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

be so simply achieved until it becomes possible to create real or virtual 3-D images in space." No protagonist of an "interocular" camera-lens inter axial spacing, certainly not I anyway, would have said that a fixed interaxial had got anything whatever to do directly per se with strain-free viewing; it should be obvious from my own paper, which follows Prof. Rule's classic lead, that the conditions for strainfree viewing to which I adhere lie primarily and essentially in the avoidance of too marked a deviation from the normal convergence/accommodation ratio in viewing, and in the avoidance of divergence of infinity points in projection. Elsewhere the Authors are very properly insistent upon the necessity of avoiding distortions. Throughout their argument it is not perhaps made obvious enough in their advocation of a technique based on a variable interaxial that the onus of providing, in major part, the illusion of three dimensions in the majority of shots, is placed by them mainly upon the minor factor of convergence; nor, correspondingly, that their adoption of an interaxial camera-lens spacing of other than interocular must lead to the irrevocable recording of such disparate images on the film hi the camera which, being once shot, cannot by any convergence or other means subsequently in the transmission procedure be made to give to the viewer that natural disparity between images which is the major and essential factor in a complete and lastingly convincing illusion of natural depth. This, to my mind, constitutes a major distortion which the advocates of a "Human-" or "NaturalVision" fixed camera-lens interaxial spacing technique would I am sure regard as one which viewing audiences would not tolerate indefinitely. However, my comments in the last paragraph above could be regarded as only an expression of a personal opinion and I would be far from expecting the Authors necessarily to agree with me, merely on my say-so! Clearly what is at issue here is the whole matter involved in the relative merits and demerits of techniques based mainly upon either a fixed or on a variable camera-lens interaxial or, if it comes to that, upon the fixed convergence technique of Dr. Reijnders as well. Such an issue cannot be much furthered by a mere exchange of letters and I hope that I may find the opportunity to sustain at length in a considered paper the argument as it appears to one of the fixed-interaxial school of thought. In that event, nevertheless, I shall be hard put to it to equal the able manner in which the Authors have put forward their own. H. Dewhurst Ministry of Supply Telecommunications Research Establishment Great Malvern Worcs., England December 12, 1952 [from the Authors] We appreciate Mr. Dewhurst's kind words about our paper, "Basic Principles of the 3-Dimensional Film," and feel that there is no fundamental difference of opinion between us, but rather a difference of approach. We are in full agreement that the vital requirements of strain-free viewing are careful control of the convergence/accommodation ratio presented to the spectator, and in particular the avoidance of divergence on background images. It j is in order to meet these requirements under the conditions of studio shooting that we have developed our two-stage technique of careful measurement of the depth-range presented to the camera, and adjustment of the interaxial separation to keep the images within the strain-free range. The shot itself will on occasion need to be altered from that first proposed by the director, if the depth range it contains is too great to be compassed even at the minimum interaxial. Mr. Dewhurst advocates that the retention of the images within the strain-free range be achieved without alteration of the lens separation from 2.5 in.; this it would be possible to achieve only at the cost of exercising far more drastic control of the permissible depth-content of the scene being filmed, and hence limitation of the freedom of expression available to the director. We feel that Mr. Dewhurst only regards this as the more desirable alternative because he is accustomed to working with very small magnifications where the depthrange limits during filming become proportionately less restrictive — the maximum screen size referred to in his paper is 21 inches only. If he were face to face with 640