US2635047A - Method of processing sound film - Google Patents

Description

April 14, 1953 s. M. BEST 2,635,047 METHOD OF PROCESSING scum: FILM Filed July 27, 1949 I5 06 ExpasueE awe/whom 7 arraeuc-V Patented Apr. 14, 1953 METHOD OF PROCESSING SOUND FILM Gerald M. Best, Beverly Hills, Oalif., assignor to Warner Bros. Pictures, Inc., Burbank, Ca1i1'., a

corporation of Delaware Application July 27, 1949, Serial No. 106,993

1 Claim. 1

This invention relates to the production of sound motion pictures, and particularly to methods of processing and printing sound records to accompany motion pictures.

In the production of composite motion picture film-the pictures and sound are first photographed and recorded, respectively, on separate films, which are separately processed and then eventually combined onto a single film for projection and reproduction. Two major types of sound recordings are in commercial use, one being the constant width, variable density type of record, and the other being a varying width, constant density type of record, also known as a variable area record. The present invention is directed to the variable area type of record, wherein distortion sometimes occurs in the form of a varying density component. In the recording of film sound records, the iilm I is sometimes subjected to undue strains and/or pressures which desensitize the original negative at the points of pressure or strain. For instance, should the film be curved sharplyas it leaves the magazine, or any other point in its path between the supply reel and takeup reel, strains may occur therein, particularly between the sprocket holes or perforations. When such a negative is printed to a print, a varying density component is introduced, and, because of the occurrence of the sprocket holes, exhibits itself as'a comparatively high level, 96- cycle tone, which renders the recording non-commercial. The variation in density may range from minus ten decibels to minus forty decibels. The present invention is directed to printing and processing methods for eliminating this variable density component, while providing asatisfactory commercialfinal print.

, The principal object of the invention, therefore, is'to facilitate the production of photographic film sound records.

-Another object of the invention is to provide an improved method of eliminating a variable density component in a variable area type of by referring to the following description, read in conjunction with the accompanying drawings, forming a part hereof, in which:

Fig. 1 is a graph showing H and D curves for difierent film emulsions on which the sound record is recorded and to which the sound record is printed, and

Figs. 2, 3, and 4 are partial sections of sound film illustrating th methods of the invention.

Referring now to the drawings, the original sound record negative used commercially for film sound recording is generally of the Eastman type 1372, with a base density of .31, and processed to a gamma of 3.70. Base density, as used herein, is the density of the Celluloid base on which the emulsion is coated at the factory. The H and D curve for this film is shown as curve A in Fig. 1. It has been found that, with the variation in density occurring from sharp bends in the film or other sources of pressure or strain, the density thereof ranges between a density of 1.5 to a density of 2.9, which variation is not detectable on the negative by the naked eye. This is illustrated in Fig. 2, wherein an unmodulated variable area film 1 with sprocket holes 8 is shown with bias lines 9, which have been exposed to light and in which no variation in density is discernible, but in which there can be a variation in density between 1.5 and 2.9. The Width of the bias lines have been slightly exaggerated for better illustration of the invention.

This range of densities between 1.5 and 2.9 is shown by bracket 5 on curve A. Thus, the variable density component is not detected until the original sound negative is normally printed to an Eastman type 1302 film emulsion, which has a base density of .05. To provide the optimum sound record in normal processing, this print is developed to a gamma of 2.32, and the H and D curve therefor is shown at B. The density range of 1.5 to 2.9 now appears on the print between the range of the base density of to .5, as shown by bracket H. Since these densities represent a variation in output from the sound record range between approximately minus ten decibels to minus forty decibels, this 96-cycle variation has a level which renders the print non-commercial. This range of variation on the print is very easily discernible by the naked eye, as shown in Fig. 3, which is a print from the film of Fig.2.

To eliminate this variable density component, chemical reduction methods have been used, but these are not commercially feasible for several by bracket I3.

