Legit Memory Reviews
AMD Socket 939 Memory Testing
|Product:||AMD Socket 939|
|Date:||Fri, Jun 18, 2004 - 09:00 AM|
|Written By:||Nathan Kirsch -|
Popular PC-3200 Memory & Socket 939 (VIA K8T800-Pro):
After being asked by our readers to look into benchmarking memory on a Socket 939 platform we have had the chance to take a look and the results are in!
Since the most popular DDR1 memory sold on the market is still PC-3200 we took a look at some memory from the "Big 3" (Corsair, Kingston, OCZ) sellers here in America. We ran all the memory at the manufacturer set timings and also tweaked the voltages to note any weird characteristics. We ran full testing and benchmarking on the following 1GB kits (2x 512mb) on our ASUS A8V Rev 1.02 motherboard:
- Corsair XMS PC-3200 LL (Winbond CH-5)
- Corsair XMS PC-3200 XL (Samsung TCCD)
- Kingston HyperX PC-3200 (Winbond CH-5)
- OCZ Platinum Ltd. Edition PC-3200 (Winbond BH-6)
Let's get on to our test system and our benchmarks.
Our Test System:
- AMD 939-Pin Athlon 64 3800+
- ASUS A8V Deluxe Rev 1.02 Motherboard (BIOS 1005)
- nVidia 5900 Ultra 256mb (nVidia reference)
- Thermaltake HSF
- Seagate 120Gb Serial ATA 150 Hard drive
- Antec True Control 550W Power Supply
- 17" Sony Trinitron monitor
All testing was done on a fresh install of Windows XP Professional build 2600 with Service Pack 1A and DirectX 9.0b. All benchmarks were completed on the desktop with no other software programs running. No overclocking was done on the video card unless noted. Our testing processor was an unlocked AMD 939-pin 3800+.
The audio, USB, Firewire, and lan features were disabled in the BIOS menu for all the testing competed during this review.
Sisoft; Sandra 2004:
Sisoft Sandra 2004 is designed to test the theoretical power of a complete system and individual components. The numbers taken though are, again, purely theoretical and may not represent real world performance. Higher numbers represent better performance in memory bandwidth.
Results: As you can tell all of the modules we tested performed virtually identically when tested under Sandra 2004. Nothing significant here so let's move on.
Next Page - Everest 1.1
Page 1 - Introduction
Page 2 - Everest 1.1