Compute & Tessellation

Moving on from our look at gaming performance, we have our customary look at compute performance, bundled with a look at theoretical tessellation performance.

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.

Unlike our gaming benchmarks, Civ V’s compute performance strongly favors NVIDIA’s Fermi family here, giving the GTX 550 a very strong showing. At 136fps it’s competitive with the much more expensive 6950 and even comes quite close to the GTX 460 768MB. The 6850 and 5770 meanwhile fall well behind.

What’s quite interesting is that with a bit more gas, the Zotac AMP really climbs up the charts. At 150fps it’s tailing the GTX 460 1GB and even AMD’s single-GPU king, the 6970. And ultimately while this benchmark is compute bound, it’s also memory bound at least part of the time, which is why the GTX 550 outperforms the GTS 450 by 21%.

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.

SmallLuxGPU also gives NVIDIA a solid standing here, however not quite as much as with Civ V. The GTX 550 beats the 5770 and even pulls ahead of the 6850 here. The benefit from the memory bandwidth increase is minimal however, leading to a lower 15% gain versus the GTS 450, and a larger 25% gap versus the GTX 460.

Our final compute benchmark is a Folding @ Home benchmark. Given NVIDIA’s focus on compute for Fermi, cards such as the GTX 560 Ti can be particularly interesting for distributed computing enthusiasts, who are usually looking for a compute card first and a gaming card second.

We do not normally classify Folding@Home to be very memory bandwidth sensitive on GPUs, which makes our results here more interesting than we expected. A 17% gain over the GTS 450 is in-line with the core clock increase, but achieving 81% of the GTX 460 is no small feat. With the AMP’s overclock, the card closes to within 10% of the GTX 460.

At the other end of the spectrum from GPU computing performance is GPU tessellation performance, used exclusively for graphical purposes. With Fermi NVIDIA bet heavily on tessellation, and as a result they do very well at very high tessellation factors. With 1 GPCs the GTX 560 Ti can only retire 1 triangle/clock however, which nullifies much of the architectural advantage on paper.

Even with extreme levels of tessellation, the lower geometry throughput of the GTX 550 keeps a lid on performance. It’s consistently around 80% of the performance of the GTX 460, however the hit for using extreme tessellation ultimately is greater here than on NVIDIA cards with more geometry throughput. As NVIDIA likes to note it has quite a bit more geometry throughput than older generation cards, but tying geometry performance to the number of SMs means parity between high and low-end cards is lost.

The DX11 Detail Tessellation sample reinforces those findings. NVIDIA does keep up their solid scaling from medium to high tessellation versus AMD though., including a very surprising score versus the 5870 and 6870 at high tessellation.

Wolfenstein Power, Temperature, & Noise
Comments Locked

79 Comments

View All Comments

  • Ryan Smith - Tuesday, March 15, 2011 - link

    Our experience with desktop Linux articles in the past couple of years is that there's little interest from a readership perspective. The kind of video cards we normally review are for gaming purposes, which is lacking to say the least on Linux. We could certainly try to integrate Linux in to primary GPU reviews, but would it be worth the time and what we would have to give up in return? Probably not. But if you think otherwise I'm all ears.
  • HangFire - Tuesday, March 15, 2011 - link

    All I'm asking for is current and projected CUDA/OpenCL level support, and what OS distro's and revisions are supported.

    You may not realize it, but all this GPGPU stuff is really used in science, government and defense work. Developers often get the latest and greatest gaming card and when it is time for deployment, middle end cards (like this one) are purchased en masse.

    Nividia and AMD have been crowing about CUDA and OpenCL, and then deliver spotty driver coverage for new and previous generation cards. If they are going to market it heavily, they should cough up the support information with each card release, we shouldn't have to call the corporate rep and harangue them each and every time.
  • Belard - Wednesday, March 16, 2011 - link

    Someone who already has a GF450 would be a sucker to spend $150 for a "small-boost" upgraded card.

    When upgrading, a person should get a 50% or better video card. A phrase that never applies to a video card is "invest" since they ALL devalue to almost nothing. Todays $400~500 cards are tomorrows $150 cards and next weeks $50.

    So a current GF450 owner should look at a GF570 or ATI 6900 series cards for a good noticeable bump.
  • mapesdhs - Wednesday, March 16, 2011 - link


    Or, as I've posted before, a 2nd card for SLI/CF, assuming their mbd
    and the card supports it. Whether or not this is worthwhile and the
    issues which affect the outcome is what I've been researching in recent
    weeks. Sorry I can't post links due to forum policy, but see my earlier
    longer post for refs.

    Ian.
  • HangFire - Friday, March 18, 2011 - link

    I wasn't really suggesting such an upgrade (sidegrade). I was just saying that each generation card at a price point and naming convention (450->550) should have at least a little better performance than card it replaces.
  • Calabros - Tuesday, March 15, 2011 - link

    tell me a reason to NOT prefer 6850 over this
  • 7Enigma - Tuesday, March 15, 2011 - link

    So basically this is my 4870 in a slightly lower power envelope with DX11 features. I'm shocked the performance is so low honestly. Thanks for including the older cards in the review because it's always nice to see I'm still chugging along just fine at my gaming resolution (1280X1024) 19".
  • 7Enigma - Tuesday, March 15, 2011 - link

    Forgot to add, which I bought in Jan 2009 for $180 (Sapphire Toxic 512meg VaporX, so not reference design)
  • mapesdhs - Tuesday, March 15, 2011 - link


    You're the target audience for the work I've been doing, comparing cards at
    that kind of resolution, old vs. new, and especially where one is playing older
    games, etc. Google for, "Ian PC Benchmarks", click the 1st result, then select,
    "PC Benchmarks, Advice and Information". I hope to be able to obtain a couple
    of 4870s or 4890s soon, though there's already a lot of 4890 results included.

    Ian.
  • morphologia - Tuesday, March 15, 2011 - link

    Why in the name of all that's graphical would you use this Noah's Ark menagerie of cards but leave out the 4890? It doesn't make sense. If you're going to include 4000 series cards, you must include the top-of-the-line single-GPU card. It's proven to be quite competitive even now, against the lower-level new cards.

Log in

Don't have an account? Sign up now