Referrin again to the drawings and to further describe the invention, the print developed-in accordance with curve B and shown in Fig. 3 is again printed to the same Eastman type 1302 emulsion to provide another print'o'r' what is known as a dupe negative, but which is developed to a gamma of 2.05, as shown by curve C. The

first print shown by curve B is printed to a normal density of 1.85, while the dupe negative is printed to a normal density of 2.80. Normal density, as used in this description, is the density determined by cross modulation tests, as described by Baker and Robinson in the Journal of the S. M.-P. volume 30, for January 1938. Non-technically, it is the density which produces the best possible sound ,prints as indicated by cross modulation tests. The normal density above stated provides a lighter than normal print,

but which is the normal density best suited to eliminate the variable density noise component.

The range of density variations causing distortion or noise has now been reduced to a narrow range between a density of 2.7 and 2.8, as shown This dupe negative will appear similar. to the original negative shown in Fig. 2, since the density variations occur above a density of 1.5.

The next step of the invention is to again make a print on emulsion 1302 from the dupe negative, as shown by curve B, at a normal density of 1.4. This final step will reduce the range of noise to such a small value as to be unmeasurable and unreproducible, since the limits will be practically the same as the base density of .05. The second print, therefore, will appear as shown in. Fig. 4,

the. improvement therein being observable by comparing Fig. i with Fig. v3. Slight variations from these gammas and normal density values willv provide satisfactory results. In the claim,

three prints are mentioned, theduplicate negative being the second print, and the. final or second print being the third print, while in Fig. 1, the second print referredto is the final print.

The above method. has been found to be satisfactory ineliminating the variable density component as illustrated in 3; but, there has been some harshness in the sound when reproduced from the final print. To eliminate this harshness, a second process is provided which utilizes the toe of the characteristics; instead of the shoulder, as just described.

Referring again to the drawings, we have the original negative processed in accordance with curve A and shown in Fig. 2, but this negative is 4 lower gamma, such as 1.23, as illustrated by curve D. The normal density thereof is 1.0. This changes the variable density range of 1.5 to 2.9 to a range of .05 to .12 on the print, as shown by bracket E5 on curve D and illustrated in Fig. 3. From this print, a dupe negative is then made on the same emulsion developed to the samegamma, which transfers the density range from 1.14 to 1.20 on the dupe negative, shown by the bracket [6 on curve D. From this dupe negative, a second print is made with a normal density of 1.05, which reduces the density component of the second print to a variation which is so smallas to be unreproducible and a noise level which is at the threshold of audibility in the neighborhood of minus forty decibels.

- This is illustrated in Fig. 4. Furthermore, not

only is thenoise eliminated, but the sound has substantially its normal quality as originally recorded.

These methods, therefore, provide satisfactory final commercial prints from non-satisfactory originalnegatives or prints, and which can be accomplished in the regular commercial laboratories with materials always available. The final commercial prints are used for rerecording to a composite master negative.

Iclaim:

The method of obtaining a variable area sound print having no variable density component from a variable area sound negative havin such a variable density component, comprising printing said original negative to a first print for shifting said variable density component to the curved toe portion of the exposure density characteristic to reduce the range of said variation, printing a second negative from said first print to shift said range to an upper linear portion of the exposure density characteristic, and printing a second now printed to Eastman typepositive 1302 to a. L

print from said second negative for shifting Said variable density component to the curved toe portion ofthe exposure density characteristic to re.- duce the density range to an unmeasurable value, the emulsions of said first print, second negative,

and second print being printed to 'a gamma of 1.23 at normal densities of substantially 1.0, 1.2 and 1.05, respectively.

GERALD M. BEST.

References Cited in the file of this patent UNITED STATESPATENTS Name 7 Date Baker Oct. 31, 1939 Steube June 20, 1944 Dip-pol July 31, 1945 FOREIGN PATENTS Number Number

Images