AMD OpenCL APP SDK Beats Intel's Own SDK On Ivy Bridge

[AnonymousCoward]

Isn't the GTX 680's Compute preformance driver limited? To drive people to the nVidia Tesla/Quadro range? I know the 7970 outpreforms it in a number of compute tests. (Was the exception direct compute? that wasn't capped somehow?)

Here is the collection of the 'Room' scene top 20 benchmarks. Top is 8* 580's, 4 7970's and you can see the 7970's are alot faster then the 580's (OpenCL cripple drive?r)
http://www.luxrender.net/luxmark/top/top20/Room

there was/is some performance regression in OpenCL driver on nVidia site. also Geforce 680 and other chips from this family is crap for GPGPU. nVidia strip down every feature which is not needed for DX/OGL rendering and it is not driver/SW limited. nVidia will bring two different GPUs. one for gaming purpose and second for Tesla GPGPU purpose. on other hand new Northern Island aka Radeon 7xxx chips are huge improvement for GPGPU on AMD side.

[wstorm]

It's funny because AMD can't even write software, so what does that say about intel.

hah, i have tried intel code composer studio once. Thats lolware.

[Ansla]

Radeon HD7970: 4 or 5
Nvidia GTX 680: 8 to 10

Where did you get those values from? Your ass? Every benchmark I saw of those cards had results ranging from GTX680 insignificantly faster then HD7970 for some tests to HD7970 being more then twice as fast as GTX680 for other tests.

[Wilfred]

Where did you get those values from? Your ass? Every benchmark I saw of those cards had results ranging from GTX680 insignificantly faster then HD7970 for some tests to HD7970 being more then twice as fast as GTX680 for other tests.

I agree with you, just look at:

http://www.geeks3d.com/20120130/luxmark-2-0-opencl-benchmark-for-gpus-and-cpus/

It appears radeon is much better for gpgpu than geforce.