Scott Herod (herod++at++aspen.evt.com)
Mon, 20 Sep 1999 10:54:11 -0600
I've seen something like this twice before but only on the
OCTANE. I am pretty sure that in at least the first case, it
is an OCTANE only problem so if you see it on O2 it may not have
the same cause.
The first problem is that on OCTANE the pointer to texture
memory locations can become confused if you create and pfDelete
many textures. To get around this problem make sure that you
not only pfDelete textures when you are through with them but also
remove their GLHandles. Before calling pfDelete, use something like,
pfList *tex_list = our_pfuMakeTexList( leaf_node );
int num_texs = tex_list->getNum();
for ( int i=0; i<num_texs; i++ ) {
pfTexture *tex = (pfTexture *)tex_list->get(i);
if ( tex != 0 ) {
tex->deleteGLHandle(); // For Performer 2.2
}
} // for each texture in this pfNode
pfDelete(tex_list);
( Notice that I've also used a different call to get the texture
list, our_pfuMakeTexList, the one that is part of Performer leaks. )
Also, you have got to have a valid GLContext when you pass through
this code.
Recently I've seen something similar occurring again only on OCTANE
that I think has the same root cause. It only happens when I have
the PFPATH environment variable set though so I haven't got a work
around yet other than not setting PFPATH.
Scott Herod
scott++at++evt.com
Bryan Walsh wrote:
>
> Hi All,
>
> In my lab we use Viewpoint Datalab's realtime library of OpenFlight
> models. All the models look great on our Onyx and Onyx2. However,
> on our Octanes and O2s the textures of certain models (f-15s,
> eurofighters...) seem to be misapplied. It looks as if the texture
> coordinates are screwed up. I've reported this problem to SGI and they
> have been able to reproduce it, but haven't been able to give me a cause
> or a solution. If anybody has seen this problem and/or can offer a
> possible solution, it would be greatly appreciated.
>
> Thank you much,
>
> Bryan Walsh
This archive was generated by hypermail 2.0b2 on Mon Sep 20 1999 - 09:54:19 PDT