Gaming Performance (Cont'd)

While our previous tests have shown a minimal impact and excellent compatibility in utilizing ATI CrossFire on the ASUS P5B Deluxe, our next tests will reveal a couple of driver bugs and a complete omission of compatibility within the current Catalyst 6.9 drivers.

Half Life 2: Lost Coast

The Half Life 2: Lost Coast level demonstration first introduced Valve's HDR technology. While certain developers have implemented HDR, there have been several issues with antialiasing (even on ATI cards) being an option. Valve placed a high value on building an HDR implementation that everyone can use with whatever AA/AF settings they want to select.

We use the built-in timedemo feature to benchmark the game. Our timedemo consists of starting at the bottom of the hill near the lake and ending in the old church. The Source engine timedemo feature is similar to the nettimedemo of Id's Doom 3 engine, in that it plays back more than just the graphics. The highest visual quality settings possible were used with HDR turned on. While the Source engine is notorious for giving great frame rates for almost any hardware setup, we find the game isn't as enjoyable if it isn't running at 35fps or above.


Our first issue with the drivers on P965 appears as HDR is not supported currently in our test benchmarks. The P965 CrossFire solution generated the same results with HDR on or off. We ran the 975X with HDR off and it consistently generated lower numbers than the P965. We turned our attention to viewing the benchmark several times and noted that the 975X CrossFire setup generates slightly better water and sunlight images, although it is difficult to see the difference without resorting to comparing individual screenshots.

At this time it seems that the P965 CrossFire solution is not generating the same amount of detail as the 975X setup. Our other issue is the abysmal performance of the 7950GX2 on the P965 in this demo. This is a total reversal of our earlier testing and considering our benchmark is more sensitive to memory and CPU choices we determined this is probably a driver related issue. We tested with a single X1950 card and the results were almost equal between the two boards in this benchmark. We are doing further driver regression testing and have contacted both ATI and NVIDIA about our issues.

Quake 4

There has always been a lot of debate in the community surrounding pure timedemo benchmarking. We have opted to try the NetTimeDemo option instead of our normal timedemo test for benchmarking Quake 4. This means our test results focus on the capability of each platform to render Quake 4 in a way that one could experience during game play, testing has shown the performance results to be relatively consistent. NetTimeDemo will usually generate lower scores than our normal timedemo. Additionally, Quake 4 limits frame rates to 60 fps during gameplay whether or not VSync is enabled. Remember that the questions we are trying to answer today have only to do with the graphics subsystem performance between the P965 and 975X chipsets.

Our benchmark utilizes the IdNetDemo demo recording. This includes mainly outdoor areas with numerous players trying to kill each other. We tested the game with Ultra Quality settings (uncompressed normal maps), and we enabled all the advanced graphics options except for VSync. Id does a pretty good job of keeping frame rates consistent so in-game frame rates above 30 are acceptable for single player and 60 for multiplayer.


This time our issue is that ATI is not providing P965 specific OpenGL CrossFire compatibility until a later driver release. We understand this driver release will occur early next year. However, although the driver does not have OpenGL capability for CrossFire on the P965 at this time you can still play the game. In fact, the OpenGL driver optimizations that ATI has recently implemented allowed us to comfortably play this game and provided excellent visuals in both outdoor and indoor sections. In a reversal from the Half Life 2: Lost Coast scores our 7950GX2 scores on the P965 are back to normal.

Company of Heroes

Company of Heroes was recently released and is proving to be a very addictive RTS game around the office. The game is extremely GPU intensive and also requires a hefty CPU at times. If this is beginning to sound a lot like Oblivion, it is because CoH is very similar to Oblivion in system requirements. The visuals and audio experience within the game will at times have you believing the game is based more on a First Person Shooter than a traditional Real Time Strategy game. We set all options to high and turned on all additional video options. We have found the ultra settings to be almost unplayable with a single card and barely playable in certain areas with SLI or CrossFire at the higher resolutions.

The game contains a built-in performance test that utilizes the game engine to generate several different action scenes along with an argument about coffee as a sideline distraction before the war starts. We found the performance test gives a good indication of how well your system will perform throughout the game, though we have found some of the in-game action sequences to be more demanding than the performance test. We generally found the game to be very enjoyable with an average frame rate above 35fps.


Our final issue with our test results is that from all indications the Catalyst 6.9 drivers have not been optimized for this game yet with the P965. The CrossFire scores on the P965 are equal to our single card scores in compatibility testing. We have contacted ATI about this issue and hopefully will have a response shortly. Even though we could not get CrossFire to work properly on the P965 board we did not notice any further issues during game play and actually found our CrossFire experience on the 975X board to be more enjoyable than the 7950GX2 from both a performance and visual aspect. Our P965 7950GX2 solution continues to generate overall results that are slightly better than the 975X board.

