Test Setup And Methodology
| System Hardware | |
|---|---|
| Processor | Intel Core Duo e4300, 1.8 GHz, 1400 FSB, 2 MB Cache |
| Motherboard | ASUS P5B, BIOS: 1604 |
| RAM | Wintec Ampo PC2-6400, 2x 1024 MB, CAS 5.0-5-5-16 |
| Hard Drive | Western Digital Caviar WD2500JS250 GB, 7200 RPM, 8 MB Cache, SATA 300 |
| Networking | Realtec onboard RTL8168/8111 Gigabit Ethernet NIC |
| Graphics Cards | EVGA GeForce 8800 GTX, 768 MB RAMSPARKLE GeForce 7900 GS, 256 MB RAM |
| Power Supply | Thermaltake Toughpower 1200w |
| System Software & Drivers | |
| OS | Microsoft Windows XP Pro 5.1.2600 |
| DirectX Version | 9.0c (4.09.0000.0904) |
| Platform Driver | Intel INF 8.0.1.1002 |
| Graphics Driver | Nvidia ForceWare 162.18 |
A few notes about the test system and our benchmarking methodology:
The ambient temperature in the lab was 23.5 degrees Celsius in our tests.
All coolers were applied with the included Thermalright thermal paste.
The coolers were tested in a closed Gigabyte Aurora case with a silent 120 mm fan below the power supply, and one in the front bottom of the case for air intake. The CPU was cooled with the CoolIT Eliminator, which also provides a 120 mm exhaust fan by the CPU.
Temperatures were recorded as reported from the video card's temperature sensor; this was the only way to achieve a "control." I would have preferred to use an IR thermometer, but the coolers were so vastly different in their configurations that there was no place from which to get a consistent reading across all the products. The on-die sensor results were consistent and repeatable.
Load temperatures were recorded after 10 minutes of stress-testing using the freeware ATITool utility, version 0.26. Few benchmarks will raise graphics processor temperatures more than ATITool's "scan for artifacts" option, so real-world temperatures experienced when gaming will likely never hit the load levels we see in these tests.
Sound levels were not tested as neither Thermalright VGA cooler included a fan. Any sound tests would have indicated the noise levels of the fans instead of the coolers.
Now onto the benchmarks.
