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

The finer detail permitted in the second picture is exemplified by the narrower spacing between scanning lines, and by the shorter dimension of the picture element along the scanning line. The tolerance to be put on the echoes depends in part, as was already noted, upon their spacing from the original picture. Before proceeding to evaluate these comparatively in A and B of the figure, we can note a suggestion made by the author in 1951 for a similar comparison in the tolerance on random noise.10 This suggestion was to the effect that allowance for the improved quality of the theater television be reflected in the engineering of the tolerances. Specifically it was proposed that the tolerance be engineered on a closer viewing distance for the better picture. The ratio suggested was 1 to \/2. In the present case this would enlarge the image area at B to that shown at C. Here the picture elements then become of exactly the same size and shape as those seen at A. If the field of view in the magnified area is cut down in the diameter ratio of \/2 to 1 as indicated by the dotted circle in C, the picture becomes identical to A. If we apply the suggestion, it means that all tolerances which have been set on envelope or phase-delay excursions can remain the same if the tolerances are measured in picture-element durations. If the tolerances, however, are measured in microseconds, they will be halved for the 8-mc band as compared with their values for the 4-mc band . The tolerances on phase-shift excursions where applicable are not a function of echo spacing and are therefore not changed in going from broadcast to theater television. These deductions have so far assumed a 740-line theater television. If the latter uses the broadcast standard 525 lines, with an 8-mc band, the magnified field of view becomes that shown at D in Fig. 20. Here the picture-element height as measured by the scanning-line spacing is greater than in C in the ratio of \/2 to 1 . In compensation, the picture element width is narrower than in C, in the ratio of 1 to \/2. This last correspqndingly narrows down the geometrical spacing of all echoes and permits increases in the tolerances on envelopeand phase-delay excursions in the ratio of \/2 to 1 . This result indicates a certain advantage for the choice of a smaller number of scanning lines in the picture. Such an idea is not new, and was one of the factors discussed when the standards were set for broadcast television.11 Appendix Consider a Fourier component of the signal voltage of frequency «/2ir: V = COS CO t (3) When this is transmitted through a distortionless system, which however delays it by the time T, it becomes : v = cos co (/ — T] (4) If the system also transmits an echo of the signal, of relative amplitude a, and relative delay T to it, the distorted signal becomes : v = cos co (/ T) + a cos co (/ T — r) = cos co (/ T) + a cos co (t T) cos COT + a sin u> (t — T) sin COT (5) = (1 + a cos COT) cos co (/ — T) + a sin COT sin co (/ — T) (6) The overall transmission, including the echo, can be interpreted as: v = a (co) cos [co/ — <f> (co)] (7) where a (co) gives the overall amplitude ratio and <f> (o>) gives the overall phase shift. These quantities are computable from equation (6), and are particularly simple when the relative echo amplitude a is small. That is: 594 May 1953 Journal of the SMPTE Vol. 60