I elected to use Promise's FastTrak TX4200 Serial ATA RAID card on our test platform because it supports both Native Command Queuing and Tagged Command Queuing (TCQ). Although the only drive that currently supports TCQ is Western Digital's Raptor WD740GD, the drive is widely available and reasonably popular among enthusiasts. NCQ is definitely the future of Serial ATA command queuing, but TCQ is certainly an alternative for now, albeit a considerably more exclusive one.
To contrast NCQ and TCQ's impact on performance, I tested the DiamondMax 10, 7200.7, and Raptor with command queuing enabled and disabled. This should give us a good idea of where command queuing matters and where it doesn't for each drive.
Finally, I should preempt purists who will no doubt be perturbed that I'm testing 7,200-RPM desktop drives against 10K-RPM Raptors. The Raptors are technically enterprise-class drives, while the 7,200-RPM the drives are desktop products, and marketing types don't like see the two thrown in the ring together. However, arbitrary labels have never stopped PC enthusiasts from running enterprise-class hardware in their desktop PCs, so they're certainly not going to stop me. I've even thrown in a 10K-RPM SCSI drive into some of our tests to illustrate how Native and Tagged Command Queuing compare with SCSI's command queuing. The horror!
Our testing methods
All tests were run three times, and their results were averaged, using the following test systems.
|Processor||AMD Opteron 246 2.0GHz (1 processor)|
|Front-side bus||HT 16-bit/800MHz downstream|
HT 16-bit/800MHz upstream
|Motherboard||Tyan Thunder K8W|
|North bridge||AMD-8151 AGP tunnel|
AMD-8131 PCI-X tunnel
|South bridge||AMD-8111 I/O hub|
|Chipset drivers||AMD chipset driver pack 2.10|
|Memory size||1GB (2 DIMMs)|
|Memory type||Corsair CM72SD512RLP-3200/S Registered PC3200 DDR SDRAM|
|Graphics||NVIDIA GeForce FX 5200|
|Graphics driver||ForceWare 61.76|
|Operating System||Windows XP Professional SP2 with DirectX 9.0c|
The Serial ATA and SCSI cards were used in different PCI-X slots on different PCI-X busses. Each card had an entire PCI-X bus to itself, so bandwidth sharing shouldn't be an issue.
We used the following versions of our test applications:
- ZD Media Business Winstone 2004 1.0.1
- ZD Media Content Creation Winstone 2004 1.0.1
- Xmpeg 5.03 with DivX Video 5.11
- ATTO ExpressPCI Utilities v163
- Intel IOMeter v2004.07.30
- Xbit Labs File Copy Test v1.0 beta 13
- TCD Labs HD Tach v3.01
- Far Cry v1.2
- DOOM 3
The test systems' Windows desktop was set at 1280x1024 in 32-bit color at a 75Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests. All of the 3D gaming tests used the high detail image quality settings.
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.