Testbed and Methods
For our today’s test session we used the following software:
- WinBench 99 2.0;
- FC-Test 1.0 build 13;
- PCMark04 Build 1.3.0;
- Intel IOMeter 2003.02.15.
We assembled the following testbed:
- Albatron PX865PE Pro mainboard;
- Intel Pentium 4 2.4GHz CPU;
- IBM DTLA-307015 15GB system HDD;
- Radeon 7000 32MB graphics card;
- 256MB RAM;
- Microsoft Windows XP SP2.
The tests were performed with the promise SATA150 TX2 Plus controller card. This way we could test ATA and SATA drives on equal terms, because this controller supports both interfaces.
The drives were formatted for FAT32 and NTFS file systems as a single partition with the cluster of default size. In some cases, which will be specifically mentioned later in this article, we tested in 32GB logical partitions, also formatted for FAT32 and NTFS systems with the default cluster.
Since the last review of a 2.5-inch hard disk drive the testing methodology has been updated. First, we started running the tests in Windows XP OS :)
Second, we removed some Intel IOMeter tests from the list of approved benchmarks (such as FileServer and WebServer patterns). Third, we included PCMark04 into the list of our benchmarks. I am sure your immediate question would be: why not PCMark05? Let me explain: PCMark05 hasn’t yet reached patch 1.20, and if we use it without this patch, we will not be able to compare the results obtained in it with the previously obtained results from our database (retesting all the drives will take too much time). Unfortunately, the patches do not allow direct comparison of the results if they have been obtained in different versions of the test.
All PCMark04 tests were run 10 times and the average result was taken for further analysis.