P3D Re Projection and mounting, terminology

[abram klooswyk <abram.klooswyk@xxxxxx>]

David W. Kesner wrote on deviation and convergent points in 
P3d 3749:
>(...)Several people have alluded to the difference in on 
>film deviation and on screen deviation ...
>(...)As I understand it, zero deviation occurs at the window 
>(...) Say you have an image with convergent points 
>(..) I know I should study the technical papers on the 
>subject, but I find it much easier to ask the questions here 
>and get an immediate and pointed answer. 


In my opinion studying a stereoglossary before studying the 
technical papers would be better :-) The only problem is that 
no real authoritative glossary exists :-(.

1. Separation 
I would recommend the term "separation" to indicate _all_ left 
to right distances of "items" in stereoscopy. This includes 
lens separation of stereocameras, camera separation in 
hyperstereo, separation of homologues on stereoslides and 
printed stereoviews, (also infinity-, far- and nearpoint 
separation), separation of image apertures in stereocameras, 
and of mask apertures of 41 x 101 masks, separation of viewer 
lenses and eventually projection lenses.

It is a good thing to have different terms for X-direction 
distances ("separation") and Z-direction distances ("distance" 
in a narrower sense). So we have near point separation and 
near point distance, far point separation and far point 
distance. This useful distinction between terms existed 
earlier in German and French, in French "écartement" and 
"distance", in German "Abstand" and "Weite". 
In Dutch there used to be no specific term for separation, so 
Ferwerda was happy to adopt it from English, in Dutch it 
became "separatie". (Ferwerda found the term in Dalgoutte's 
1967 glossary). 
A lot of confusion is avoided by being able to use these 
different terms for different concepts.  

2. Deviation
By strict definition this should be the difference between
on slide infinity separation and any smaller on slide 
separation. 
In this way we get near point deviation and window deviation, 
the latter being the difference between infinity separation 
and mask aperture separation. Less strictly applied deviation 
could also mean the difference between far point separation 
and any smaller on slide separation, as in many scenes no 
infinity is included. In projection, on screen deviation 
should also refer to the difference of the separations of the 
same points as on the slide. (Then of course the same 
magnification applies to all deviations).

3. Convergence 
In stereoscopy the term convergence applies to lines or 
directions, like lines of sight or optical axes. These lines 
or directions converge if they meet at a common point closer 
than infinity. 


>(...) zero deviation occurs at the window (...)
According to the definition this has zero meaning :-)

>Say you have an image with convergent points in the center of 
>the scene ... 
Speaking of convergent points is pointless (:-) without 
indicating which lines or directions converge at this point.

>It has been suggested that by changing the horizontal 
>convergence of the projector (and thereby moving the 
>placement of the window from the screen to somewhere in 
>front of it) you will be able to compensate ...

"Horizontal projector adjustment" seems a better way to say it 
(for "convergence" see above :-)). By this adjustment 
obviously all on screen separations change by the _same_ 
amount. When it is done to diminish too large on screen 
infinity separations, the percepted stereo window will 
necessarily come in front of the screen. This is easy to 
visualize when you imagine a birds-eye view of a spectator 
looking at the screen.

"Negative deviation" and "positive deviation" are strictly not 
part of the definition of deviation. (However, in some cases 
it is useful to speak of the "direction of deviation").

In viewing stereoprojection, we have to do with crossed and 
uncrossed fusion (the terms often used by physiologists). 
Normally, the left infinity homologue is to the left of the 
right hand infinity homologue, they are seen with uncrossed 
fusion. Any two homologues of which the left one is to the 
right of the right hand one will be seen with crossed fusion, 
and therefore in front of the screen (as the stereo window in 
the discussed case).

In viewing anaglyphs, and indeed in viewing slides in 
stereoscopes, similar effects can occur, but the projection 
case is easier to visualize. (In stereoscope viewing 
everything is in virtual image space.)

Abram Klooswyk