[photo-3d] X-ray stereos

[Peter Abrahams <telscope@xxxxxxxxxx>]

>But the funny thing about stereo-x-rays is that both work with
>parallel as well as with cross-eyed viewing.

Not funny if you're trying to mount them.  When I had the rare opportunity
to shoot the shell xrays (using a radiology lab at a college, & borrowed
shells), I was very inexperienced at mounting; and also I didn't mark the
original film L & R.  I had a few days to mount a series of slides for the
NSA meeting in Bellevue, and had a terrible time determining the correct
mounting.  
You might think that either way - correct or reversed - would do, but the
orthoscopic view looked better & was easier to view, than the pseudo view.
This might be because the foreground was slightly more magnified than the
background; also we used a very large disparity between L & R (after all,
some shells are almost spherical).  But I believe there was one other
reason why there was a subtle difference: almost all shells corkscrew in
the same direction, like a 'dextral' corkscrew -- not like a 'sinister'
corkscrew.  If I remember correctly, pseudo mounting causes the spiral to
reverse -- but sometimes even that was hard to tell.

One other problem with seashells: because they spiral, the top & bottom
tips of the L & R images are offset vertically, so determining the correct
vertical placement is tough.

And one other problem: in the radiology lab, we would set the film down,
set the shell on the film, and shoot the L image.  Then lift the shell &
set new film down, then lower the shell.  We were careful not to rotate the
shell, but we didn't use a jig -- and should have; because there was slight
rotation in some images.  This absolutely drove me crazy in mounting -- a
little rotation in a rounded seashell is mighty tough to eliminate.  Again,
you'd think you could leave it imprecisely mounted, but it just didn't look
right until everything was correct.

It was a real trial by fire, actually induction & introduction by fire; but
one thing for sure; mounting has been shorn of its mystery & terror.  I use
RBTs and take my time; but I'm not hesitant to monkey with the chips.

Also, the image of the rose that we're discussing, where L & R can be
reversed, is a rather chaotic image.  The very simple flowers like
daffodils might be a problem if reversed, especially if the camera was very
close to them.

There is a stereo book of rocks & minerals, with crystallographic models
made of wire & cellophane.  If you take a closeup stereo photo of a clear
cube with an X on the front face, then another stereo photo with the cube
reversed, and compare a pseudo of the one with an ortho of the other --
will they look the same?  They might, but I think there will be conflicting
cues between the two.

Using an x-ray gun, you either rotate the object; or slide & tip the gun.
Generally, with transparent objects, you need less disparity between the
images, because you're seeing the front & back surfaces of the object.  You
can tip the gun so that it rotates about a point at the middle of the
object, or rotates about the front or back surfaces.  You could use a
simple lateral motion of the gun to make L & R images with less disparity
than tipping the gun; with the seashells we needed more distinctions
between the L & R.  With transparent objects, you can place the window
mid-object, or front or back.

>know how the real flower looked there is no way to tell
>which pair is pseudoscopic and which orthoscopic, because 
>the x-rays pictures are totally transparant.

If you foolishly don't mark your film L & R; there are a lot of variables:
the object could be facing towards or away from you; the chips could be
reversed L-R; the chips could be reversed front to back (I was working with
35mm copies of 8x10 xrays); and the operator could be confused.  It is all
a very interesting pastime, but I was in a panic as the NSA meeting
approached to some dozens of hours away.  If I hadn't had the help of John
Dennis of Stereo World (who converted them to 35mm) & members of the
Cascade Stereo Group, I'd have been sunk.  I also wrote some messages to
P3D, which caused George T. to respond, 'you've never been to one of these
meetings have you?' -- I didn't know what he meant until I got there & saw
that standards of mounting were not uniformly high.
Anyway, the presentation was a great success.
  
I hope to figure out a way to display the 8x10 originals at the NSA meeting
in Portland in 2004.  I have enough frosted mylar, and a few large viewers,
but I need a window to mount them on, or a sufficient quantity of
background lights.

I might not have another opportunity to do more xrays, but I have a
collection of transparent mineral crystals to photograph, and various
prisms & lenses.  I just shot a roll at an ice sculpture show; I'd never
seen absolutely clear ice -- it was beautiful & I hope the stereos work,
unfortunately there were no plain backgrounds to be had.  Also cut crystal
vases, art glass, maybe ghosts if I can find some.
--Peter
_______________________________________
Peter Abrahams   telscope@xxxxxxxxxx   The history of the telescope & 
   the binocular:   http://www.europa.com/~telscope/binotele.htm


 

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