[photo-3d] Re: focus and convergence

[Abram Klooswyk <abram.klooswyk@xxxxxx>]

In answer to ron labbe October 13, 2000:
> No matter how perfect a stereo projection is, the audience 
> members must disengage their normally linked focus and 
> convergence mechanisms,
John A. Rupkalvis wrote 13 Oct 2000: 
>Focus and convergence functions of the human eyes are 
>not normally linked, although they most often are used 
>in a manner that might suggest that.

When Masters speak I fear to rush in, but I believe the issue 
is a little more complicated.

In France and Germany, and later elsewhere, has been said 
that the depth zone in stereoviewing (I prefer the concept of 
stereo viewing space) should be limited because convergence 
and accommodation are linked. For many years in the first 
half of this century (this still is the 20th, for another 11 
weeks) many stereoscopists believed that convergence and 
accommodation were linked as toothed gears. Therefore, in 
their opinion, the viewing depth range should be limited by 
the depth of focus of the human eye, with they measured at 
about 70 min of arch (expressed as convergence difference), 
and from there the 1 in 50 rule emerged (indeed one-in-fifty, 
not one-in-thirty, which holds for the 5P system).

[George Themelis wrote Oct 14, 2000,  "... still has to 
decouple accommodation (focus) and convergence (...) 
convergence changes while the accommodation is fixed. This 
is true anytime we look at stereo slides through a viewer." 
This is in a strict sense not quite true, because
accommodation 
can change a little while keeping clear images, because of 
the eye's depth of focus, which is rather comparable to 
lenses in general, and also depending on pupil size]

About the accommodation-convergence link, the funny thing 
was that the Dutch ophthalmologist Donders already had shown 
in the 19th century how loose the coupling in fact was. 
One of his pupils wrote a thesis on the subject, which most 
people would find hard to read, not because it was in Dutch, 
but because it was in Latin... 
In his famous Treatise on Physiological Optics, Hermann [von] 
Helmholtz has quoted Donders' results. 

In recent years much work has been done on accommodative 
convergence and the reverse, convergence accommodation. 
There is a "zone of clear single binocular vision", which
means 
that to each value of accommodation belongs a range of 
possible vergence values, and the reverse is also true. 

However, the range is rather broad. For practical purposes in 
stereoscopy this means that the link can be almost ignored. 
Supposing a fixed accommodation value in using a stereoscope, 
vergence can be changed over several degrees. And then, as 
said, in addition a slight accommodation alteration is
possible 
because of the eye's depth of focus.

For viewing this means that the possibly range in practice is 
mostly more than can be achieved by the stereocamera, 
because of its limited depth of focus.

There is another point of view to look at all this.
The function of accommodation is to provide sharp retinal 
images, and the function of convergence to avoid double 
images, that is to provide single vision. 

It should not be surprising that both functions prefer to 
continue to do what they are invented for in the first place: 
provide sharp respectively single images, and that they 
easily give up their coupling when necessary. 
And it is necessary in stereoviewing.

But stereoviewing is not the only example of uncoupling, 
another is when myopic people take of their glasses. I can 
assure you that zero accommodation and convergence at 
25 cm (10 inch) does not give eyestrain, I have done that 
most of my life when I took off my glasses (I recommend 
myopia to freeviewing stereoscopists...).

Abram Klooswyk