During our benchmarking sessions we used the configuration listed below. Between our benchmarking sessions the only variable device was the video card being tested.

Hardware Configuration

• Motherboard: DFI LAN Party 925X-T2 (Intel 925X, Sep. 20th, 2004 BIOS)
• CPU: Pentium 4 3.4 GHz LGA 775
• Memory: Two 512 MB DDR2-533 CM2X512-4200 CL4 from Corsair memory modules
• Hard Drive: Maxtor DiamondMax 9 Plus (40 GB, ATA-133)
• Screen resolution: 1024x768x32@85 Hz

Software Configuration

• Windows XP Professional installed using NTFS
• Service Pack 2
• Direct X 9.0c
• Intel inf driver version: 6.0.1.1002
• ATI video driver version: 4.10 (6.14.10.6483)
• ATI video driver version: 4.12 (Radeon X850 Platinum Edition)
• ATI video driver version: 5.3 (Radeon X800 GT)
• ATI video driver version: 5.9 (Radeon X800 GTO)
• ATI video driver version: 6.1 (Radeon X1600 XT from HIS)
• ATI video driver version: 8.173 beta (Radeon X1000 series)
• ATI video driver version: 8.203.3.0 (Radeon X1900 XTX)
• nVidia video driver version: 66.93
• nVidia video driver version: 77.72 (GeForce 7800 GTX)
• nVidia video driver version: 78.02 (GeForce 7800 GT)
• nVidia video driver version: 81.98 (GeForce 6800 GS)
• nVidia video driver version: 84.21 (GeForce 7600 and 7900 series)
• nVidia video driver version: 91.29 (GeForce 7950 GX2)

Used Software

• 3DMark2001 SE
• 3DMark03 Business 3.50
• 3DMark05 Business 1.10
• Doom III
• Far Cry 1.3

We adopted a 3% error margin. Thus differences below 3% cannot be considered relevant. In other words, products with a performance difference below 3% should be considered as having similar performance.