part of the editor of the JOURNAL will be required to unearth this potential material. It would seem desirable to keep the number of pages published each month at a fairly constant level. We therefore hope that in the future more contributed papers will be available.

TABLE III

Members and Samples and Month Subscribers Back Orders Stock Special Total

Jan. 661 497 770 300 2228

Feb. 705 246 1081 45 2087

March 759 223 913 100 1985

April 809 195 1020 25 2049

May 769 238 1100 60 2157

June 834 111 1089 45 2079

July 871 147 1068 35 2121

Aug. 908 76 1075 45 2104

Sept. 951 68 1068 20 2107

Total 7267 1801

Moreover there is undoubtedly a large number of foreign articles which are well worth translating and printing in the JOURNAL. Here again, a regular editor with more time to devote to the search for such material would be an advantage. We should also like to see an expansion of the abstract section and an increase in the number of book reviews, provided books of sufficient value continue to appear from time to time as they undoubtedly will. It seems reasonable at the present time to plan upon a journal of approximately 150 pages per month. It will be noted that within the past nine months several of the issues have fallen considerably below this number of pages. If we assume a 150 page issue each month there is therefore space to accommodate more contributed and translated articles and some expansion of the abstract and book review sections.

The committee feels that the most important step now in the evolution of the JOURNAL is the appointment of an editor with suitable assistants to carry on the work and to develop the JOURNAL along the general lines as indicated.

LOYD A. JONES, Chairman J. W. COFFMAN H. T. COWLING

J. H. KURLANDER W. C. HUBBARD

Jan., 1931] COMMITTEE REPORTS 95

REPORT OF PUBLICITY COMMITTEE*

The present Publicity Committee was appointed directly following the Fall Meeting of 1929 at Toronto and has been actively function- ing since that time.

The work of the committee naturally divides itself into two parts, namely: providing news to the trade press and newspapers of the semi-annual meeting and the activities of the Society throughout the year.

The present Publicity Committee has served during one meeting, the May, 1930, meeting at Washington, D. C. At this meeting two news releases were issued each day of the Convention. As a result of this work at the May Meeting, 1400 inches of news or ap- proximately 20 newspaper columns were carried by trade papers and newspapers of which the Publicity Committee has accurate record. However, it is certain that a great deal more space was obtained since several of the stories were put on Associated Press and United Press wires and published in many newspapers throughout the country.

In furnishing news of the activities of the Society between con- ventions, the Publicity Committee has released more than 25 stories to the trade press.

When a story is released it is sent to all trade papers in the United States, all technical journals dealing with the motion picture in- dustry, and foreign motion picture trade papers and technical jour- nals. The list includes more than 30 publications altogether. Special stories have also been written for a number of publications, and reports of the May meeting were supplied to a number of tech- nical journals. Abstracts of all papers read at the May meeting were mimeographed and supplied to papers in this country and abroad.

Another duty of the Publicity Committee was to establish ex- changes of the Society's monthly JOURNAL with more than 30 motion picture and technical publications in this country and abroad. This exchange of publications has resulted in a great deal of publicity not only in this country but in some of the finest technical journals in European and other foreign countries.

Whatever success the present Publicity Committee may have ob-

* Presented at the Fall 1930 Meeting, New York, N. Y.

96 COMMITTEE REPORTS [J. S. M. P. E.

tained has been due not so much to its own work as to the splendid cooperation given by the motion picture trade-press. The Publicity Committee has found that the motion picture trade-press is extremely willing to offer every possible cooperation in publishing news regard- ing the activities of the Society and that its pages are always open for any legitimate news concerning the Society.

The Publicity Committee therefore wishes to express its apprecia- tion to the motion picture trade-press for its splendid cooperation in reporting the activities of the Society.

The Publicity Committee also wishes to thank all those in the Society who have cooperated with it and who have helped to supply the Publicity Committee with details of the Society's activities, for transmission to the press for publication.

WILL WHITMORE, Chairman

F. C. BADGLEY B. W. DEPUE

G. E. MATTHEWS G. F. RACKETT O. A. Ross

REPORT OF STUDIO LIGHTING COMMITTEE*

There has been little change in the methods of studio lighting since the report given at the Washington convention last May with but one exception, that there seems to be a tendency on the part of many of the studios, where incandescent lighting has been used to a very large extent, to increase the number of high intensity spots and sun arcs for floodlighting purposes. This has been rendered possible by the efficient silencing devices which have been installed on d. c. generating equipment and arc lamps in the various studios.

Manufacturers of arc lamp equipment are advertising new equip- ment for high intensity arcs which is claimed to be free from many of the causes of noise present in the older lamps.

