Single page Print

Benchmarking methodology
The following configuration was used in our tests:

Processor: AMD Thunderbird 1.2GHz

Motherboard: Abit KT7-RAID (Highpoint 370 ATA100 RAID controller, version 1.03 drivers)

Memory: 1x 256MB Micron Cas3 SDRAM

Video: Asus v7100 GF2 MX (Detonator 3 version 6.50 drivers)

Audio: Creative SoundBlaster Live! Value

Networking: D-Link 10/100 Ethernet

Storage: 2x Maxtor 20GB 7200RPM ATA/100 hard drive

OS: Windows 2000 Professional SP1

The following benchmarking software was used.
  • SiSoft Sandra 2001.3.7.50
  • ZD Winbench 99 1.2
  • ZD Winstone 2001 1.1
  • Bapco Sysmark 2001
  • Quake 3 Arena 1.17
  • 3DMark 2000 1.1
  • Intel Iometer 1998.10.08
The test system's Windows desktop was set at 1024x768 in 32-bit color at a 75Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests. Tests were run under using single-drive ATA66, single-drive ATA100, and two-drive ATA100 RAID 0 and RAID 1. Each hard drive was on its own IDE channel for the duration of testing to ensure the most possible bandwidth. As always, tests were run several times and the results averaged to ensure accurate data. Just to be thorough, we've also benchmarked with both NTFS and FAT32 file systems. For RAID 0, a block size of 64k was used. Because only two drives were used in the testing, RAID 10 was not tested.

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.

SiSoft Sandra
Sandra is synthetic, but she should give us a good idea of how much bandwidth is available across the different configurations.

Synthetically speaking, RAID 0 easily outdistances the competition and comes out twice as fast as RAID 1 and ATA100. Granted, this is an ideal situation, but it's good to get an idea of what kind of transfer rates are theoretically available before we delve into some real-world benchmarks. It's clear RAID 0 has a lot of potential. Let's run some more tests and see how things pan out.