Testbed and Methods
We are going to test the graphics performance using the following universal testbed:
- Intel Core i7-975 Extreme Edition processor (3.33 GHz, 6.4 GT/s QPI);
- Scythe SCKTN-3000 Katana 3 CPU cooler;
- Gigabyte GA-EX58-Extreme mainboard (Intel X58 Express chipset);
- Corsair XMS3-12800C9 (3 x 2 GB, 1333 MHz, 9-9-9-24, 2T);
- Samsung Spinpoint F3 HDD (1 TB, 32 MB buffer, SATA II);
- Ultra X4 850 W modular power supply;
- Dell 3007WFP monitor (30", 2560x1600 @ 60 Hz max display resolution);
- Samsung BX2331 LED monitor (23", 1920x1080 @ 60 Hz max display resolution);
- Microsoft Windows 7 Ultimate 64-bit.
As you can see there is a new component in our testing rig. We decided to use three Samsung BX2331 LED monitors for a number of reasons. In order to keep overall costs down we opted to use smaller caliber FullHD-enabled displays instead of a much pricier WQXGA 2560x1600 high-end model. Secondly, Samsung BX2331 uses LED backlighting and therefore boasts lower power consumption and offers slimmer-looking setup. This means that there will be more room on your desk after three monitors like that settle there for good. Please keep in-mind that this monitor cannot be rotated and has no wall mount in the back.
We used the following ATI Catalyst and Nvidia GeForce drivers:
- Nvidia GeForce 280.26 WHQL for Nvidia GeForce;
- ATI Catalyst 11.8 for ATI Radeon HD.
The ATI Catalyst and Nvidia GeForce graphics card drivers were configured in the following way:
- Anti-Aliasing: Use application settings/4x/Standard Filter
- Morphological filtering: Off
- Tessellation: Use application settings
- Texture Filtering Quality: High Quality
- Enable Surface Format Optimization: Off
- Wait for vertical refresh: Always Off
- Anti-Aliasing Mode: Adaptive Multi-sample AA
- Other settings: default
- Texture filtering – Quality: High quality
- Vertical sync: Force off
- Antialiasing - Transparency: Multisampling
- CUDA – GPUs: All
- Set PhysX configuration: Auto-select
- Ambient Occlusion: Off
- Other settings: default
The image quality in the game was set to the maximum.
The games configuration files weren’t modified in any way, because the ordinary user doesn’t have to know how to do it. We ran our tests in the following premium resolutions: 1920x1080, 2560x1600 and 5760x1080.
Here are the cards that participated in our today’s test session.
- AMD Radeon HD 6990
- AMD Radeon HD 6970
- Nvidia GeForce GTX 590
- Nvidia GeForce GTX 560 Titanium SLI
We selected these cards for a number of reasons. Triple-monitor configurations will most likely be built by those enthusiasts who intend to invest a substantial amount of money into their gaming system. On top of that Nvidia’s approach requires two physical graphics chips, so in most cases an SLI configuration will be the only way to go.
We measured the average and minimum performance using Fraps utility version 3.4.6. Each test scenario was repeated three times and the average value of the three runs was taken for the analysis, as always. An extreme resolution like 5760x1080 requires a lot of horsepower and in a way it is an uncharted territory not only in terms of picture quality but performance as well. That is why we decided to run the tests in two modes: traditional high-quality with full-screen antialiasing and the less stressful mode with FSAA disabled.