Testbed and Methods

We set up our standard AMD64 platform to test the ASUS V9999 GE:

• Athlon 64 3400+ CPU (2.20GHz, 1MB L2 cache);
• ASUS K8V Deluxe mainboard;
• 2x512MB OCZ PC3200 Platinum EB DDR SDRAM, CL2.5-3-2-8;
• Seagate 7200.7 HDD (SerialATA-150, 8MB buffer);
• Creative SoundBlaster Audigy 2 sound card;
• Microsoft Windows XP Pro SP2, DirectX 9.0c;
• ATI Catalyst 4.8, NVIDIA ForceWare 61.77 (version 60.85 for 3DMark03).

We used the following benchmarks and games:

First Person 3D Shooters:

• Call of Duty;
• Doom III;
• Unreal Tournament 2004;
• Halo: Combat Evolved;
• Far Cry;
• Painkiller;
• Counter Strike: Source Beta
• Highly Anticipated DirectX 9 Game 1;
• Highly Anticipated DirectX 9 Game 2.

Third Person 3D Shooters:

• Splinter Cell: Pandora Tomorrow;
• Prince of Persia: Sands of Time;
• Max Payne 2: The Fall of Max Payne.

Simulators:

• IL-2 Sturmovik: Aces in the Sky;
• Lock On;
• Colin McRae Rally 04;

Strategy:

• Command & Conquer Generals: Zero Hour;
• Perimeter.

Semi-Synthetic Benchmarks:

• Half-Life 2 Stress Test;
• Aquamark3.

Synthetic Benchmarks:

• Futuremark 3DMark03 build 340.

In each game we selected the maximum possible graphics quality settings, the same for products from ATI and NVIDIA. We disabled anisotropic and tri-linear filtering optimizations in the ForceWare driver settings for NVIDIA’s GPUs.