Three-Dimensional Photography - Principles of Stereoscopy

(Frankie) #1

82 THREE-DIMENSIONAL PHOTOGRAPHY


are stamped with the consecutive number of the films as well as
the semi-circular and square notch symbols so that there is no
danger of confusing the right and left images. They are marked
in the camera and cannot be confused.
Because so many 35mm stereograms are destined for projection,
and because projection slides should be more precisely aligned
than others, what are the limits?
This all depends upon the individual spectator, so all that can
be said is to give a low average for the limits which apply to the
alignments discussed below.
The two images should not have more than one-half prism
diopter vertical deviation, that is a deviation which a one-half
diopter prism would compensate. This is equal to one unit de-
parture from identical positions for each 200 units of screen dis-
tance. So, if the screen is located four meters from the spectator,
the greatest advisable vertical difference would be two centi-
meters, or not quite one inch. Most stereographers prefer half
that as a working limit. And remember the angular displacement
of the visual axis; the amount of movement demanded of the eyes
depends upon the distance from screen to spectator, not screen
to projector.
Laterally the deviation should be less than three prism diopters
or about 12cm, say five inches, for the most comfortable viewing,
but usually the spectators can tolerate four and sometimes five
prism diopters provided the relative positions are such that this
means convergence, never divergence. Most people find a half
prism diopter divergence uncomfortable, while one prism diopter
divergence is impossible for many people. Exact limits of all
vergences depend, as has been repeated so often, upon the indi-
vidual. (Distances given assume qoocm screen distance.)
Figure 5-7 shows the most frequent mounting errors in exag-
gerated and diagrammatic style. A is the normal mount with the
two images correctly aligned.
B shows the result of having the camera tipped endways. Such
a pair must be mounted as shown by the dotted lines in the right
figure so as to preserve the original horizontal alignment, but if it
is more than a degree or so off vertical, no correction should be
attempted because painful eye “twisting” will result.

Free download pdf