Testing NVIDIA's Linux Threaded OpenGL Optimizations
Phoronix: Testing NVIDIA's Linux Threaded OpenGL Optimizations
With the NVIDIA 310.14 Beta driver introduced at the beginning of this week there are some OpenGL performance improvements in general plus an experimental threaded OpenGL implementation that can be easily enabled. In this article are benchmarks from the NVIDIA GeForce GTX 680 with this new Linux driver release.
Impressive decrease of performance "in applications that heavily rely on synchronous OpenGL calls such as glGet*". At least for vdrift I can tell that it is using glGet calls in the GL2-1 backends. But it is interesting to see how much difference it makes.
This is the thing I don't like about OpenGL api. It makes it very easy to write very inefficient code.
Can you confirm that WoW without that patch but with this on delivers the same performance?
Well, I never got ae's patch to work, so I can only refer to others benching of that. That said, I can confirm that the WoW without ae's patch, but with nvidias threading provides vastly better performance (when you need it, think during raid and battleground) than without threading optimization. The effect is supposedly twice what you get for ae's patch, but the big bone is that it actually works for everybody and it is stable. You definitely do not want to use ae's patch together with nivida's optimization, they both try to achieve the same.