Mailinglist Archives:
Infrared
Panorama
Photo-3D
Tech-3D
Sell-3D
MF3D

Notice
This mailinglist archive is frozen since May 2001, i.e. it will stay online but will not be updated.
<-- Date Index --> <-- Thread Index --> [Author Index]

Re: [photo-3d] Re: Newbie question-near point distance


  • From: "Dr. George A. Themelis" <drt-3d@xxxxxxx>
  • Subject: Re: [photo-3d] Re: Newbie question-near point distance
  • Date: Fri, 27 Apr 2001 10:30:33 -0500

Bruce Springsteen writes:

> Both are recommendations to reduce base when increasing focal length, or
> add base when subtracting focal length.  This is just the opposite of the
> old PePax rule of increasing (or decreasing) base and focal length
> together.  Though based on a fallacy suggesting that parallax and
> perspective are reciprocal parts of orthostereo, PePax still served the
> useful purpose of counteracting unpleasant "stretch" and "squeeze"
> effects.  What purpose does the opposite advice serve?

Thank you Bruce, you are great!!!  I love how you are asking
questions to which you know the answers, pretending to be
a newbie, just to challenge us.

OK, here is my response:

First, ask yourself why is someone using 135mm lenses instead
of 35mm lenses?  Long lenses are used in specialized situations,
like: 1) portraits, 2) wild-life (even of the zoo variety), 3) action
(when the photographer cannot come closer to the action),
4) to isolate details in a building, monument, etc.

You will notice that none of these cases includes infinity.
The purpose of the long lens is to isolate a detail (face,
animal, action, etc.) when the photographer cannot
or does not want to come closer to the subject.
The depth range in these situations is very limited.

If you apply the full "maximum OFD" treatment, you will
have to plug near *and far* distances.  Because far
distances are limited, you will get more stereo base
than this formula calculates: "New base = Original 
base * (Original Focal Length / New Focal Length)"
because this formula assumes infinity.

It so happens that McKay's PePax rule (of increasing
base and focal length together) works very well in
most practical situations.  And the results are very
ortho-like, despite the fallacy of ortho-stereo as you
say.

What bothers me with the "maximum on-film-deviation"
approach is that it is "artificial".  There is no physical
reason why the on-film-deviation has to be constant
or maximum or even be a serious consideration in a
photographic situation.  On the other hand, McKay's
PePax rule has an explanation (as you said, serves the
"useful purpose of counteracting unpleasant 'stretch' 
and 'squeeze' effects).

The PePax rule really deals with increasing the focal
length.  What happens at the other extreme, i.e. very
wide lenses?  You will notice that in these situations
people tend to come very close to the near object
so usually they do not have to either increase or
decrease the stereo base.  For example, quite a few
people are experimenting with the RBT SLR type
cameras and super wide lenses with good results, 
while fitting 200mm lenses on an RBT camera will 
lead to disappointments.  

George Themelis


 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/