Stephan Huber wrote on Friday, September 08, 2006 12:06 PM: > Hi Bryan, > > Thrall, Bryan schrieb: >> I finally got around to looking at this, but I'm having some trouble >> creating a simple test case; I don't see how you can have two >> Geometry objects with different textures (as Stephen describes) that >> MergeGeometryVisitor will merge, because they will have different >> StateSets so the LessGeometry comparator will see them as different >> and they won't be put into the same DuplicateList. > > >> Stephen, can you create a simple test case that demonstrates the >> problem you're having? > > After looking into my code I described my problem a little bit > inaccurately. My setting is as follows: There is one geode part of the > scenegraph with two geometry-drawables, both without a stateset. Later > in the application, both geometries get a newly created stateset with > different textures and a blending-animation starts. > > I ran into the problem, that the two geometries got merged by the > Optimizer, and my code holds a reference to a geometry, which is not > part of the scenegraph anymore, so the animation did not work anymore. > > What I want to do is to "mark" the geode/geometry, so the Optimizer > can still do its work, but leaves this particular geode with its two > geometries untouched. I thought this is why Object::Datavariance > exists, but it seems, that every object in the scenegraph is DYNAMIC. > > So there is no bug in the optimizer, but I have a small feature > request to exclude certain nodes from being optimized.
Well, the bug is that you should be able to use Optimizer::setPermissibleOptimizationsForObject() but can't because MergeGeometryVisitor doesn't check that ;) I have submitted a fix for that. In the mean time, you can use setPermissibleOptimizationsForObject() to disable MERGE_GEOMETRY on the Geode, as Robert suggested. -- Bryan Thrall FlightSafety International [EMAIL PROTECTED] . _______________________________________________ osg-users mailing list [email protected] http://openscenegraph.net/mailman/listinfo/osg-users http://www.openscenegraph.org/
