Re: [photo-3d] Luring Rupkalvis

["John A. Rupkalvis" <stereoscope@xxxxxxxxxxxxx>]

Bruce,

An addendum to my previous reply.  I got onto a tangent that I felt was
important, and now realize that I didn't really answer your specific
question.   Although, I am not sure there is a specific answer that applies
to all cases.

Imax photography must consider Imax projection, and Imax audiences.  The
projectors are aligned such that the screen image has a positive parallax
offset of 65mm at the screen.  That is, if the optical axes of the camera
lenses were parallel, photographic infinity would have a positive parallax
of 65mm.   In "normal" situations, this would allow for a very comfortable
parallel (non-convergent, non-divergent) eyeline to the farthest parts of
the image.   This would prevent the infamous "walleye" condition and the
resultant image-splitting or double images in far parts of the screen.  Zero
degrees is well within the 2 degree limit you mentioned.

Unfortunately, this "recommendation" has not always been followed (see my
previous message).

Also, the Imax 3D system has been designed such that, when the photography
is correct (a not always valid assumption), the viewer in the center seat
(in all directions) of the typical Imax auditorium will experience an
orthostereo condition.  That is, all dimensions of any object will appear to
be proportionately correct.  An image of a cube will appear to have equally
dimensioned faces, as originally photographed.

In practice, it is possible to deviate from some of the recommended
specifications sometimes, depending upon the subject matter.  An example
would be an actor close to the camera(s) with only a blue sky background.

In this instance, convergence could be set closely behind the actor, which
would put the sky background way beyond 2 degrees (beyond infinity!).   But,
who could tell?  There are no definable points in the background to split or
give headaches.

Therefore, we must consider the characteristics of the subject matter before
deciding on things such as a "2 degree" limit.

I have bookshelves full of books on stereoscopy by a wide variety of
authors, spanning many, many years.

Most of these contain "tables" for figuring parallaxes.   Every one of these
tables is different.

Which one is correct?

They all are!   Given specific circumstances, it is possible to find (or
create) situations that will make any set of numbers work.

In the case of interior photography (or exteriors with a limited depth
range) infinity might no longer be a consideration, but rather the farthest
definable point in that particular scene.

Yes, your comment regarding "limiting depth" is very important.  > Do the
folks who shoot 3D movies worry about "limiting" depth in the scene? <
Probably not usually.  But they should.

The most important thing to remember in stereoscopic imaging, is that
stereographic images are extremely subjective.

The decision of where to set convergence must be made depending upon all of
the characteristics of the subject matter to be imaged.   A very important
such characteristic is the existing depth range (of the actual scene, or
"set").  What we do with that range, how and where we determine our limits,
are extremely important to comfortable viewing.  [:-}

JR

----- Original Message -----
From: "Bruce Springsteen" <bsspringsteen@xxxxxxxxx>
To: "Photo-3D Maillist" <photo-3d@xxxxxxxxxxx>
Sent: Monday, October 09, 2000 6:29 PM
Subject: [photo-3d] Luring Rupkalvis


> John Rupkalvis said:
> >Picky?  Perhaps.  But, only when everyone can agree on the usage of the
> >various stereoscopic terms can we discuss these things with each other
> >without misunderstandings.
>
> In that case, welcome to Photo3D, John Rupkalvis! :-)
>
> We've had some terrific one-of-a-kind additions to the list this last year
> - Sheldon of the USA comes to mind, for example - and now Rupkalvis, who
> with long-time P3Ders Zone and Alderson form an Eternal Golden Braid in my
> mental pantheon of California stereo mavens.  I first saw those three
> names joined in my delightful 3D comics from the 80s.  Now John proves in
> a few postings what Ray said of him, that he's one of the unique resources
> in our 3D community.
>
> Having greased the skids, John, I have another question about Imax and
> Iwerks 3D, and 3D movies in general.  What kind of maximum deviations are
> tolerated by the best practitioners in the format - in other words what is
> the total depth generally considered permissable in commercial motion
> pictures?  An approximate total convergence range, from near point to
> infinity, of about 2 degrees has been put forward as the "limit" in
> still-3D projection, but I swear that many scenes in "Mark Twain's
> America" were well beyond that, without unreasonable strain.  Do the folks
> who shoot 3D movies worry about "limiting" depth in the scene?  Have I
> asked this querstion understandably? :-)
>
> Springsteen, 66044
>
> __________________________________________________
> Do You Yahoo!?
> Get Yahoo! Mail - Free email you can access from anywhere!
> http://mail.yahoo.com/
>
>
>
>