None of the information which your committee has been able to obtain in the past six months is of a character which advances the real knowledge of studio lighting to any considerable extent. Basic information which is available with regard to the various sources

* Presented at the Fall 1930 Meeting, New York, N. Y.

Jan., 1931] COMMITTEE REPORTS 97

of light is given in many articles which have appeared in the Trans- actions of the Society, of which a bibliography was presented in the last committee report. Much additional knowledge can be obtained from the studios themselves, but in spite of very earnest efforts of the committee it has been impossible to obtain this information up to the present time.

We understand that in past years attempts have been made to utilize photometric measuring devices in the studios, but that none have been found satisfactory or useful for one reason or another. Some recent work has again been done on this problem, but up to the present time little progress has been made in the practical applica- tion of these instruments.

Continued efforts on the part of the committee should be made to obtain information from the studios which will permit the establish- ment of standards for desirable levels and types of illumination for the various kinds of sets encountered in the production of motion pictures. The methods which can be applied in this work probably lie in the determination of levels of illumination coupled with the photographic values of the light actually used and micro-density determinations on films taken with the various types and mixtures of illuminants.

A. C. DOWNES, Chairman L. J. BUTTOLPH R. E. FARNHAM K. C. D. HICKMAN M. W. PALMER

REPORT OF THE COLOR COMMITTEE*

The May report of the committee gave a list of producers of color pictures and the systems used. At that time the Photocolor Cor- poration1 report had not been received. Mr. A. G. Waddingham, technical director of the corporation, supplied the following descrip- tion of this system:

"The color camera is of special design, photographing a pair of images in conjunction with special taking-filters and an optical sys- tem employing the split beam method of photographing.

* Presented at the Fall 1930 Meeting, New York, N. Y. 1 Letter dated July 11, 1930.

98 COMMITTEE REPORTS [J. S. M. P. E.

"The negative is printed upon a specially designed optical printer which prints the two respective images upon duplitized positive stock.

"The print is then transferred to the green processing room where the film receives an application of the blue-green complementary dye on the side containing the image from the red sensation negative.

"It then passes through a red processing machine, wherein the orange-red dye is applied to the image from the green sensation nega- tive. At the termination of this operation the film is removed and sent to the assembly room where it is assembled and finally projected and inspected upon the screen."

According to Mr. Waddingham, the process is adaptable for the production of sound prints in color, either by the disk method or the sound-track on film method. The company is equipped with a thoroughly up-to-date laboratory, and a new sound studio is in the course of construction.

The Reporter, Hollywood, October 8, 1930, says the Photo- color Corporation of New York is planning to build a plant in Holly- wood with a capacity of a million feet of film a week and expects to be in operation soon after the first of the year.

FILM PACK

A specially made negative is being marketed, for use with the Film Pack system, known as Red Ortho Front Negative.2 This has a blue sensitive emulsion on the surface of which is a layer containing a red coloring matter.

In making color sensation negatives by this system, Red Ortho and a panchromatic negative are placed emulsion to emulsion in the camera and exposed simultaneously. The light from the lens passes through the Red Ortho, recording the blue end of the spectrum. The red colored layer then filters out the blue; the red end of the spectrum passing through is recorded by the panchromatic negative.

The red coloring matter on the Red Ortho is removed from the de- veloped, fixed, and washed negative by bathing in a 3 per cent solution of hydrosulfite of soda.

NEW COLOR PROCESS

A new color process is being introduced from Germany, known as the "New Color Process." It is claimed that this is usable for either motion picture or stills, although in the description the method of using it for motion pictures was omitted.

2 English Provisional Patent No. 333,933, August 25, 1930.

Jan., 1931] COMMITTEE REPORTS 99

Successive exposures are made in a special camera fitted with tri- color filters. The color value negatives are printed on positive films which have their respective dyes incorporated in the emulsions. The films are then developed, fixed, washed, and subjected to a warm water treatment. No formulas were given. The silver images are then reduced, leaving pure dyed images which, it is claimed, can be either transferred to an individual support, or the three films can be placed in register and bound. The printing is accomplished by printing through the celluloid side of the film.

THREE-COLOR ADDITIVE PROCESSES

In the Herault Color Process a three-color sector wheel is rotated in front of the camera and the contact print negative is dye tinted so that each successive group of frames is tinted one of the primary colors. The three-color positive is then projected with a continuous projector (Continsouza-Combes) . The method is said to suppress the chromatic flicker when projected at 24 frames per second; only spherical lenses are used in this projector. This plan is somewhat similar to that now being suggested by Wolf-Heide.

HORST SYSTEM OF COLOR PHOTOGRAPHY

In this system three pictures are taken simultaneously with three- color filters, using a prism system in the camera. In the positive, each