Testbed and Methods

To test the performance of ATI Radeon HD 2900 XT 1GB CrossFire tandem we assembled the following standard test platform:

• Intel Core 2 Extreme X6800 processor (3.0GHz, FSB 1333MHz x 9);
• DFI LanParty UT ICFX3200-T2R/G mainboard (AMD RD600) for AMD/ATI graphics cards;
• Asus P5N32-E SLI Plus mainboard (Nvidia nForce 680i SLI) for Nvidia graphics cards;
• Corsair TWIN2X2048-8500C5 (2x1GB, 1066MHz, 5-5-5-15, 2T);
• Maxtor MaXLine III 7B250S0 HDD (250GB, Serial ATA-150, 16MB buffer);
• Enermax Galaxy DXX EGX1000EWL 1000W power supply;
• Dell 3007WFP monitor (30", 2560x1600@60Hz max display resolution);
• Microsoft Windows Vista Ultimate 32-bit;
• Nvidia ForceWare 162.22;
• ATI Catalyst Beta 7.7.

Since we believe that the use of texture filtering optimizations is not justified in this case, especially for high-end graphics solutions, the AMD and Nvidia graphics card drivers were set up to provide the highest possible quality of tri-linear and anisotropic texture filtering. We have also enabled transparent texture filtering to achieve best image quality by selecting Transparency antialiasing in multisampling mode for Nvidia GeForce and Adaptive antialiasing for AMD Radeon. As a result, our Nvidia ForceWare and AMD Catalyst settings looked as follows:

AMD Catalyst:

• Catalyst A.I.: Standard
• Mipmap Detail Level: High Quality
• Wait for vertical refresh: Always off
• Adaptive antialiasing: On
• Temporal antialiasing: Off
• High Quality AF: On
• Other settings: default

Nvidia ForceWare:

• Texture Filtering: High quality
• Vertical sync: Off
• Trilinear optimization: Off
• Anisotropic optimization: Off
• Anisotropic sample optimization: Off
• Gamma correct antialiasing: On
• Transparency antialiasing: On (multi-sampling)
• Other settings: default

We selected the highest possible graphics quality level in each game using standard tools provided by the game itself. The games configuration files weren't modified in any way. Performance was measured with the games' own tools or, if not available, manually with Fraps utility version 2.9.1. We also measured the minimum speed of the cards where possible.

Since CrossFire tandem including two Radeon HD 2900 XT 1GB is a high-end solution we performed our tests in 1600x1200, 1920x1200 and 2560x1600 resolutions. The games that didn't support 16:10 ratio were run in 1920x1440 or 2048x1536 resolutions respectively. We used "eye candy" mode everywhere, where it was possible without disabling the HDR or Shader Model 3.0. Namely, we ran the tests with enabled anisotropic filtering as well as MSAA 4x. We enabled them from the game's menu. If this was not possible, we forced them using the appropriate ForceWare or Catalyst driver settings.

Besides the Radeon HD 2900 XT 1GB CrossFire tandem we also tested one of these cards in single-card mode. For the sake of more illustrative comparison we also included the following AMD and Nvidia solutions:

• AMD Radeon HD 2900 XT (R600, 742/1650MHz, 320sp, 16tmu, 16rop, 512-bit, 512MB)
• Nvidia GeForce 8800 Ultra (G80, 612/1512/2160MHz, 128sp, 32tmu, 24rop, 384-bit, 768MB)
• Nvidia GeForce 8800 GTX (G80, 576/1350/1900MHz, 128sp, 32tmu, 24rop, 384-bit, 768MB)
• Nvidia GeForce 8800 GTS (G80, 513/1188/1600MHz, 96sp, 24tmu, 20rop, 320-bit, 640MB)

For our tests we used the following games and benchmarks:

First-Person 3D Shooters

• Battlefield 2142
• Call of Juarez
• Far Cry
• F.E.A.R. Extraction Point
• Tom Clancy's Ghost Recon Advanced Warfighter
• Half-Life 2: Episode One
• Prey
• S.T.A.L.K.E.R.: Shadow of Chernobyl

Third-Person 3D Shooters

• Hitman: Blood Money

• Tomb Raider: Legend
• Splinter Cell: Double Agent

RPG

• Gothic 3
• Neverwinter Nights 2
• The Elder Scrolls IV: Oblivion

Simulators

• X3: Reunion

Strategies

• Command & Conquer: Tiberium Wars
• Company of Heroes
• Supreme Commander

Synthetic Benchmarks

• Futuremark 3DMark05
• Futuremark 3DMark06