Hi,
I’m working with a fragment shader that has run fine on ATI/NVidia on Windows and with NVidia on Mac OS X, but recently discovered that with ATI on Mac OS X (10.5), the gl_FrontFacing variable is always true. I have reported this to Apple, but received no comments. Has anyone else run into this?
(I have also discovered that attempting to use edge flags with glDrawArrays(GL_TRIANGLE,… ) ATI / Mac OS X is futile…)
Cheers
I remember this being problematic on the X1600.
Have you tried using GL_VERTEX_PROGRAM_TWO_SIDE_ARB and writing to gl_FrontFacing in the vertex shader? That’s a possible work around.
Other than that file a radar with Apple, or mention it on the Apple OpenGL List as Chris and others from ATI tend to lurk there and will reply. 
Writing to gl_FrontFacing in the vertex shader? How would I know what to write to it?
I did check out GL_VERTEX_PROGRAM_TWO_SIDE and I figured that I could use the color to code what is front / back as a workaround.
I’ll drop a message on the OpenGL list…
Noticed your post on the list… Hopefully Chris will respond to you.
I am guessing you’ve found this and read it…
http://lists.apple.com/archives/mac-opengl/2009/Mar/msg00001.html
That was kind of what I was referring to in the post above. I meant of course to use the colours but my fingers got ahead of my brain!!. 
Thanks for the advice to post on mac-openl scratt! ATI are helping me out now.
I successfully implemented a workaround with secondary back and front colors for gl_FrontFacing emulation that works great. The edgeflags thing was a confirmed bug.
No worries. For all their little niggles with drivers ATI seem far more responsive and genuinely concerned about issues than any other vendor. Glad you got some help.
This topic was automatically closed 183 days after the last reply. New replies are no longer allowed.