TBH, I thought my system performance was worse when I did my last install of Vista with this mode enabled in BIOS. I don't run RAID, and I don't have any e-SATA devices, but hey, things can change.
First difference I noticed was a loss of desktop snappiness, esp. in Explorer. In AHCI mode, there seems to be a bit of lag in opening Explorer windows and loading up icons/thumbnails for my files. Actually, a lot of lag loading up those icons/thumbnails.
Running HDTune reveals that Burst Mode performance ranges from somewhat to much worse:
ST3500320AS: 93.7 MB/S AHCI, 146.5 MB/S IDE :Q
ST3500630AS: 125.1 MB/S AHCI, 140.9 MB/S IDE
ST31000333AS: 140.4 MB/S AHCI, 157.8 MB/S IDE
CPU usage in AHCI mode was a little better across the board, but CPU usage in IDE mode never exceeded 3.7%. Other performance benchmarks were essentially identical. I couldn't generate any synthetic write speed benchies (I'm too cheap).
The ST31000333AS drive is new (Seagate firmware FUD = good deals on HDDs for me). I started copying the contents of ST3500630AS to it, but aborted the process after 15 minutes or so after noting an xfer speed that barely exceeded 25 MB/Sec on the copy operation. This was a 200 GB copy, with file sizes ranging from a handful of bytes to DVD sized ISOs and everything in between. Restarting the operation in IDE mode netted a write speed pretty much twice that fast.
Did I configure something incorrectly? Is this yet another case of Newer != Better for folks who run simple configurations?
First difference I noticed was a loss of desktop snappiness, esp. in Explorer. In AHCI mode, there seems to be a bit of lag in opening Explorer windows and loading up icons/thumbnails for my files. Actually, a lot of lag loading up those icons/thumbnails.
Running HDTune reveals that Burst Mode performance ranges from somewhat to much worse:
ST3500320AS: 93.7 MB/S AHCI, 146.5 MB/S IDE :Q
ST3500630AS: 125.1 MB/S AHCI, 140.9 MB/S IDE
ST31000333AS: 140.4 MB/S AHCI, 157.8 MB/S IDE
CPU usage in AHCI mode was a little better across the board, but CPU usage in IDE mode never exceeded 3.7%. Other performance benchmarks were essentially identical. I couldn't generate any synthetic write speed benchies (I'm too cheap).
The ST31000333AS drive is new (Seagate firmware FUD = good deals on HDDs for me). I started copying the contents of ST3500630AS to it, but aborted the process after 15 minutes or so after noting an xfer speed that barely exceeded 25 MB/Sec on the copy operation. This was a 200 GB copy, with file sizes ranging from a handful of bytes to DVD sized ISOs and everything in between. Restarting the operation in IDE mode netted a write speed pretty much twice that fast.
Did I configure something incorrectly? Is this yet another case of Newer != Better for folks who run simple configurations?