CrossFireX Arrives: First Look at 3 and 4 GPUs in 2 Card Setups
by Derek Wilson on March 7, 2008 12:05 PM EST- Posted in
- GPUs
S.T.A.L.K.E.R. Performance
Version: 1.0005
Settings: full dynamic lighting, everything maxed without AA and no grass shadows.
For this test, we walk in a straight line for about 30 seconds and use FRAPS to measure performance. We use the same save game every time and the path doesn't change. Our performance measurements are very consistent between runs. We do two runs and take the second.
S.T.A.L.K.E.R. is a strange one when it comes to performance. At some resolutions, quad-GPUs helps somewhat, but generally speaking the sweet spot is three GPUs. What's really perplexing is that we see quad help more at 1280x1024 than at 1600x1200 and 1920x1200. The performance benefits of three GPUs is also a bit erratic; though clearly at 2560x1600 the difference is noticeable, that resolution also demonstrates better scaling from two to three cards than from one to two cards - not something we would normally expect to see.
What we can say for certain is that the latest version of STALKER is not performing as well as we would expect in a variety of ways. It's not unusual to see multi-GPU technologies run into CPU limitations at lower resolutions and offer better scaling at higher resolutions, but that's not what we're seeing. Instead, we have our lowest and highest resolutions benefiting more from CrossFire (and CrossFireX) than our middle resolutions. We are looking into the matter more to see if we can determine what is actually happening; this could be something caused by the Skulltrail platform, the 1.0005 patch, the AMD drivers… or more likely than not all of these things working together.
CrossFireX has the potential to add value at all resolutions in S.T.A.L.K.E.R., but at present there are some anomalies (Ed: the kind they pay good money for around Chernobyl, we hear). Due to the nature of performance, gameplay and location in the game could make for huge variations in performance. Looking at the sky will push frame rate way up into the hundreds, so keeping more sky in your view than ground will improve your framerate. Thus, people who normally look higher while running around will experience better performance than those who look lower.
It's very hard with data like this to accurately assess the value of CrossFireX for this particular game. Based on the data we do have, it again seems that three GPUs is the sweet spot rather than four, in spite of the fact that four GPUs can help at both higher and lower resolutions.
| S.T.A.L.K.E.R. Performance | ||||
| 1280x1024 | 1600x1200 | 1920x1200 | 2560x1600 | |
| NVIDIA GeForce 9600 GT SLI | 94.3 | 72.8 | 62.6 | 39.9 |
| NVIDIA GeForce 8800 Ultra SLI | 110.3 | 94 | 85.2 | 61.6 |
| NVIDIA GeForce 8800 Ultra | 62.3 | 53.5 | 48.8 | 33.3 |
| NVIDIA GeForce 9600 GT | 51 | 38.7 | 33.3 | 21 |
| AMD Radeon HD 3870X2 (x 2) | 111 | 84.3 | 67.3 | 39.2 |
| AMD Radeon HD 3870X2 + 3870 | 103.5 | 83.4 | 68.2 | 36.9 |
| AMD Radeon HD 3870X2 | 86.3 | 68.5 | 59.6 | 24.1 |
| AMD Radeon HD 3870 | 43.7 | 34.3 | 29.5 | 17.5 |
Performance makes sense in general, but it is important to note the sharp relative drop of the 3870X2 at 2560x1600. This is a case where the 3870X2 doesn't appear to have the power to reach the next performance plateau, and adding in another graphics card really made the difference. That, or it's just another driver glitch that needs some ironing.
Facebook
Twitter
RSS
36 Comments
View All Comments
DerekWilson - Saturday, March 8, 2008 - link
that is key ... as is what ViRGE said above.in addition, people who want to run 4 GPUs in a system are not going to be the average gamer. this technology does not offer the return on investment anyone with a midrange system would want. people who want to make use of this will also want to eliminate any other bottlenecks to get the most out of it in their systems.
not only does skulltrail help us eliminate bottlenecks and look at the potential of the graphics subsystem, in this case i would even make the argument that the system is a good match for the technology.
Sind - Saturday, March 8, 2008 - link
I agree, I don't think the Skulltrail is doing anyone favours of how they could judge utilising these MGPU solutions in a "average" system that the reader on Anand would be using. X38 seems very popular as is 780i, I really don't think even more then 1% of your traffic would ever utilise the system you used to do this review. I've read the other CrossfireX reviews from around the net, and most had no problems at all, and infact most noted that it worked straight out with no messing around with the lengthy directions that were indicated in the article to get it to work.ViRGE - Saturday, March 8, 2008 - link
Something very, very important to keep in mind is that Skulltrail is the only board out right now that supports Crossfire and SLI. If AT wants to benchmark both technologies without switching the boards and compromising the results, this is the only board they can use.Cookie Monster - Saturday, March 8, 2008 - link
No 8800Ultra or GTX Tri-SLI for comparison?DerekWilson - Saturday, March 8, 2008 - link
we were looking at 2 card configurations here ... i'll check out three and four card configs laterJarredWalton - Saturday, March 8, 2008 - link
Unfortunately, Tri-SLI requires a 780i motherboard. That's fine for Tri-SLI, but CrossFire (and CrossFireX) won't work on 780i AFAIK. I also think Skulltrail may have its own set of issues that prevent things from working optimally - but that's conjecture rather than actual testing. Derek and Anand have Skulltrial; I don't.Slash3 - Saturday, March 8, 2008 - link
...graphs are both using the same image. The Oblivion Performance and 4xAA/16AF Performance line graphs (oblivionscale.png) are just duplicates and link to the same file. :)JarredWalton - Saturday, March 8, 2008 - link
Fixed, thanks.slashbinslashbash - Saturday, March 8, 2008 - link
Graphics really are fairly unique in the computing world in that they are easily parallelized. While we're pretty quickly reaching a point of diminishing returns in number of cores in a general-purpose CPU (8 is more than enough for any current desktop type of usage), the same point has not been reached for graphics. That is why we continue to see increasing numbers of pipelines in individual GPU's, and why we continue to see effective scaling to multiple cards and multiple GPU's per card. As long as there is memory bandwidth to support the GPU power, the GPU looks like it is capable of taking advantage of much more parallelization. I expect 1000+ pipes on a 2-billion-transistor+ GPU by 2011.So, I expect multi-GPU to remain with us, but any high-end multi-GPU setup will always be surpassed by a single-GPU solution within a generation or two.
DerekWilson - Saturday, March 8, 2008 - link
that's not the issue ... graphics is infinitely parallelizeable ...the problems are die size and power.
beyond a certain die size there is huge drop off in the amount of money and IHV can make on their silicon. despite the fact that every chip could have been made larger, we are working with engineers, not scientists -- they have a budget.
multiGPU allows IHVs to improve performance nearly linearly in some cases without the non-linear increase in cost they would see from (nearly) doubling the size of their GPU.
...
then there is power. as dies shrink and we can fit more into a smaller space, will GPU makers still be able to make chips as big as R600 was? power density goes way up as die size goes down. power requirements are already crazy and it could get very difficult to properly dissipate the heat from a chips with small enough surface area and huge enough power output ... ...
but speading the heat out over two less powerful cards would help handle that.
...
in short, multigpu isn't about performance ... it's about engineering, flexibility and profitability. we could always get better performance improvement from a single GPU if it could be built to match the specs of a multiGPU config.