Setting up the Test

AnandTech prepared two identical test beds for each individual hard drive, the first was a setup running Windows 98 with a single FAT32 partition occupying the full size of the hard disk. The second setup was one running Windows NT 4.0 Service Pack 4 with a single 4GB NTFS partition. In both cases, the only installed software was the operating system, and the benchmark utilities.

The complete test system configuration was as follows:

  • Intel Pentium II 400 w/ L2 Cache ECC Checking Disabled
  • ABIT BX6 Revision 2.0 Motherboard
  • Promise Ultra66 Ultra ATA/66 PCI Hard Disk Controller
  • Ultra ATA/66 40-pin 80-conductor HDD cable
  • Microsoft Windows 98
  • Microsoft Windows NT 4.0 Service Pack 4
  • Ziff Davis Winbench 99
  • Ziff Davis Winstone 99

The Promise Ultra66 controller card was used in all of the tests, regardless of whether or not the drive being tested was Ultra ATA/66 compliant in order to remove any performance variations between the BX6’s on-board PCI IDE controller and the Ultra66 controller. The Promise Ultra66 driver revision tested was 1.42

Each hard disk was partitioned and cleanly formatted right out of the packaging as to prevent any skewing of the data. For purposes of consistency, each benchmark was run a total of 5 times, and the average then taken from those 5 benchmark trials.

Ziff Davis’ Winbench 99 was used to show the real-world transfer rates achieved by the individual drives, measured in megabytes per second. Ziff Davis’ Winstone 99 was used to show the real world performance increase by using one drive as opposed to another while running Business or High-End applications depending on the test bed. Both tests are designed to represent real world cases including sequential reads/writes to the disk (audio/video editing applications), random data access, as well as CPU utilization and disk access time.

Index Windows 98 Business Performance

Log in

Don't have an account? Sign up now