American cinematographer (Oct 1933)

October 1933 • American Cinematographer 213 Control In the Processing of Motion Picture Film by Emery Huse, A. S. C.* Editors Note: This is the concluding article in the series written by Emery Huse, A. S. C., on Sensito- metric Control in the Processing of Motion Picture Film. In November Mr. Huse will start a new series of articles written in his usual interesting style. VARIABLE DENSITY TRACK In the consideration of the Western Electric system of recording a much more detailed sensitometric discussion is necessary. In processing variable density sound records which utilize the straight line portion of the H and D curve, it is necessary that the overall gamma characteristic as determined by plotting projection densities versus the logarithm of light valve openings be held to the ideal value of unity. This might also be expressed by saying that the product of the positive and negative gammas multiplied by the projection factor should be 1.00. Before specifying numerical values for positive and nega- tive gamma, several factors entering into this computation must be explained. The gamma value obtained from a series of exposures on a light valve recorder and plotted against the logarithm of the light valve openings will be designated as the light valve gamma, LVy. The gamma valve obtained on the control strip which has been ex- posed on the Type lib sensitometer and developed with the light valve gamma strip will be called the negative control gamma and designated as Cy. In the measurement of positive gamma it is necessary to measure the apparent printer gamma, which is obtained by printing the negative control strip on a printer and develop- ing this print with the positive sound track print. This valve will be called APy. When the positive sound track print will be developed the positive control strip which has been exposed on the Type lib sensitometer and the gamma value obtained on this strip will be designed as PCy. Further, inasmuch as the positive sound track in pro- jection will be scanned by a photo-electric cell, it is neces- sary to determine experimentally the difference in gamma as determined by visual measurement of the diffuse densities and the quasi-specular measurements of the photo cell. This factor has been measured for standard projection conditions and found to be 1.30. The conditions for correct reproduction as recommended by representatives of the Electrical Research Products, Inc., are given in the following series of data: Overall gamma=LVy x APy x projection factors 1 (1 ) If LVy=a x NCy and (2) APy=b x PCy, then (3) Overall y=NCy x PCy x a x b x projector factor (4). Hourly Control Tests For Positive Film ^26/33 Fig. 7 at top; Fig. 8 at bottom. It is customary to assume that the factor a=l, i. e., the difference between the negative control gamma and the light valve gamma is negligible. In practice the light valve gamma is found to vary by plus or minus 5% from this. The printer factor, b, is usually measured daily in most laboratories, at least such a procedure is recommended. While the printer factor may be as much as 10% it is found in practice that this value is approximately equal to the factor, a, and opposite in direction so that the two tend to cancel each other. Omitting these factors then from equation (4) we have Overall gamma=NCy x PCy x projection factor (5) and substituting 1.—NCy x PCy x 1.30, or (6) NCy x PCy=0.76 (7). In other words, any combination of negative and positive control gammas which gives a product approximating 0.76 would be correct for straight line recording. Tests have been made in four Hollywood studios, which are listed by letter, of the actual Type lib sensitometer nega- tive and positive gammas. In table 6 these data are given. TABLE 6 It will be observed that the maximum variation from 0.76 is 0.15 higher. One studio at the time these tests were made operated slightly lower than the desired 0.76, their visual product being 0.72. A mean exposure of the light valve that will permit 90% modulation without going into the toe of the negative H and D curve is recommended. This results in an average density numerically equal to the (Continued on Page 238)