Timeline for Optimal memory access when using lookup tables on GPU?
Current License: CC BY-SA 3.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 24, 2016 at 9:22 | vote | accept | russ | ||
| May 19, 2016 at 22:40 | answer | added | GroverManheim | timeline score: 6 | |
| Feb 28, 2016 at 4:51 | comment | added | Alan Wolfe | Not a full answer but the smaller amount of memory you use the better, as it will be more likely to fit in caches and have fewer cache misses. If you have interpolatable values, like you are baking out points on a curve into textures, you might check this out as a way to get higher quality curve lookup tables with less memory: blog.demofox.org/2016/02/22/… | |
| S Oct 28, 2015 at 13:29 | history | suggested | David Kuri | CC BY-SA 3.0 |
Edited out empty phrases
|
| Oct 28, 2015 at 10:28 | review | Suggested edits | |||
| S Oct 28, 2015 at 13:29 | |||||
| Oct 27, 2015 at 17:41 | comment | added | RichieSams | In addition, give this presentation a look: cvg.ethz.ch/teaching/2011spring/gpgpu/cuda_memory.pdf It's for CUDA, but it should give you a better idea of what is happening on the underlying hardware | |
| Oct 27, 2015 at 17:33 | comment | added | RichieSams | If it's a read-only lookup table, you can just use a buffer/texture. You could either pack it into one of the normal texture formats, or you can use some of the newer features of DX11 / OpenGL to have a custom format. UAV in DX11 land, or a texture / shader_image_load_store in OpenGL land. | |
| Oct 27, 2015 at 6:54 | review | First posts | |||
| Oct 27, 2015 at 10:00 | |||||
| Oct 27, 2015 at 6:52 | history | asked | russ | CC BY-SA 3.0 |