
Slackmaster2K
Gone
About a year ago I posted some numbers on ATA(UDMA)33 vs. ATA66. At that time, with my 7200RPM Maxtor drive, I found absolutely no benefits to ATA66. I was seeing sustained transfer rates in the 27+MB/sec range.
With my 5400RPM Quantum drive, I actually noted a decrease in performance on my ATA/66 controller (a Highpoint HPT366) and was seeing performance under 19MB/sec. A few others on the BBS noted similar results.
Well, I just did a quick test with a brand new IBM 30GB Deskstar 7200RPM drive on an Abit BE6-II motherboard, and found a substantial increase in speed when using the Highpoint ATA/66 controller vs. the standard UDMA/33 controller. The difference was a leap from 27MB/sec to 34MB/sec. In multitrack tests using Dskbench, I saw an increase of over 1MB per second with typical multitracking block sizes. This relates to several more tracks (assuming a hard disk bottleneck).
I'm not sure why exactly I'm seeing these changes, but the 34MB/sec turned in by the IBM drive on the Highpoint would lead me to believe that drives are simply getting faster, which is of course to be expected. With this increase in speed, perhaps the higher peak transfer rates of ATA/66 do help with sustained transfers.
I would be interested if anyone would be so kind as to run some experiments with a newer (literally newer) 7200RPM hard drive and a couple controllers.
Slackmaster 2000
With my 5400RPM Quantum drive, I actually noted a decrease in performance on my ATA/66 controller (a Highpoint HPT366) and was seeing performance under 19MB/sec. A few others on the BBS noted similar results.
Well, I just did a quick test with a brand new IBM 30GB Deskstar 7200RPM drive on an Abit BE6-II motherboard, and found a substantial increase in speed when using the Highpoint ATA/66 controller vs. the standard UDMA/33 controller. The difference was a leap from 27MB/sec to 34MB/sec. In multitrack tests using Dskbench, I saw an increase of over 1MB per second with typical multitracking block sizes. This relates to several more tracks (assuming a hard disk bottleneck).
I'm not sure why exactly I'm seeing these changes, but the 34MB/sec turned in by the IBM drive on the Highpoint would lead me to believe that drives are simply getting faster, which is of course to be expected. With this increase in speed, perhaps the higher peak transfer rates of ATA/66 do help with sustained transfers.
I would be interested if anyone would be so kind as to run some experiments with a newer (literally newer) 7200RPM hard drive and a couple controllers.
Slackmaster 2000