• What
    is this?
    You've landed on the AMD Portal on AnandTech. This section is sponsored by AMD. It features a collection of all of our independent AMD content, as well as Tweets & News from AMD directly. AMD will also be running a couple of huge giveaways here so check back for those.
    PRESENTED BY

Compute Performance

Moving on from our look at gaming performance, we have our customary look at compute performance. With GCN AMD significantly overhauled their architecture in order to improve compute performance, as their long-run initiatives rely on GPU compute performance becoming far more important than it is today.

With such a move however AMD has to solve the chicken and the egg problem on their own, in this case by improving compute performance before there are really a large variety of applications ready to take advantage of it. As we’ll see AMD has certainly achieved that goal, but it raises the question of what was the tradeoff for that? We have some evidence that GCN is more efficient than VLIW5 on a per-shader basis even in games, but at the same time we can’t forget that AMD has gone from 800 SPs to 640 SPs in the move from Juniper to Cape Verde, in spite of a full node jump in fabrication technology. In the long run AMD will be better off, but I suspect we’re looking at that tradeoff today with the 7700 series.

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. Note that this is a DX11 DirectCompute benchmark.

Theoretically the 5770 has a 5% compute performance advantage over the 7770. In practice the 5770 doesn’t stand a chance. Even the much, much slower 7750 is ahead by 12%, meanwhile the 7770 is in a class of its own, competing with the likes of the 6870. The 7770 series still trails the GTX 560 to some degree, but once again we’re looking at the proof of just how much the GCN architecture has improved AMD’s compute performance.

Our next benchmark is SmallLuxGPU, the GPU ray tracing branch of the open source LuxRender renderer. We’re now using a development build from the version 2.0 branch, and we’ve moved on to a more complex scene that hopefully will provide a greater challenge to our GPUs.

SmallLuxGPU is another good showing for the GCN based 7700 series, with the 7770 once again moving well up the charts. This time it’s between the 6850 and 6870, and well, well ahead of the GTX 560 or any other NVIDIA video cards. Throwing in an overclock pushes things even farther, leading to the XFX BESDD tying the 6870 in this benchmark.

For our next benchmark we’re looking at AESEncryptDecrypt, an OpenCL AES encryption routine that AES encrypts/decrypts an 8K x 8K pixel square image file. The results of this benchmark are the average time to encrypt the image over a number of iterations of the AES cypher.

Under our AESEncryptDecrypt benchmark the 7770 does even better yet, this time taking the #2 spot and only losing to its overclocked self. PCIe 3.0 helps here, but as we’ve seen with the 7900 series there’s no replacement for a good compute architecture.

Finally, our last benchmark is once again looking at compute shader performance, this time through the Fluid simulation sample in the DirectX SDK. This program simulates the motion and interactions of a 16k particle fluid using a compute shader, with a choice of several different algorithms. In this case we’re using an (O)n^2 nearest neighbor method that is optimized by using shared memory to cache data.

It would appear we’ve saved the best for last, as in our fluid simulation benchmark the top three cards are all 7700 series cards. This benchmark strongly favors a well organized cache, leading to the 7700 series blowing past the 6800 series and never looking back. Even NVIDIA’s Fermi based video cards can’t keep up.

Civilization V Theoretical Performance
POST A COMMENT

155 Comments

View All Comments

  • Articuno - Wednesday, February 15, 2012 - link

    Kepler really, really needs to come out soon, and I'm saying this as an AMD fan. Reply
  • Malih - Wednesday, February 15, 2012 - link

    Exactly, I'm a fan of AMD myself, and I can't wait for Kepler, AMD needs a kick in the (you know what) Reply
  • DimeDeviL - Wednesday, February 15, 2012 - link

    "Theoretically the 5770 has a 5% compute performance advantage over the 5770." Reply
  • eminus - Wednesday, February 15, 2012 - link

    uhhh really? same card performs better hehehehe Reply
  • tynopik - Wednesday, February 15, 2012 - link

    lobbing off -> lopping off Reply
  • mattgmann - Wednesday, February 15, 2012 - link

    I feel like reviewers were blinded by performance with the 7970/7950 cards. They offer the same lack of competitive pricing as these lower end cards. The 7950 can be compared directly (in price and performance) to the Nvidia GTX 580, a card that was available a year ago.

    I'm still rocking a pair of 4870s that set me back ~$400 a few years back. To get a substantial performance upgrade, I'd have to spend $450 on a 7950 today. Where is the value in that? Yes, power consumption and features are important but are tertiary to raw performance in almost every user scenario when it comes to gaming.

    To say the least, the lack of competitive pricing between nvidia and amd currently smells a little fishy.....it wouldn't be the first time there was price fixing in the graphics card industry.
    Reply
  • maniac5999 - Wednesday, February 15, 2012 - link

    I'm convinced that the pricing for 4870s in late 2008/early 2009 was the sweet point to buy. Yes, power consumption sucks, but other than that, a card that cost me $160 then, is pretty competitive with a card that costs $120 today. I think we may have just lucked out with the 4870s, and may need to wait until at least Kepler, if not 8XXX to upgrade. Reply
  • chizow - Wednesday, February 15, 2012 - link

    Just be careful not to set expectations on the 4870's pricing for 2 reasons:

    1) 4870 pricing was a mistake imo. AMD will say it was a calculated one in fluff pieces like the RV770 story and it was true in some degree that AMD needed to recover mindshare/marketshare and consumer confidence after the R600 debacle and a weak performing RV670. Still, when you set your single-GPU flagship at $300 and your second SKU at $200, there's not much room to go down on pricing.

    2) Late 2008 early 2009 was the height of the recession. Wall Street, Real Estate, Auto Industry, all that. Nvidia and AMD were feeling it too and got involved in a highly publicized price war. That's why you saw "new" high-end performance parts like the 4890 and GTX 275 launching for $230-$250 that occupy the $350+ market today, with cards like the 4870 and GTX 260c216 selling at tremendous value for $150 or less.

    Its obvious AMD is doing its best to correct their 4870 price mistake over the last few years, but with the overall performance of Southern Islands stack, the 7-series was the wrong time to do it. They should've just stuck to their old pricing scheme or at worst, matched Nvidia's pricing ($500, $380) with their Tahiti parts. Then you might see the rest of the stack priced reasonably.
    Reply
  • Zoomer - Wednesday, February 15, 2012 - link

    Yeah, at this point of the business cycle, it sucks to have to buy anything. Everything is expensive due to all the capacity cuts.

    Recessions are good (for buying things, cars, houses, whatever). Btw: Housing is an anomaly due to efforts to stop/slow the process.
    Reply
  • Oxford Guy - Wednesday, February 15, 2012 - link

    Recessions are good for the rich... not so good for everyone else. Reply

Log in

Don't have an account? Sign up now