Archives Discussions

[jonmarbach]

Didn't work in 9.8 either

I'm using a geometry program to simultaneously render triangles to two layers of a texture array (similar to render-to-cube-map). The second layer has incorrect depth testing artifacts, as if using the depth test from layer 1, despite the fact that a two-layered depth texture is attached to the framebuffer as the depth buffer.

Another thing that seems to be going wrong is that calling glClear with the layered framebuffer bound is only clearing the first layer of the texture array. I would assume that glClear would clear the whole framebuffer, but maybe I'm wrong (and/or Nvidia's driver is not following the spec.)

Has anyone else done this successfully on ATI hardware yet?

Thanks,
Jon

[jonmarbach]

...and on 9.10.

Further testing confirms my suspicion that the first layer of the layered (3D texture) depth buffer is the only one being used.

I've sent a code sample to an AMD employee I've met recently, so we'll see if there's an official word.

[Byron]

Where can we dload a sample of this so that we may look at it.

9.11 is out, is the problem still there?

[jonmarbach]

Just had a chance to check into this again now that 10.3 is out. Still not working properly - I sent a sample to Brad G at AMD/ATI a while back but will have to ping him again.