Journal of the Society of Motion Picture Engineers (1930-1949)

ment may be only slightly affected, the phototube of the usual sound reproducer will see a reduced image contrast with definite effect on the reproduction of sound. Measurements of silver sulfide sound tracks by densitometers with phototubes having S-4 response or by visual instruments are to be discouraged. Standardized Densitometry The interchange of useful information on sound track printing conditions would be greatly facilitated by standardized densitometry. The method most likely to produce an acceptable degree of correlation requires the use of an interference filter2*3 to restrict the response of the densitometer to a bandwidth of a few millimicrons. The density characteristics of one type of interference filter are shown in Fig. 4. With such a filter, the effects of variations of the wavelength of peak sensitivity, shift of peak due to color temperature of the light source, the effects of aging, and differences in spectral sensitivity of phototubes are virtually eliminated. In fact, practically the only sources of error that remain are due either to the geometry of the instrument or to error in the determination of the wavelength of peak transmission of the filter. The effects of tube responses are eliminated so that the same density is indicated whether an S-l or S-4 response is used. Of course, the sensitivity with an S-4 response is so much less that it would seldom, if ever, be used. The filters are relatively dense so that as much as two density ranges may be lost. This restricts the choice of wavelengths for standardized densitometry. If 550 m/i were chosen, the required unfiltered range for the densitometry of a sound track on Kodachrome film would be 0 to 6, since it would be necessary to be able to read to a density of about 3.2. However, at 800 mp. the unfiltered density range required would be only a little more than 3, since the maximum density at 800 mju of the reversal silver sulfide sound track on Kodachrome film is now under 1.0. Acknowledgment The author wishes to acknowledge the contribution of Jack Pinney and Edward Letzer of the Color Control Division in obtaining materials and data for study of the problems of densitometry of silver and silver sulfide deposits. References 1. Zworykin and Ramberg, Photoelectricity and Its Applications, John Wiley, New York, 1949, p. 15. 2. Bruce H. Billings, "Narrow band optical interference filters," Phot. Eng., 2: 45-52, 1951. 3. Harry D. Polster, "A symmetrical alldielectric interference filter," /. Opt. Soc. Am., 42: 21-23, Jan. 1952. Discussion John G. Frayne (Westrex Corp.} : I'd like to ask if those figures given for density are diffuse density? Mr. Lovick: They're diffuse. These densities are measured on the Western Electric RA 1100B Densitometer modified for this purpose. Dr. Frayne: With the interference filter in the optical system, what density are you able to measure with the existing amplifier? Mr. Lovick: We have modified the amplifier so that even with the additional filter we still get a density range of 4. Dr. Frayne: Is that information going to be available? We'd like to know about it. Mr. Lovick: You mean how to modify the amplifier? Dr. Frayne: Yes. Mr. Lovick: It's only necessary to change a few resistors in the preamplifier. Dr. Frayne: Even that's worth knowing. Are the interference filters available? Mr. Lovick: Yes. We've obtained some from Bausch & Lomb. The tolerance which I think we should require is about plus or minus 2 millimicrons in order to get good correlation. Robert C. Lovick: Densitometry of Sulfide Tracks 93