[MF3D.FORUM:1677] Re: Using front shift to simulate camera movements?

[Tom Hubin <thubin@xxxxxxxxx>]

Don Lopp wrote:
> 
> I think a better solution is found on page 113 ofthe book The World of 3-Dby
> Jac G. Ferwerda 1990 printed in the Netherlands in English- it requires 2
> cameras and a semi translucent ( half silvered) mirror I assumethe intended
> use is for macro stereo- I am up at this time of the morniing because  the
> pain is such that I can not sleep- I successfully  detoxed  from 4 years of
> morphine  but have not found a good replacement as of yet for my 3 back
> surgeries   so my thinking is still a bit muddled sorry about that  -DON.

Hello Don,

I don't have the book so I will just throw out some thoughts on using a
plate beam splitter to feed the light to 2 cameras.

On the down side you loose half the light as compared to the two mirror
symmetrical arrangement discussed earlier in this thread. In low light
conditions this might be a problem.

On the up side you can set any interocular distance desired. Positive,
zero, negative. Zero if you just want duplicates without stereo.
Negative if you want to swap the roles of the right and left cameras.
Although I cannot think of a good reason for doing this. 

Any positive distance allows for the lenses to appear to overlap. That
could be very useful for closeups where you want the interocular
seperation to be less than the lens diameter. Can you think of a
situation where this might apply?

When a single beam splitter is used the reflected image is mirrored
while the transmitted image is normal. Assuming you want both images
normal you would need a mirror between the splitter reflected light and
the camera lens, just to mirror the mirror image to make it normal.

Tom Hubin
thubin@xxxxxxxxx