P3D Re: boris starosta's site

[bercov@xxxxxxxxxxx (John Bercovitz)]

Dr.T writes:

> My advice to Boris is to go beyond "ortho" and explore ALL faces
> of the stereo experience.

I agree in the sense that hypos and hypers can be fun sometimes.
In fact, it is difficult to do macros with much depth in them unless 
you do them as a hypos.  Not to mention the potential vergence problems 
when you project a correctly-windowed ortho macro.  Also, it is difficult 
to see the depth in a really large scene without doing a hyper.  This 
condition is exacerbated by lower-than-ideal resolution in 35 mm format 
and very low resolution on the computer screen.

> It is interesting that the inexpensive Pinsharp viewer is dictating 
> what lenses he uses in his stereo cameras.  I would go the other way
> around:  Use what lenses I have available and then worry how to view
> the images.  Boris acknowledges that most people will not view his
> computer images in a way that satisfies the ortho conditions.  The
> same is true for projected images.  Then, why over-worry about ortho?

It is easier to get the viewer lenses and then match the camera to them 
because good viewer lenses are hard to find but camera lenses are made 
in virtually all focal lengths (notable exception: 70 mm).  So I think
Boris has taken the right approach here.  He could have gone with the
Red Button and Minolta 45 mm lenses or he could have gone with a more
expensive (than Pinsharp) 50 mm viewer and the more common 50 mm lenses.

Here George is talking about a different ortho error than stereobase
mismatch: Here he is talking about perspective mismatch which changes
only the depth dimension and does not rescale all three dimensions 
equally like stereobase mismatch does.  I think perspective mismatch
is a problem only for people who are new to stereo and thus not inured 
to this distortion and for people who have poor depth perception.  Our
projectionist at OCC has been doing stereo since the dawn of time, and 
has good depth perception, but a perspective mismatch of 4X doesn't
seem to faze him.  I would guess George is in this class by now.
I'm not.  Call me a sensitive guy.  8-)

> Let's see if I get this right... Slide film has limited resolution 
> (compared to what?  Our eyes?)  And so does the computer monitor.
> So, I only shoot objects within 2-3 feet from the camera, or 10-20
> feet the most.  I do this so that partially stereo blind people
> can enjoy my images.  So I restrict my photography to what will
> appear good to partially stereo blind people who are viewing my 
> images in a computer monitor.  This comes from a person who is only 
> using 50 mm lenses in his cameras because they match HIS 50 mm
> viewer lenses.  Am I the only one who sees a problem with this?

Probably.  Maybe?  8-)  I don't know.  What problem do you see here,
George?  I would say that Boris doesn't do this solely so that partially 
stereo blind people can enjoy his images.  Pumping the disparity is 
good for computer stereo.  You can do that with hypers or closeups.

> I use slide film and take many pictures where there is nothing in
> the first 10-20 feet (yes Ron, I do!)  I like those pictures and 
> they look great in my viewer.  There is plenty of depth 20 feet 
> away from the camera.  These pictures would not have worked well 
> in a computer monitor.  I would not use them for this purpose but
> would share them with others through a viewer and even in 
> projection.

"I would not use them for this purpose" and neither would Boris.

>> Architecture
>> I consider stereo photography to be of borderline utility in 
>> architecture. Most buildings are too large and must be placed
>> too far from the camera for stereo cues to make much of a 
>> contribution to spatial perception.

> Some of my best images are of buildings.  Stereo photography has been
> and could be a great tool for teaching architecture.  (I wonder if
> our resident professor of Architecture agrees with that...)
> We have all seen the models that architects make to show as how
> a future site will look like.  Hyperstereos will give you the same
> effect.  So, what is wrong with that?

I agree that hyper is the way to go accompanied by a caveat/explanation 
for the newcomers that this is a hyper so they won't wonder why the 
building looks like a model instead fo a full-sized building.  If you 
can afford the luxury of medium format, a normal stereobase is great 
for architecture as long as the building is of reasonable size (not the
Taj Mahal).  I would recommend a medium format camera with name 80 mm
lenses and a viewer which uses Supplier: Edmund Scientific's Catalog 
Number 45105.  And no, I'm not a shill for Edmund; I think their prices
are too high.

John B


------------------------------