DFI LT X48-T2R
ICH9R southbridge, running in AHCI mode
Nothing else really matters...
7200.11 No AAM
SE16 Default AAM (128, slowest)
I was not able to change my AAM settings, as I did not want burn the feature tool, mount an IDE DVD drive, reboot and change my settings from AHCI, use the feature tool to change AAM, then boot back into Windows for further testing. It didn't matter though, as you will see from the results:
Everest Disk Benchmark
Shockingly, the SE16 destroys the 7200.11 at buffered reads. Guess the size of the cache is not all that matters
The older 3-platter 7200.11 750GB model has the best access times, so there seems to be a hardware or firmware issue with the 2-platter 640GB model from Seagate. Of the two 640 drives, the 7200.11 leads with access times, but that was to be expected. Turning off automatic acoustic management (AAM) on the SE16 will drop its access dramatically, to approximately 12-13ms, enough to take the lead over the 750GB drive.
While CPU usage is marginal across the board, the SE16 does indeed consume a little more than expected. This may, however, be partially caused by AAM.
HD Tach 126.96.36.199
What we're seeing here is probably a little more accurate. The burst speed advantage of the 7200.11 might be because of AAM on the SE16, or simply because of its 32MB cache. Can't say, but regardless, burst speed is fairly unimportant with such a small margin of difference, particularly because of the repeatedly-mentioned AAM stuff.
Read speeds seem to be roughly even across both drives at first, but as we start getting further into the test the SE16 starts to take the lead. As you can see from the final test results:
SE16 Min. Speed: ~40MB/s (dipped 3x)
7200.11 Min. Speed: ~30MB/s (dipped 5.5x)
SE16 Avg. Speed: 99.1MB/s
7200.11 Avg. Speed: 96.8MB/s
SE16 Access Time: 16.6ms (with AAM)
7200.11 Access Time: 15.3ms
SE16 Burst: 237.4MB/s
7200.11 Burst: 250.1MB/s
Okay, there is definitely some weirdness going on here. Purple represents the 7200.11, grey (the very stable line at the bottom) is the SE16. When it comes to writing, there is something REALLY wrong with the 7200.11. It averages with ~35ms write access time, approximately three times slower than the SE16!
Linear writes show the 7200.11 being faster overall, quite a bit faster when it comes to the minimum write speed it attained. This should be because of the additional 16MB cache it has.
Acoustics / Temperature
I do not have anything to with which I can definitively test acoustics here, but I do have my ears. My system fans drown out both hard drives. Putting my ears to each drive at idle and under load, the SE16 is clearly quieter (yeah, I know, what a shocker). It sounds like a whisper, just air flowing through the drive, whereas the 7200.11 sounds more "powered", if you will. Seek sounds were fairly quiet on the 7200.11, but I wonder what the SE16's seeks would be like once I turn it off.
Unfortunately, I can't test with AAM off (explained above), so this may be slightly skewed. I can't say whether or not the SE16 will be quieter or louder than the 7200.11 with AAM off.
In regards to temperatures, the 7200.11 is considerably cooler. Running them outside of my case, the 7200.11 ran a full 2-3 degrees cooler than the SE16 (39-42 vs. 41-43; ambient is approx. 30 degrees).
I'll keep this brief. The 7200.11 writes faster than the SE16, this should be due to its additional cache. However, there was that odd write access time test, not sure why it was so weird. Didn't seem to have an effect on the write speed itself, so I wonder if it was just a fluke.
The SE16 reads faster than the 7200.11 if you turn off AAM (or even reduce it). Access times are pretty abysmal on both sides, but at least the SE16 can be improved with just AAM settings.
It's pretty obvious that the drive to get is the SE16. Compared to the 7200.11, it has AAM (so, inherently it is quieter), vastly superior access times with AAM off, read times are just as good if not better, and write speeds are only marginally slower.
I wish I had gotten two of each of these drives for RAID testing. Sorry, fellas.