Robert, The two cards are in two separate machines, on which I am getting a similar error. However, the worse by far is the machine with the 8800 GTX, so that is the one on which I have focussed my testing efforts.
It seems that the driver does not freeze permanently; sometimes I can regain control by holding down escape (in osgviewer) which eventually quits the program returns apparently normal control. This definitely leaves the OpenGL driver in a corrupted state though, because from the first time osg breaks, all other OpenGL programs show similar geometry glitches and freezes. This persists until reset. I want to stress however that this never happens if I do not run an osg-based application. I agree that it definitely *looks* like a driver issue, but the facts that I am using unexceptional hardware, the latest nvidia driver (162.18) that I presume most windows developers here are also using, and that no other OpenGL or Direct3D applications show any problems makes me scratch my head a bit. Is there anyone else here/do you know of anyone using osg under WinXP with an 8-series nvidia card? I should add that the system dual boots with Ubuntu, under which I get no such error. This steers me away from the conclusion that this is a hardware fault. All in all I'm pretty stumped. Chris. On 7/31/07, Robert Osfield <[EMAIL PROTECTED]> wrote: > Hi Chris, > > This really sounds like a OpenGL driver issue. W.r.t other OpenGL > apps, are they opening up on both cards at one time like the OSG > examples will try to do later, it could be the driver is struggling > with handle two graphics cards at once. > > Things you could try is forcing the apps to open up on a particular > screen by setting the OSG_SCREEN env var to 0 or 1. > > Robert. > > > > On 7/31/07, Chris Davoren <[EMAIL PROTECTED]> wrote: > > Thanks for replying Robert. > > > > I am using the latest version of OSG: 2.0, released on the 15th of June. > > > > The two cards in question are a 6600GT and an 8800GTX, both of which are > > using the latest stable (not beta) drivers from Nvidia. My first thought > > was a driver issue too, except that I can't get any other OpenGL programs to > > fail. Is there a way I can identify this kind of issue? > > > > As far as I can tell, pretty much all of the examples can or will fail at > > some point. Specifically, the osgsimulator example runs very poorly on the > > 6600GT (low framerate, globe reflection flickers and is backed by a black > > rectangle) and bringing text up on the screen, such as by pressing H in > > osgsimulator, seems to encourage the error to happen more often. > > > > Thoughts? > > > > Chris Davoren. > > > > _______________________________________________ > > osg-users mailing list > > [email protected] > > http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org > > > > > _______________________________________________ > osg-users mailing list > [email protected] > http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org > _______________________________________________ osg-users mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
