Testing (continued):
HD Tune 2.53:
HD Tune 2.53 is another benchmark that tests drive performance, and it is very similar to HD Tach. I have just recently started using it, and to date it has pretty much echoed the findings of HD Tach. The graph below details the results for access time, CPU usage, average transfer rate, and burst rate, where lower values are better for the first two categories and higher values are better for the other two categories.
Here we see a major drop in the reported performance of the FreeAgent Pro while connected via eSATA. While every other drive configuration performed about as well as reported by HD Tach, the Average Transfer Rate and Burst Rate of the FreeAgent Pro dropped so far that both its USB and Firewire performance figures beat it with ease. The Cooler Master enclosure tested for reference still maintained about the same speeds in this test, so I became suspicious about what might be going on.
I then took the FreeAgent Pro to another test system and ran both HD Tach and HD Tune while connected via eSATA and USB. The results achieved by the second system followed the same trend as seen in the charts above. The eSATA results were just horrible according to HD Tune, while they were not so bad in HD Tach. HD Tune allows you to take a snapshot of the drive's information, so I decided to compare the Maxtor drive found in the Cooler Master enclosure to the FreeAgent Pro, while both were connected via eSATA. Many features that are enabled on the Maxtor drive are not on the Seagate drive (Read Look-Ahead and NCQ), and while the Maxtor drive is listed as ATA-7, the Seagate drive is listed as ATA-6. If the information on these pages is correct, it could explain the performance issues, but it doesn't explain why HD Tach reports much better performance.
 
Being less familiar with HD Tune than I am with HD Tach, I assumed it was something in the software, and moved on to real world testing to set the record straight.
|
|
