Looking At The OpenCL Performance Of ATI & NVIDIA On Linux

[phoronix]

Phoronix: Looking At The OpenCL Performance Of ATI & NVIDIA On Linux

Recently we provided the first Linux-based review of the NVIDIA GeForce GTX 460 graphics card. Overall, this Fermi-based graphics card was a great performer for selling around $200 USD and is complemented by great video playback capabilities with VDPAU acceleration and great proprietary driver support. In that review we primarily looked at the OpenGL performance under Linux, but with NVIDIA's Fermi architecture bringing great GPGPU advancements for CUDA and OpenCL users too, in this article we are looking more closely at the Open Computing Language performance of this GF104 graphics card as well as other NVIDIA and ATI graphics cards.

http://www.phoronix.com/vr.php?view=15257

[kernelOfTruth]

thanks for those tests !

why weren't the Cypress GPUs not included (5850, 5870) and only the smaller Junipers ?

[rudl]

And what about SmallLuxGpu ?

[Michael]

why weren't the Cypress GPUs not included (5850, 5870) and only the smaller Junipers ?

No access to such hardware...

[numasan]

What about the FirePro's? Could be fun to see if they are that much faster in OpenCL than the consumer cards.

[kernelOfTruth]

No access to such hardware...

ah - thanks for the clarification !

it's just that the results look rather in favor of Nvidia

do you have a 5830 available (afaik that's a cypress LE) ?

[Michael]

ah - thanks for the clarification !

it's just that the results look rather in favor of Nvidia

do you have a 5830 available (afaik that's a cypress LE) ?

Nope... Any and all GPUs you would have seen in a review. Only Evergreen ASICs I have are the HD 5550 and HD 5570 and an HD 5450 - that I actually bought and am in the process of reviewing.

[ssam]

do any of those benchmarks use double precision floating point?

(and as always: it would be nice if you could put error bars on the plots)

[vrodic]

Thank you Michael!

It would be nice to compare it with some CPU. So we can actually see if low end cards make sense for that processing work.

I'm not sure which OpenCL CPU implementation is efficient and uses SSE2 etc. Maybe Intel has such implementation of the OpenCL compiler, or LLVM has some OpenCL frontend/parser.

[brent]

Michael, please keep in mind that SmallPtGPU contains a bug/incompatibility that seriously limits performance on NVidia hardware, especially pre-Fermi.

Here's a diff that fixes it. This improves performance more than ten-fold on G80/GT200.