A note on the testing
Product reps have a habit of freaking out whenever we throw an orange or banana into an apples-to-apples comparison, but that's not going to stop us from testing the WD740GD against not only the WD360GD, but also a pair of 7,200RPM drives, a couple of 10K SCSI disks, and even a two-drive "parallel" ATA RAID 0 array.
With a 2,800RPM spindle speed advantage, the Raptors are easy favorites over the other Serial ATA disks. However, it's important to note that the Raptors are only available in 37 and 74GB capacities, and both models are quite expensive compared to Serial ATA offerings from Maxtor and Seagate.
Since the Raptors are built for high-end workstations and servers, comparing their performance to 10K RPM SCSI drives is also appropriate. However, there are a few things to keep in mind when we throw SCSI into the mix. First, our Adaptec 29320-R SCSI controller supports command queuing, which should give our SCSI drives an edge over the WD740GD, at least until the Raptor's TCQ gets host controller support. Also, it's important to note that Serial ATA drives support a WRITE_THROUGH flag that demands that data is written directly to the disk rather than to the drive's cache. For reasons I outline in this section of our 10K-RPM hard drive comparison, I haven't disabled WRITE_THROUGH for our SCSI disks.
Our testing methods
All tests were run three times, and their results were averaged, using the following test system.
|Processor||Intel Pentium 4 2.26GHz|
|Front-side bus||533MHz (4x133MHz)|
|Motherboard||Tyan Trinity GC-SL|
|Chipset||ServerWorks Grand Champion SL|
|North bridge||ServerWorks CMIC-SL|
|South bridge||ServerWorks CSB5|
|Memory size||512MB (1 DIMM)|
|Memory type||CAS 2.5 PC2100 ECC DDR SDRAM|
|Graphics||ATI Rage XL|
SIIG Serial ATA PCI
|3ware Escalade 7500|
Silicon Image 188.8.131.52
Western Digital Raptor WD740GD 74GB
Maxtor Atlas 10K IV 147GB
|Maxtor 740X-6L 40GB|
2-drive RAID 0
|Operating System||Windows XP Professional SP1|
The Serial ATA, SCSI, and RAID cards were each used in the motherboard's PCI-X slot and had the entire PCI-X bus to themselves throughout testing.
We used the following versions of our test applications:
The test systems' 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.
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.
|Apple's A9 impresses and the Nexus strikes back: The TR Podcast 188||0|
|Google Fiber has arrived in Damage Labs||11|
|Silverstone's PT18 chassis lets NUCs run fan-free||2|
|Intel to begin shipping Skylake CPUs with SGX enabled||6|
|Premium HDMI cables will be ready for next-generation media||23|
|Microsoft acquires Havok physics engine from Intel||83|
|AMD unleashes mobile Tonga with the FirePro W7170M||14|
|Deals of the week: Crucial's MX200 500GB SSD and more||12|
|Report: TSMC makes around 6 in 10 Apple A9 SoCs||19|
|You don't need moral fortitude to see that "Intel" is an anagram for "Let In." Applying similar standards, if you look closely at "AMD", you'll notice...||+35|