Re:Suitability of telescope oculars for stereoscopes

[T3D john bercovitz <bercov@xxxxxxxxxxx>]

Sam Smith writes:
 
> Has anyone ever done any experiments with photo-sensitizing 
> concave surfaces, such as the inside of half a ping-pong ball?
 
Been done but it's difficult.  It's very desirable because lenses 
naturally have curved fields so making the film plane fit the field
rather than vice versa gives the lens designer more freedom.  The
practicality is the problem, as you say.  It's got to be worth a lot 
to you before you'd go to this much trouble.  Sometimes a camera
maker will go half way to a spherical surface by making a cylindrical
surface to fit a simple curved-field lens.  Examples are my first
camera (a Brownie) and Kodak's Stretch camera.  To answer the exact
question you put, I can't think of an advantage from a perception
standpoint to having a curved surface.  You might think there wouldn't
be any lateral shift of the pupil of your eye relative to the film 
as you scan.  But a good ocular will put the scene far enough away to
make this a non-problem.
 
The human eye is just another transducer of course.  It's a small, 
stopped-down system with so so resolution off axis.  It's probably
spherical because that's easier to fabricate biologically, and keep
inflated, than a cubic eyeball.
 
A photographic lens is in fact a device for mapping object points' 
angular displacements off the optic axis onto a flat plate.  That's 
why it's so important to view from the right place.  (This isn't to 
take anything away from artistic license which allows using a 17 mm 
lens to take a portrait just for the effect.)
 
I think you could make a camera with a curved field both on the object 
side and on the image side by using a glass sphere for a lens.  One 
problem is the iris.  In which direction should its axis point?  Maybe 
you could obviate this one with Larry B's suggestion of a CCD array: 
To each of the pixels you attach a small black tube (made by micro-
channel plate technology?) and that tube only allows the CCD pixel to 
see the center of the ball from its (the pixel's) point of view.  The 
other big problem would be focussing.  If you don't build a really 
stopped down, fixed-focus system, you'll need to grow and shrink the 
sphere the pixels are attached to.  Rubberized inflatable silicon chip 
balloon?  8-)
 
John B
 
PS: Did you notice in my diagram that there is a singular case where
change of magnification does not distort the perspective?  The original 
poster hints at it.  The case is flat-to-flat: a planar object to a plane
image.  So if you take a picture of a flat surface parallel to the film 
plane with, say, circles on it, you will get a photo with nice round 
circles on it and you can move closer to it or farther from it with no 
distortion.  But let one of those circles be out of that plane, whether 
tipped to that plane or in another plane at a different distance from the 
lens, and all bets are off.  So there is such a thing as free lunch but 
it's pretty limited fare.


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

End of TECH-3D Digest 161
*************************
*************************
 Trouble? Send e-mail to 
 wier@xxxxxxxxxxxxxxx 
 To unsubscribe select one of the following,
 place it in the BODY of a message and send it to:
 listserv@xxxxxxxxxxxxxxx 
   unsubscribe photo-3d
   unsubscribe sell-3d
   unsubscribe mc68hc11
   unsubscribe overland-trails
   unsubscribe icom
 *************************