Hi Robert,
I ran the tests on my PC at home, and got some pretty weird results for
the last one. See below. I think we're seeing that video card drivers
are very inconsistent across versions (which we knew from experience,
but now we can quantify it).
AMD Athlon 64 X2 4200+, 2GB RAM, Vista SP1, NVidia GeForce 7900GTX,
Driver 174.74
_________________________________________________
1) osgmemorytest --window -c 1000
At 695 I started getting
Windows Error #3221684230: [Screen #0]
GraphicsWindowWin32::makeCurrentImplementation() - Unable to set current
OpenGL rendering context
and at 697 my graphics driver crashed and I had to kill osgmemorytest
manually (it seemed to have hung). My main memory was filled up too, and
it was swapping like crazy.
_________________________________________________
2) osgmemorytest --pbuffer -c 1000
GraphicsContext 547
PixelBufferWin32::init, wglCreatePbufferARB error: The operation
completed successfully.
Error: Unable to create graphis context, problem with running
osgViewer-2.7.6, cannot create windows/pbuffers.
Exception caught, contexts completed = 547, gl objects successfully
applied = 0, error = Failed to create PixelBuffer
_________________________________________________
3) osgmemorytest --texture 256 256 -g 10000
Exception caught, contexts completed = 0, gl objects successfully
applied = 0
(same error we've seen before on Windows for this test)
_________________________________________________
4) osgmemorytest --geometry 32 32 -g 10000
Successful completion, contexts created = 1, gl objects applied = 10000
Duration = 8.162292 seconds.
_________________________________________________
5) osgmemorytest --geometry-vbo 32 32 -g 10000
Successful completion, contexts created = 1, gl objects applied = 10000
Duration = 2.741584 seconds.
_________________________________________________
6) osgmemorytest --geometry-va 32 32 -g 10000
Successful completion, contexts created = 1, gl objects applied = 10000
Duration = 2.411383 seconds.
_________________________________________________
7) osgmemorytest --fbo 256 256 -g 1000
Successful completion, contexts created = 1, gl objects applied = 1000
Duration = 12.410262 seconds.
_________________________________________________
8) osgmemorytest --fbo 1024 1024 -g 1000
At 513, there seemed to be a null pointer access, no OpenGL error
though. It never filled up my main memory (system memory usage was 1.5GB
when running, 900MB when idle)
_________________________________________________
9) osgmemorytest --fbo 2048 2048 -g 1000
Very weird: it went up to ~100 FBOs pretty slowly, like one per second,
then it blurted out about 400 more in less than 5 seconds and exited
without any error message. So the total was 484, but I'm not sure that's
correct given the behavior.
_________________________________________________
J-S
--
______________________________________________________
Jean-Sebastien Guay [EMAIL PROTECTED]
http://www.cm-labs.com/
http://whitestar02.webhop.org/
_______________________________________________
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
