Re: [osg-users] CullVisitor locks up on NaN matrix

> I'm not aware of any reason why 0.9.8 should be "senstive" to NaN's where
> 0.9.6 wasn't, it might simply by an issue that compiler has optimized the
> code in a slightly different way with the two code basis and the a subtle
> difference is enough to trip over in a CPU or compiler bug. Have you tried
> disabling optimization in the release build?

No. I could try that, but I'm not sure if I'll get around to it
anytime soon, now that it's not really bothering me anymore. (I've
got lots of other stuff to do too :-) ).

When you brought up compilers, you reminded me of one issue that I
didn't mention. The 0.9.6 version I had was compiled with gcc-3.2
on Red Hat Enterprise 3, while the 0.9.8 version was compiled with
gcc-3.4 on Fedora Core 3 (0.9.6 didn't compile on FC3 because of the
stingy new compiler). So, the problem may also lie with the
difference in compiler versions.

I'll see if I can look into it more when I get some time.

Thanks!

--"J"

_______________________________________________
osg-users mailing list
osg-users@openscenegraph.net
http://openscenegraph.net/mailman/listinfo/osg-users
http://www.openscenegraph.org/
Received on Wed Jan 26 12:07:31 2005