Journal of the Society of Motion Picture and Television Engineers (1950-1954)

Mr. Athey: We do not have any proper evidence, but it is my impression that the situation is exactly backwards from that. However, perhaps some of the others have some experience with it. Dr. Frayne: I was going to add that the only problem that we've had so far is with so-called tonying of the film.* If you get a film with a bad tony in it, then, of course, you get a very bad contact, but if the film is in normal condition then there's no particular problem. Walter T. Selsted (Ampex, Redwood City, Calif. This statement read by Charles H. Wirth, Ampex Corp., New York District Office}: "As most of you know, Ampex Corp. has recently developed studio and field recording equipment for the Magna Theatre Corporation. Recently we have developed a complete line of theater sound reproducing equipment designed specifically for the reproduction of CinemaScope pictures. "During our work on these projects, we made extensive investigation of the relationship which pre-emphasis and postemphasis bear in respect to weighted signalto-noise ratio and distortion. The preemphasis and post-emphasis characteristics established through the NARTB were based on studies made relating soundsource energy distribution with saturation characteristics of magnetic recording media. The studies of energy distribution made by Bell Telephone Laboratories in 1929 were recently repeated by Ampex Corp., which found excellent agreement. Recently repeated tests confirm that the NARTB standard characteristics are an optimum compromise between unweighted signal-to-noise ratio and high-frequency tape saturation. Great improvement can be realized in the listening quality of narrow-track magnetic recordings now being introduced for motion-picture sound, if the record pre and post-emphasis at high frequencies are reconsidered, placing more importance on weighted signal-tonoise ratio and putting less importance on occasional high-frequency compression due to magnetic saturation. In design of the * Fluting or waviness along the edge of the film, usually on one edge only but sometimes on both. See Common Causes of Damage to 35mm Release Prints, Motion Picture Film Dept., Eastman Kodak Co., Rochester 4, N.Y. equipment for Magna Theatre Corp., we appreciably altered the pre-emphasis characteristics in the recorder with a resultant improvement in weighted signal-to-noise ratio of better than 10 db. Recently, studies we have made indicate that, using similar techniques and suitably choosing the recording material, an improvement of approximately 16 db can be achieved on a weighted basis. These improvements in signal-to-noise ratio are primarily in the upper end of the audio spectrum where tube and tape hiss are the objectionable quality of which the listener is immediately conscious. The techniques we have used indicate an improvement of at least 7 db throughout the rest of the spectrum." Ampex has had such good results with these recently developed changes in equalization that we should like at this time to propose that the Ampex equalization characteristics be accepted as an industry standard as well as a standard adopted by the Society of Motion Picture and Television Engineers. With your permission, I should like the opportunity to show you the two slides I have brought with me on which the equalization curves which Ampex proposes are shown. The first slide shows the record amplifier equalization characteristic. The one presently used by the motionpicture industry is shown on Curve 1, the only exception being that the lowfrequency pre-emphasis used with CinemaScope films Ijas been added (Slide 1 ). Curve 2 shows the increase in highfrequency pre-emphasis which we find is not only acceptable but highly desirable, and the 6-db spread between the two at low frequencies merely illustrates the approximate increase in output from the tape realizable by means of the new oxide developed by Minnesota Mining. Slide 2 shows the necessary playback equalization curves to complement the record characteristics in each of the two former cases. As you will notice, the spread between the two curves beyond 1000 cycles increases quite materially. Since it is in this region where the most audible hiss occurs, the improvement in signal-to-noise ratio at any particular frequency is approximately equal to the spread between these two curves. We wish to recommend these Ampex Stereophonic Equipment Discussion 231