Single page Print

Our testing methods
All tests were run three times, and their results were averaged, using the following test systems.

System Athlon 64 Dual Opteron
Processor AMD Athlon 64 3200+ 2.0GHz AMD Opteron 246 2.0GHz (2 processors)
Front-side bus HT 16-bit/800MHz downstream
HT 16-bit/800MHz upstream
Motherboard Abit KV8-MAX3 Tyan Tiger K8W
North bridge VIA K8T800 AMD-8151 AGP tunnel
South bridge VIA VT8237 AMD-8111 I/O hub
Chipset drivers Hyperion 4.53 AMD chipset driver pack 2.10
Memory size 512MB (1 DIMM) 2GB (4 DIMMs)
Memory type Corsair XMS3500 PC3000 DDR SDRAM Corsair CM72SD512RLP-3200/S Registered PC3200 DDR SDRAM
Graphics ATI Radeon 9800 XT NVIDIA GeForce 6800 Ultra
Graphics driver Catalyst 4.9 ForceWare 61.76
Optical drive

Liteon 16X DVD-ROM

SCSI Controllers


Adaptec 29320-R

SCSI Driver


Adaptec 3.00S4


Hitachi Deskstar 7K250 SATA 250GB

Maxtor 740X-6L ATA/133 40GB
Maxtor Atlas 10K IV SCSI 147GB
Seagate Cheetah 10K.6 SCSI 147GB
Western Digital Raptor WD360GD SATA 37GB
Western Digital Raptor WD740GD SATA 74GB

Operating System Windows XP Professional SP2 with DirectX 9.0c

I put together a couple of test systems for this comparison. The first is a moderately taxing Athlon 64 3200+ with a Radeon 9800 XT and few other accessories. The second is a monster of a dual Opteron workstation with a GeForce 6800 Ultra, 2GB of RAM, and a stack of 10K-RPM hard drives. The DC voltage and AC ripple of each PSU was tested at idle and under load for both tests systems. I also used the Athlon 64 system for our temperature and noise level tests inside an Antec Lanboy enclosure with a Vantec Stealth 80mm exhaust fan.

Systems were tested at idle and under a load consisting of Folding@Home, HD Tach's read transfer test, and a looping 3DMark03 Mother Nature demo at 1600x1200 with antialiasing and anisotropic filtering cranked all the way up. I used the following versions of each application:

The test systems' Windows desktop was set at 1280x1024 in 32-bit color at an 85Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests.

All the tests and methods we employed are publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.