588 HANDBOOK OF PHOTOGRAPHY 



Transparencies vs. Prints. — The question of the use of transparencies as opposed 

 to prints is largely a matter of personal preference. 



The light effect of transparencies gives a much greater illusion of the actual exist- 

 ence of light-filled space. They also have a brilliance which cannot be obtained with 

 prints. Furthermore they permit the use of color in the same way that it is used in 

 colored films. For these reasons they are generally considered to be superior. 



On the other hand prints are much easier to handle and not so susceptible to dam- 

 age. They are much easier to view as individual photographs. They require less 

 light intensity and can usually be viewed in any position without searching for a source 

 of light, up to which they can be held. A number may be viewed, as single photo- 

 graphs, at once for purposes of selection. For these reasons they usually give greater 

 satisfaction to the beginner. 



Hjrperstereoscopy. — If the separation between camera lenses is increased beyond 

 the interocular distance the two resultant photographs of any object will be the same 

 as could have been taken, using the interocular distance, of an exactly similar object 

 both smaller and nearer the camera. 



On convergence principles if the base is increased m times the object will appear 

 1/m the distance and l/m the size. Thus, if the base is doubled, the resultant object 

 should appear as a "reduced model" of half the size at half the distance. 



Since the ability to detect depth is much greater at greater convergences (see Table 

 I) the objects in this "reduced model" will appear to stand out from their backgrounds 

 much more obviously than in an orthostereoscopic view. The perception of depth is 

 consequently enhanced. 



As the base increases, the convergence angle consistent with a reduced model inter- 

 pretation becomes greater and soon passes beyond any value natural to binocular 

 vision or stereoscopic viewing. Certainly at some point the eyes cease interpreting 

 the image as a reduced model and begin interpreting it as a similar object of increased 

 depth. This exaggeration of depth then increases as the base is further increased. 



The entire phenomenon of the interpretation of hyperstereoscopic images needs 

 more thorough investigation before more exact statements can be made concerning it. 



In aerial photography it is usually assumed that the most natural relief, or the 

 nearest approach to a properly shaped image, is obtained when the two views of a 

 point on the ground appear the interocular distance apart in relative position on the 

 two plates. The base for this most natural relief can be calculated from the formula 



B=f (5) 



where B = distance between exposures in feet; 

 e = interocular distance in inches; 

 d = height of camera above ground in feet; 

 / = focal length of camera in inches. 



If the base as obtained from this formula is exceeded, the resultant stereoscopic 

 image is certainly exaggerated in depth. Bj^ this means low buildings can be made to 

 appear as skyscrapers and normally imperceptible details of relief can be made 

 apparent. 



The hyperstereoscopic effect, if judiciously used, can add greatly to a stereograph's 

 effectiveness by its strong emphasis of the depth quality. It is best used on mountain 

 or landscape views which have few foreground objects. 



From considerations of the limits of stereoscopic vision, objects over 2000 ft. away 

 must be photographed with an increased base if any stereoscopic effect is to be 

 obtained. The photographer need not expect any pronounced binocular effect of 

 depth on objects over 1000 ft. unless he uses an increased base. 



