> osgFX has this same problem, you don't actually know what the > graphics context supports until you've got a valid graphics > context to test against - which means the draw traversal, > which is after the cull traversal where you need to make the > descision. The way osgFX deals with this is to have one > frame which the test is done, and the next frame the result > can be used.
I was afraid of that. This is one of OpenGL's drawbacks, in my opinion. I understand why it is the way it is, but that doesn't make me stop wishing there was a better way to query capabilities from the underlying graphics system without having to have a current context. > The other way is to have the LightPointSpriteDrawable test > for the extension in its drawImplementation and if it isn't > supported do a graceful fallback locally. This is what I'd do. Ah! Good call. -Paul _______________________________________________ osg-users mailing list [email protected] http://openscenegraph.net/mailman/listinfo/osg-users http://www.openscenegraph.org/