Quick Take

Our experience to date with CrossFire capability on the ASUS P5B Deluxe board has been surprising to us. We expected a lot less performance along with a lot more issues. We are in the middle of doing compatibility testing on over three dozen games at this time and so far the vast majority have worked flawlessly. While not having OpenGL compatibility is disappointing, we can understand why ATI has chosen to remove this feature until they can guarantee performance and stability in OGL games. We are more concerned about the lack of HDR CrossFire capability in an older title like Half Life 2; hopefully this can be corrected quickly, or it might be a system/motherboard specific issue. We are still testing this title along with Episode One and should have a resolution shortly. In regards to the Company of Heroes issue, until we hear back from ATI it will be difficult to determine if it is a driver or game issue. It was not until the 1.02 patch that SLI worked properly with the game so we will give ATI the benefit of doubt at this time, even though CrossFire worked perfectly on the 975X based ASUS P5W-DH Deluxe board.

Overall, a three to nine percent performance hit in our limited testing is not that bad, although the more performance minded enthusiasts will likely consider the 975X boards for CrossFire performance. However, with the recent price reductions on the Asus P5-B Deluxe this board becomes an attractive alternative. We believe ASUS has done a remarkable job in such a short period of time in getting a CrossFire capable BIOS ready for the P5B Deluxe. We have to give extra credit to ASUS as within a couple of days after the initial bios release they were able to enhance performance and implement a CrossFire auto-sensing function to enable DMI optimizations. These optimizations directly led to performance improvements of up to 9% in most of our game titles compared to the previous bios. Hopefully, they will have optimizations ready for CrossFire solutions other than the X1900 and X1950 chipsets shortly although with the lack of peer-to-peer writes in the P965 it may be impossible. Our next comparion will look at the performance of the 6.10 drivers, new 7.10 beta bios, and ATI X1950 CrossFire on this board. We still have a significant amount of testing to complete but at this time it appears ATI CrossFire works as advertised on the P965 platform and maybe just that much better on the ASUS P5B Deluxe.

Gaming Performance
Comments Locked

22 Comments

View All Comments

  • Madellga - Monday, September 25, 2006 - link

    http://www.xtremesystems.org/forums/showthread.php...">http://www.xtremesystems.org/forums/showthread.php...
    Coolaler has posted on XS some benchmarks for SLI on 975 and 965 chipsets.
    True, there are small differences between the setups (FSB and multis), but CPU speed was close enough.

    You can see an impact between 15% and 20% for going from the 975 to 965 when running SLI with hacked drivers.

    This other thread has SLI on the 965 for a 7600GT, but no comparison to the 975
    http://www.xtremesystems.org/forums/showthread.php...">http://www.xtremesystems.org/forums/showthread.php...
  • Gary Key - Monday, September 25, 2006 - link

    I have seen basically the same results in internal testing here, just that I cannot publish those numbers. ;-)
  • Maruta731 - Saturday, September 23, 2006 - link

    It will be worthwhile to test the X1900GT and X1800GTO. Dongle-less Crossfire depends entirely on the PCIe interface for inter-card communication; it may expose a significant bottleneck in the 965 board's PCIe x4 slot. And realistically, many users who decided to save a few bucks with a 965 board will probably want to use these mainstream cards.

    I have also heard somewhere that running two X1900GTs in Crossfire requires a RD580 board. Always wondered if it's true, because then Conroe users effectively lose this option.
  • Gary Key - Monday, September 25, 2006 - link

    quote:

    It will be worthwhile to test the X1900GT and X1800GTO.
    I will get a board up to Derek for further testing shortly. I do not have the dongle-less CrossFire cards at this time.
  • jive - Monday, September 25, 2006 - link

    Interesting phrase, as I thought SLI is not officially supported on any of these two platforms:

    This simply means that as games become increasingly complex and data bandwidth increases then the differences between the P965 and 975X in CrossFire or SLI operation will widen.

    Any comments?
  • Gary Key - Monday, September 25, 2006 - link

    This was more of a forward looking statement (already addressed the unofficial SLI driver situation) and also to address any issues with readers saying they can run SLI on their Intel boards now. If NVIDIA and Intel ever come to an agreement then the statement holds true, if you decided to run the unofficial SLI drivers on the P965 then the overhead from the hacked drivers along with the DMI slowdown will create a performance delta between the 975X and P965.
  • Madellga - Saturday, September 23, 2006 - link

    Gary, nice article. I think it is the first time I see a comparison on the performance hit caused by the x4 bottleneck. It would be interesting also to see in a future article how does a pair of 7900 GTXs running in SLI using the hacked drivers stacks against the XFire solution.

    I wonder if this move will drive Nvidia in the future to allow also SLI on the 975 or 965 chipsets. We know it can be done, perhaps it is a commercial or licensing problem that is blocking it.
  • nlr_2000 - Sunday, September 24, 2006 - link

    But like the article pointed out, it's not the x4 PCIe slot but the interface between the Northbridge/Southbridge causing the performance hit.
  • formulav8 - Saturday, September 23, 2006 - link

    Please bring back the bar graphs. That is all.



    Jason
  • Frumious1 - Saturday, September 23, 2006 - link

    Hell no! Not for this type of article - leave the scaling graphs in place. I would much rather see one graph (with numbers below) showing how cards scale from 1280x1024 through 1920x1200 instead of three separate graphs. Single bar charts are only effective when showing one data set.

Log in

Don't have an account? Sign up now