Compute Performance

Moving on from our look at gaming performance, we have our customary look at compute performance. With AMD’s architectural changes from the 5000 series to the 6000 series, focusing particularly on compute performance, this can help define the 6990 compared to the 5970. However at the same time, neither benchmark here benefits from the dual-GPU design of the 6990 very much.

Our first compute benchmark comes from Civilization V, which uses DirectCompute to decompress textures on the fly. Civ V includes a sub-benchmark that exclusively tests the speed of their texture decompression algorithm by repeatedly decompressing the textures required for one of the game’s leader scenes.

New as of Catalyst 11.4, AMD’s performance in our Civilization V DirectCompute benchmark now scales with CrossFire at least marginally. This leads to the 6990 leaping ahead of the 6970, however the Cayman architecture/compiler still looks to be sub-optimal for this test. The 5970 has a 10% lead even with its core clock disadvantage. This also lets NVIDIA and their Fermi architecture establish a solid lead over the 6990, even without the benefit of SLI scaling.

Our second GPU compute benchmark is SmallLuxGPU, the GPU ray tracing branch of the open source LuxRender renderer. While it’s still in beta, SmallLuxGPU recently hit a milestone by implementing a complete ray tracing engine in OpenCL, allowing them to fully offload the process to the GPU. It’s this ray tracing engine we’re testing.

There’s no CrossFire scaling to speak of in SmallLuxGPU, so this test is all about the performance of GPU1, and its shader/compute performance at that. At default clocks this leads to the 6990 slightly trailing the 6970, while overclocked this leads to perfect parity with it. Unfortunately for AMD this is a test where NVIDIA’s focus on compute performance has really paid off; coupled with the lack of CF scaling and even a $240 GTX 560 Ti can edge out the $700 6990.

Ultimately the take-away from this is that for most desktop GPU computing workloads, the benefit of multiple GPU cores is still unrealized. As a result the 6990 shines as a gaming card, but is out of its element as a GPU computing card unless you have an embarrassingly parallel task to feed it.

Wolfenstein Power, Temperature, and Noise: How Loud Can One Card Get?
POST A COMMENT

130 Comments

View All Comments

  • HangFire - Thursday, March 10, 2011 - link

    As usual no reply after the first few pages... or on the topic of Linux.

    If they only had any clue how many Linux workstations get ordered with "are you sure this has the top processor, top video card and most RAM available?" on a post-it note stuck on the Req.
    Reply
  • Azfar - Wednesday, March 09, 2011 - link

    Crysis Killer that is.....Finally !! Reply
  • Euchrestalin - Wednesday, March 09, 2011 - link

    Why is this card named after Wedge Antilles? Why not Vader, Skywalker or my personal favorite Mitth'raw'nuruodo? Reply
  • gorgid - Wednesday, March 09, 2011 - link

    I'm big RADEON fan. I had many ATI cards ( 3870 x 2 , 4870x 2, 5970 etc). Most of them water cooled. So i don't care about noise. My temperatures always at 40C ( for the chips) overclocked. I was planning to sell 5970 and get 6990, but after first reviews i decided to wait to see what NVIDIA will bring in form of GTX 590.

    The AMD made great first step with DirectX 11, cards came 6 month earlier then Nvidia.
    First time ( may be not) ATI cards were on the top charts for long time.
    But this time almost 1 1/2 years past since 5970 came to the market. 6990 shows about 20% gain of performance comparing to 5970. Why somebody would pay $700 for that kind of performance?
    In my opinion NVidia will beat 6990 pretty easy, will take a crown and will keep it for a while.
    Thank you all.
    Reply
  • slickr - Wednesday, March 09, 2011 - link

    I've read the review on most of the other English sites and there seems to be a big fluke in anand's battlefield benchmark.

    Seems like yet another shady bench.
    Reply
  • Ryan Smith - Thursday, March 10, 2011 - link

    We of course strive to make the best benchmark possible. So if you believe there is a problem, I'd like to hear what you think is amiss with our benchmarks. We can't fix things unless you guys chime in and let us know what you think is wrong. Reply
  • HangFire - Monday, March 14, 2011 - link

    So, you do read past page 5 of the comments... but you still can't bring yourself to even mention Linux.

    We are working our contacts with AMD but can't get much out of them. I guess no news is bad news.
    Reply
  • andy5174 - Thursday, March 10, 2011 - link

    Is this performance based on AMD's image quality cheat?

    http://www.guru3d.com/article/exploring-ati-image-...
    Reply
  • Ryan Smith - Friday, March 11, 2011 - link

    That article is out of date as of Catalyst 11.1a. Reply
  • noxyucT(RUS) - Tuesday, March 15, 2011 - link

    куда стока мощностей это же пздец,295 жифорс в несколько раз рвет.Интересно какой процессор участвовал в тесте, и7 ито не потянет такова монстра.Жесть кароче. Reply

Log in

Don't have an account? Sign up now