One thing I haven't tried yet is to mimic SLI at the viewer level by
creating two cameras and two contexts, one per card, then slit the up
final window between the two windows i.e. top half, bottom half then
manually composite the final result by copying the data from the
second window to the first one.

You could probably set up osgViewer to do this via several slave
cameras and custom camera post draw callback which does the copying
from the second card to the first one.  The copy would be real slow as
it'd be read back to memory and write back to the first card, but for
certain models it might be a useful gain such as volume rendering.
Ideally you'd want this copy done over the SLI bridge, but without
extensions to access it this won't be possible.

Robert.
_______________________________________________
osg-users mailing list
[email protected]
http://openscenegraph.net/mailman/listinfo/osg-users
http://www.openscenegraph.org/

Reply via email to