Introduction

Day 2 of Computex 2005 was a hive of activity. Not only did we manage to bring you more Crossfire benchmarks, as hinted at in our coverage of Day 1, but we also caught up with more vendors, specifically ASUS, Corsair and GeCube.

One observation we've made whilst roaming the show floor is the lack of nVidia G70 samples. G70 is nVidia's new graphics core, which will be labeled the 'GeForce 7' series. (ed. note - we did see G70 this morning with nVidia but now we cannot say much as we are under NDA but nVidia's usual technology demonstration was pretty fine looking) Interestingly, whilst discussing this with a product manager at ASUS, we learnt that the reason for this has much to do with attempted strong-arm tactics by nVidia. Apparently, they refused to provide reference samples to vendors that were displaying ATI Crossfire solutions, and since Crossfire is arguably the biggest announcement of the show, it's not surprising that many vendors chose to stick with ATI - hence the lack of G70.

We'll definitely be investigating this issue further, but for now let's continue with our coverage of Day 2.

Doom 3 ATI Crossfire Benchmarks

We teamed up with Hardinfo.dk for a full Doom 3 test on one of only three working Crossfire systems currently available (as reported to us by DFI's project leader). DFI was kind and hospitable enough to give us free reign of their system for most of the afternoon, whereby we proceeded to install Doom 3 and ran a variety of tests at multiple resolutions and graphics settings. We also ran UT2004 benchmarks, but found that since UT2004 is highly CPU dependent, our average FPS barely changed throughout multiple resolutions and thus, we decided to stick with Doom 3.

The specific demo we ran was one compiled by Hardinfo, called 'doom3_hardinfo_boss', which is available for download from their website. This is an in-game recording rather than the usual flyby demos that most sites use, and therefore is far more accurate at measuring in-game frame rates. The system configuration consisted of an AMD Athlon 64 4000+, two X850XT Platinum Edition graphics cards and a DFI motherboard based on the ATI X200 chipset. The graphics driver used was a beta version of Catalyst dated May 27, 2005.

Since these results on their own are meaningless, we also ran the same tests on an nVidia GeForce 6800 Ultra SLI system. The only difference between this and the Crossfire system were the motherboard, which was a Gigabyte nForce 4 motherboard and obviously the video cards.

Finally, please note that all tests were run with Doom 3 set to Ultra Quality.

X850XT Crossfire

Normal

1024x768 - 85.4

1280x1024 - 83.1

4xAA 8xAF

1024x768 - 69.6

1280x1024 - 52.2

GeForce 6800 Ultra SLI

Normal

1024x768 - 83.8

1280x1024 - 82.5

4xAA 8xAF

1024x768 - 67.8

1280x1024 - 53.4

As you can see, we found that there's little difference as far as performance is concerned between a high-end SLI and a high-end Crossfire setup. However, please keep in mind that the Crossfire system used early beta drivers (and probably early BIOS revisions), so there should be much more room for improvement.

The competition between SLI and Crossfire is definitely beginning to heat up, regardless of the performance differences (or lack thereof). DFI's project leader told us that it's possible to run dual GeForce 6 cards in an ATI X200-based motherboard, but SLI can only be enabled when an older driver set is used (he didn't specify which one). Interestingly, if two ATI cards are installed on an nForce 4 board, ATI's drivers won't install. Unfortunately, we weren't able to test this for ourselves due to time constraints and a lack of resources, but it's definitely something that we'll be keeping an eye on in future.