4.5GB Ultra SCSI Drive Comparison

Article contributed by Anand Tech Affiliate: The Storage Review

SCSI drives have always come to mind to those looking for the ultimate in performance and expandability. The fastest hard drives available today are 10,000rpm platter drives, available only with SCSI interfaces. In addition, Wide SCSI controllers can handle up to 15 devices per channel with multiple devices using the bus simultaneously. ATA, on the other hand, has a limit of two devices per channel, one of which may be in use at any given time. SCSI is a much more intelligent interface, with queuing features that enhance performance in multithreading and multitasking operations. SCSI drives are much more expensive, however, let alone the cost of a SCSI host interface, which is not typically found on motherboards.

Debates rage on in newsgroups how much advantage SCSI hard drives have over their ATA counterparts and whether or not they are worth the extra money for the "average" user. There is no denying that the average 4.5GB SCSI hard drive seems to cost more than twice what a 4.5GB ATA drive does. SCSI drives usually have better specs than ATA drives such as higher rotation speeds, lower seek times, larger buffers. It often seems that benchmarks, including Ziff Davis' Winbench 98, do not show a large difference in performance between SCSI and ATA drives. Even the best benchmarks often seem to miss key performance-feel issues though. Here at Storage Review, we feel that the benchmark figures provided for SCSI and ATA drives are quite useful in comparing drives within the respective classes. Directly comparing figures of a SCSI drive to an ATA device is less concrete. Do so with caution.

For this test, we used an ABIT LX6 motherboard (v1.1, bios v C7Q), a 266 MHz Intel Pentium II processor, a 64MB 10ns SDRAM DIMM, and a Matrox Millennium II PCI 4MB (bios v1.2, PowerDesk v3.80). The SCSI controller was the Adaptec AHA-2940U2W. Read caching and write caching were enabled on all tested drives. The boot drive (Seagate Hawk 4XL ST34555) contained Windows 95 OSR 2.1 patched with Intel's 82371xB INF Update. The tests were run at 1024x768 with 24 bit color at 85 Hz using small fonts. Here is a link to the test bed information. All drives were formatted with Fat 32 to the largest possible drive size. ZDBop's Startup Manager was used to prevent loading of background applications. ZDBop's WinBench 98 v1.0's Disk Test Suites were run on all test drives. These test results represent the average of five runs.

IBM Drives
Comments Locked


View All Comments

Log in

Don't have an account? Sign up now