PDA

View Full Version : FBO / memory limitation?


sdiverdi
10-26-07, 07:30 PM
I'm experiencing some weird behavior in a program I'm developing, and I'm hoping someone here can clue me in. I have a GPGPU program that works on 9 floating point values per pixel stored across three RGBA32 textures, and it has to ping-pong between them, so there's 6 textures total. Each pass I attach 3 of them to an FBO and bind the other 3 as textures, activate my shader, and draw a quad. Rinse, repeat. This works fine with textures of 1024x1024, and is quite speedy on a GeForceFX 8800GTX (with 768Mb RAM, driver 163.71 on Windows XP). However, when I bump up to 2048x2048, it fails. Specifically, the first pass works fine, and then when the second pass gets to attaching textures to the FBO, the third attachment ends up causing a FRAMEBUFFER_UNSUPPORTED error.

I'm suspicious it's a shortage of memory somehow, but that doesn't make too much sense - 6 RGBA32 textures at 1024x1024 is roughly 100Mb, and at 2048x2048 should be 400Mb, which is only half of what I have available, so it really should be okay. Another thought I had was that maybe the problem was that the 2048x2048 texture was larger than my window (1024x1024), but when i run with a 1024x1024 texture in a 512x512 window, it's no problem.

Any thoughts on what's going on? Or how I might go about debugging this? Thanks!

-stephen diverdi
-stephen.diverdi@gmail.com

sdiverdi
10-26-07, 07:32 PM
Oh, I'm sorry, I forgot to mention...I'm writing this in OpenGL, and GLSL for my shader. And the three textures hooked up to the FBO are written to simultaneously using MRT with a single shader.

sdiverdi
10-30-07, 02:24 PM
Talking on the OpenGL.org forums, all signs point to this problem being a hardware or driver limitation on the amount of memory attached to an FBO. Can anyone at NVIDIA confirm this?

jolle
10-30-07, 02:29 PM
Talking on the OpenGL.org forums, all signs point to this problem being a hardware or driver limitation on the amount of memory attached to an FBO. Can anyone at NVIDIA confirm this?
I think you might have to contact Nvidia for that sort of stuff.
Dunno how to do that, but this page might be a good starting point.
http://developer.nvidia.com/object/contact_us.html

sdiverdi
10-30-07, 05:46 PM
I think you might have to contact Nvidia for that sort of stuff.
Dunno how to do that, but this page might be a good starting point.


Hmm, well there used to be NVIDIA ppl on these forums, but maybe that's not the case anymore. In the meantime, I've found the NVIDIA Developer's forums and asked on there. http://developer.nvidia.com/forums/index.php Thanks!