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

be respected in order for successful stereography to result. The necessary relationship between the 3-D film and human vision is immediately apparent: the errors in transmission cannot be greater than the limitations imposed by the spectator's visual ability. Steadiness of the Composite Image The two aspects on the screen will usually vibrate with respect to each other. Such vibration is caused by a lack of film registration in cameras, printers and projectors, and by mechanical vibrations in apparatus. Under such conditions the fundamental frequency of the unsteadiness in each aspect is 24 cycles/sec. In order to determine whether such unsteadiness can be adequately handled by the average spectator's visual abilities, it is necessary first to define such unsteadiness in terms related to physiological optics, and secondly, to determine the magnitude of the unsteadiness. This unsteadiness is adequately expressed in visual terms by the variation in the angle formed by straight lines passing through the right and left eyes and their respective point pairs on the screen. In the horizontal plane this angle is referred to as the convergence angle, while in the vertical plane it is known as the angle of vertical divergence. As regards the magnitude of the intermittent variation in these angles, it has been determined that even in the extreme case of 16mm filming without registration pins, enlargement to 35mm, and projection on a standard 24-ft screen, the maximum intermittent variation in the convergence angle for a spectator seated 75 ft from the screen is about 6 ' of arc, while the intermittent variation in vertical divergence is no more than 3.5 ' of arc.1 Is this order of variation visually acceptable? Since the values cited will probably be the maximum en countered in the 3-D film, these values will be referred to the limitations of the visual apparatus, for, if it can be shown that these extreme values can be handled, it will be known that less demanding cases will certainly be acceptable. First a determination should be made of whether such a variation creates the perception of a vibrating composite image. With the discriminating device (Polaroid glasses, for instance) removed, both aspects are seen by both eyes. Now the two aspects are seen to vibrate with respect to each other, for the visual Snellen acuity of the average spectator for normal motion-picture brightness is about one minute of arc, while the visual vernier acuity, which is more significant, is even less. With the application of the discriminating device can the spectator still detect movement in the composite image? With the discriminating device inducing fusion, individual perception of either aspect is absent. Thus, the detection of unsteadiness can only be the detection of unsteadiness of the composite image. Conceivably this unsteadiness might be detected if, by fixation, each eye were to follow its respective vibratory aspect movement. However, the reaction time of the extraocular muscles is about one-tenth of a second,2 so the eyes cannot possibly follow the 24-cycles/sec image movement due to a lack of film registration. Whether the reflex period is influenced by certain cerebral rhythms3 or by the time delay at the synapses (junction points between the brain and the receptors in the retina) is not of particular importance for this discussion except that if it is the latter, there exists the indication that movements of points other than those fixated could be detected even less because of the added number of synapses for receptors outside the fovea. Nevertheless, stereopsis exists even in the absence of willful innervation of the 200 March 1954 Journal of the SMPTE Vol. 62