So, here's the latest update after several hours of testing 8 different configurations:
1) Standard , metal to metal mount
3) Standard mount using EAR Grommets
5) NoVibes3 in 5 1/4" drive bay
7) Modified shock cord suspension.
2,4,6,8.) Repeat above, but with AAM enabled.
My original goal was to get a good handle on performance, sound, as well as temperature. However, in order to be satisifed with temperature analysis, I'm going to have to spend more time, than just a 10-15 min test. So I can give you a "general" look at temperature at this time, but that's about it.
I modified the shock cord suspension to be a bit more standard:
It's much more "stable" this way, but the main difference is that it provides room for the front case fan.
The benchmark program I decided to use is called h2benchw (zip file is
here), which I found by looking at
StorageReview's FAQ. It's described as:
StorageReview wrote:Probably the best freely available hard drive benchmark is German c't magazine's h2benchw, which they use themselves when reviewing hard drives. It measures the interface and physical transfer rates, access time, includes some application patterns and can do data integrity testing.
Now, for the most accurate testing, this requires an empty, unpartitioned disk to do
write testing (And I will be doing this perhaps later tonight with a couple configurations.) Since my first round of tests is really for noise analysis, I decided to skip the writing aspect for now.
This is a command line test, and I ran it with these parameters:
2benchw.exe 0 -english -c 0 -s -p -w test
This runs 3 of 4 groups of tests. I omitted one test because that particular one (sustained read transfer) took an additional 18 minutes by itself, and it didn't make any "noise". (So who cares?
)
On my system each test run (all three tests) takes about 10 minutes.
During the test, I could generally hear two distinct types of "seek" noise. One was higher pitched, and one was lower pitched. At one particular point in the test, the combination of seek noises was particularly loud. If you want to try this yourself, that portion of the test is when the screen reads:
"Measuring Access Time (whole disk)
reading..."
This occurs about 30 seconds into the test, and it lasts for about 30 seconds. The first 15 seconds of this test is the loudest part. (I'll henceforth refer to this as the "f'n loud part"
). Then there's about 15 seconds or "normal seek" noises before the screen updates for the next test.
General noise results:
Up front, I'd like to say there is no appreciable "idle whine" from the raptor. It's there if you put your ear up next to it, but inaudible over the case fans.
1)
Direct metal-to-metal mounting, AAM disabled. There is a muted hum eminating from the case, and you can feel some vibration in certain places of the case panels. It's a faint hum, but it's there. Seek noise is in general sharp and pronounced. It's not hollow and there's no residual resonance, just sharp, pronounced seek noise. The f'n lound part is....
F'N LOUD! Sounds like a truck rumbling through the case. Low frequencey "thuds" combined with the normal seek noises.
I'd hate to hear the raptor direct mounted to an aluminum case without padding...must sound like absolute hell.
2)
Direct metal-to-metal mounting, AAM Enabled. Doesn't seem to make a difference. Sounds just as loud to me.
3)
Direct mounting with Ear Grommets, AAM Disabled The grommets succeeded in eliminating the hum, and I can only feel a "faint" case vibration on certain parts of the side panels at idle. Normal seek noises are subdued quite a bit...but while the f'n loud test isn't quite as obnoxious, as metal to metal, but it's still what I would consider "real loud".
4)
Direct mounting with Ear Grommets, AAM Enabled No major difference from number three.
5)
NoVibes3 suspension in 5 1/4" bay, AAM Disabled.
Up to this point, I was feeling quite uneasy about my Raptor purchase. That "f'n loud" seeking section was very irritating. Of course, one would expect improvement with a suspension, but I was pretty floored by the order of magnitude difference the NoVibes suspension made with the "f'n loud" seek. It was practically gone. I don't mean inaudible..but now it sounded almost no different than the "normal seeks", which have also been significantly reduced in volmume and sharpness. Seeks in general are still clearly audible ouside the case, but they are muted and dull. I have to have my ear within inches of the case in order to tell the difference between the "f'n loud" seek, and the normal seeks.
This is the first time I said "now, THAT'S acceptable!"
6)
NoVibes3 suspension in 5 1/4" bay, AAM Enabled. To boot, this is the first time I believed I heard an additional improvement in sound with AAM enabled.
7)
Shock Cord suspension underneath drive cage, AAM Disabled. There is now absolutley no difference bewteen the "f'n loud" and "normal" seeks. The normal seeks sound the same or might be a touch louder compared to the NoVibes suspension. The fact that the drive is closer to the bezel opening in the case is likely causing a little more noise to escape.
8.)
Shock Cord suspension underneath drive cage, AAM Enabled. Again, I believe I can hear an improvement in the seeks with AAM Enabled. Not drastic, but an improvement.
General Performance. In general, I'd say enabling AAM decreasing performance about 3%. The "App index" scores I was getting with AAM Enabled were in the high 23's. The App Index scores with AAM disabled were in the low 24s.
I didn't see any significant performance difference across different mounting methods. For a more definitive look at performance, I will run these tests in write enabled mode. (I'll have to reinstall WinXP on another hard drive, and blank out the Raptor.)
General Temperature Can't really comment too much, since I really haven't run one configuration long enough to get stable temp readings. As expected, temps were generally higher when suspended (and highest in the 5 1/4" bay). I'll get better temp detail when I do the performance testing.
I also hope to get some WAV files recorded for you, so you can get an idea of the differences I'm talking about.
Morals of the story:
* I don't think anyone here would be happy with the Raptor in any type of direct mounting situation.
* If you are willing to suspend the drive in some way, the Raptor becomes "very acceptable" IMO. I am satisfied with the current noise levels, without resorting to a drive enclosure.
* If you want close to absolute silence, you won't get it from the Raptor...with the possible exception of a drive enclosure.
As a final note, keep in mind that the ARM Stealth case is steel construction, has acoustipac lining, and generally good airflow. You milage may vary with other cases....
).
