[photo-3d] Re: Constant infinity mounting

[Abram Klooswyk <abram.klooswyk@xxxxxx>]

George Themelis Jul 23, 2000: 
> ...two schools... 
>Mount with a constant infinity separation 
(...) 
>...[this] school of thought claims that this is more 
>natural and comfortable to the eyes.

There are different professors in that school...
Some of them (which I like most) think that the
"natural and comfortable" argument is nonsense
because the slight difference in viewing convergence
is not sensible. BTW, a similar debate is on projection, 
the "normal infinity" school, the "window-on-the-screen" 
school, the "superimpose important feature" school and 
others, again different approaches but only significant 
different with large screens.

With your tool:
>automated mounting is not very difficult based 
>on this method.  

When Koo Ferwerda held his enquiry in 1973-1974 the aim
was to design a stereosystem with "automatic" mounting,
that is mounting without viewing the result before it
is finished, using a method which any novice can use.
With 35 mm film (135) this is mostly based on perforation
mounting, which requires reliable positioning of the 
perforations in the camera, also excluding tilted running
of the film in the camera (this is often not reliable 
in the old camera's).

A number of proposals were discussed at the technical 
meetings at the ISU founding congress, Wageningen 1975.
Constant infinity separation is more or less a 
prerequisite for "automatic" systems. (The "natural and 
comfortable" argument is not needed :-))
The "system" concept of course isn't only about
mounting, the viewing and projection also should be
automatic: no changing of lens separation, no horizontal
adjustments in projection. From natural object scene to
cyclopean image scene, all "automatic", the "you press 
the button..." philosophy. 

>If the scene does not include
>infinity (close ups where the distant background is
>blocked, macros, etc.) I have no clue how "constant
>infinity" is practiced and what is the reasoning
>behind it. 

The answer to the practicing is of course: automatic
is automatic. When you use your tool for scenes with
no far objects, the result remains the same, but 
there is only a "virtual" infinity separation.

Suppose you could put a mark on the film on the spot 
where the optical axes of both camera lenses intersect  
with the film, then this would mark the infinity 
separation in the camera, it would be seen on both chips, 
in mounting, and also on the screen in projection.
Then the infinity separation would always be directly
visible on the slides, regardless of the actual content 
of the scenes.
(In the past I have suggested to call this "axial 
infinity separation", but few others have used this.)

>what is the reasoning behind it. 
Again: automatic mounting and projection. 

For macro (not only for specially designed macro systems 
or camera's) the same concept holds, as for hypers, for
you can always imagine the marks of optical axes , also 
with two positions of a single camera. 

(Problems occur of course in using a normal stereocamera
for close-ups, close objects will protude through a 
window at normal distance.)

So automatic mounting is a strong argument for "constant 
infinity mounting", but, as you found out in using the tool 
(but knew it in advance, I suspect :-)) in some cases 
it doesn't work. All the better for stereoscopists which 
still can use their brains :-)

Abram Klooswyk