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

to the human interocular. However, it is emphasized that the rule just given for the limitation of the image in the theater is only approximate, and research is needed to determine the extent, in both magnitude and period, to which infraction of this rule is possible for the unique conditions of the 3-D film. Experimentation should be undertaken in the theater, not the clinic, so that usual motion-picture conditions apply. It may well be that for such conditions some divergence can be tolerated with comfort, particularly if the nearest plane is not as close to the spectator as has been shown possible. Such a possibility has its advantages in the mechanics of filming.17 Vertical Disparity of Point Pairs So far only intermittent movements of point pairs have been considered. However, mechanical, optical and human errors in transmission will also result in steady errors in the placement of point pairs. Such steady errors will usually last for at least the duration of the shot. The effect of steady errors in the distance between point pairs in the horizontal direction is to create an error in the intended location of the composite image. The extent of this error in location is a function, partly, of the psychological condition of the spectator,1 and hence will not be considered in this paper. The effect of steady errors in the distance between point pairs in the vertical direction is to create a condition forcing one eye to be angled upward. This unnatural act may cause either fatigue or diplopia, depending on its magnitude. The extent to which one point of a point pair can be elevated without causing diplopia will now be investigated. Ellerbrook separated point pairs in the vertical direction in order to find out at what separation diplopia first occurs.8 He found that fusion still maintains as long as the amplitude of vertical divergence is not much greater than 1°, a value corroborated by Davson.7 Again, for the same extreme case of 16mm filming with 35mm projection in a regular theater, it has been computed that the addition of all normal mechanical, optical and human errors cannot cause a vertical divergence greater than 14' of arc for a spectator seated 75 ft from the screen.1 This maximum value is for point pairs located at the top or bottom of the screen; vertical disparity between point pairs at the center of the screen will be considerably less. Even for the closest spectator to the screen, the vertical divergence induced by a vertical disparity in point pairs is much below that necessary to cause diplopia. It has already been mentioned that Ellerbrook found that the ability to fuse vertically disparate point pairs increases as the fusional stimuli approach the periphery of the visual field.8 It is fortunate that a greater amplitude of vertical divergence is possible for peripheral stimuli, because vertical disparity of point pairs increase: at the corners of the screen due to the convergences of cameras and projectors (keystoning) . Disparity in Magnification It can be shown that the convention of always keeping longer lenses on the same side, say the left, in both filming and projecting would help minimize the effect of the disparity in the magnification of the two aspects due to a disparity in lens focal lengths.1 Nevertheless, such a disparity in magnification may exist, and, as can be shown, may cause an error in the apparent location of the composite image, particularly at the side borders where horizontal disparity is most evident.1 Magnification disparity may also cause headache, fatigue or diplopia, particularly for images near the top and bottom borders where vertical disparity is most pronounced. Still another aberration may be caused 204 March 1954 Journal of the SMPTE Vol. 62