more difficult to use than just rendering two
separate images. To see what I mean, try
rendering the standard benchmark scene
"skyvase". When I tried I could not get
a good stereo pair using Stereo POV, but
it was easy to make two separate left/right
images using my usual method.

To get a good stereo pair from "skyvase.pov"
I only need to use the 1:30 rule (one unit of
stereo base separation for every 30 units of
distance to the main subject of the image).
This 1:30 rule results in two renderings,
with the x dimension of location and look_at
set at 3.33 (right) and -3.33 (left). This makes
a stereo pair with just a little too much stereo
separation because it doesn't take into account
that skyvase.pov has a direction z of 2, which
would be the equivalent of a lens with a long
focal length.

Hermann tried and also had difficulty using
Stereo POV with the skyvase.pov scene,
but with some effort he found settings that
worked. To me it seemed much easier to do
it the old way, but I'll give Stereo POV a few
more tries. Hermann suggested that the problem
with skyvase.pov is that the original author
chose the camera location by trial and error.
I think Stereo POV should work as easily as
my old way, no matter what units are used for
the camera. I imagine a stereo patch to POV
that only requires the user to calculate (or guess)
the stereo base separation and provide a
location for the stereo window.

Give Stereo POV a try with the skyvase scene.

Harolddd

Vic wrote:


> The metric is offten confusing to me, but your problem is not clear to me.
> In Stereo Pov these properties can be changed by setting stereo_base and
the
> right and up vectors according to it's documentation. What are the
> difficulties with your scenes?

Post a reply to this message