[visionegg] Re: GL state responsibility (warning: may contain theory)

Kevin J. MacKenzie wrote:
> The only possible drawback I see here is for stereo geometry.  When
> rendering stereoscopic stimuli, I employ 2 projection matrices which rely on
> differences in the camera position.  However, the geometry here is
> straightforward, and probably not difficult to change.

Should I check these changes into CVS (they are substantial enough that
I wanted to give people a period to comment first), you'd need to make
only minor modifications.  In fact, your code may work exactly the way
it is, it's just that now you'd know it's not The Right Way.

To do things properly, the idea is to move all of your camera position
and orientation information to the new "camera_matrix" parameter of
Viewport, which in turn is used as the base modelview matrix to which
other, stimulus-specific, transformations are applied.  (The field of
view and aspect ratio would continue to be part of the "projection"
parameter of Viewport.)

Cheers!
Andrew

======================================
The Vision Egg mailing list
Archives: http://www.freelists.org/archives/visionegg
Website: http://www.visionegg.org/mailinglist.html

Other related posts: