Re: SL3D per John B

[P3D Larry Berlin <lberlin@xxxxxxxxx>]

>Date: Tue, 27 May 97
>From: P3D John Bercovitz writes:
>..........................  
>Of course wide open isn't very wide on a TV camera because focal 
>lengths are short because CCDs are small. ..........................
>
>I just thought of something.  You could use a lens with a large 
>aperture, say a lens for a Hassie, and then demagnify its image down
>onto the CCD.  That would get you back some modest amount of stereobase.
>It would also exacerbate the problem of fuzzy images.  Am I kidding
>myself or would this work?
>

Good thinking! (and by extension) How about some even larger lens for a
greater aperture size. Then using the fact that CCD's are small, you could
arrange two side by side optical reducing systems behind the primary
*single* lens. You're back to having to combine two signals, but that's part
and parcel to the process of observing a stereo image. Since you still need
two images it may be better to use the smallness of CCD's to create
*completely separate* image paths from the start. Less weight due to small
lenses and greater flexibility for lens placement (Just keep the horizontal
matched). Perhaps multiple sets thereof... choose your own stereo base or
perspective point while you watch or as you mix a recorded output. Or feed
Head Tracking Displays with appropriate pairs within some available range.
MPDA (Multiple Point Data Acquisition) could provide some measure of a sense
of looking behind  or around something by small movements of the head (all
that's needed to be convincing). How about multi-channel synchronized
broadcasts (MCSB)? Possible with digital broadcast methods?  

I conclude that two is better than one, and three or more is better than two
(fills in the gaps), never mind the associated complexities... that's what
computers are for... ;-)

Larry Berlin

Email: lberlin@xxxxxxxxx
http://www.sonic.net/~lberlin/
http://3dzine.simplenet.com/


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