Samsung Spinpoint F3 1Tb random access times

[Kisakuku]

Hi,

I have two brand new Samsung Spinpoint F3 1Tb drives (HD103SJ), one of which I need to return. They post the same sequential speeds in HD Tune, but seem to have slightly different random access times.

Drive #1 - system drive, indexing disabled, all non-essential processes killed, soft-mounted on NoiseMagic NoVibes III (http://www.frozencpu.com/products/3038/) horizontally with PCB facing up.

Transfer Rate Minimum : 74.6 MB/sec
Transfer Rate Maximum : 139.5 MB/sec
Transfer Rate Average : 114.1 MB/sec
Access Time : 13.8 ms
Burst Rate : 153.3 MB/sec

Drive #2 - empty drive, laid out horizontally on a hard surface outside the case with PCB also facing up.

Transfer Rate Minimum : 72.2 MB/sec
Transfer Rate Maximum : 139.4 MB/sec
Transfer Rate Average : 114.4 MB/sec
Access Time : 13.4 ms
Burst Rate : 151.9 MB/sec

I've run HD Tune at least 5 times on each drive from within Windows 7 loaded from the first drive, with the benchmark slider all the way to "Accurate". The results have essentially zero variability. System drive is always at 13.7-13.8ms random access time, the empty drive is always at 13.4ms. Both drives are brand new and show no errors / bad sectors / etc in HD Tune / ActiveSMART.

Are both random access time within the normal range for these drives?

Is the difference between them a function of
1) one drive being soft-mounted and the other being on a hard surface OR
2) one drive being a system drive and the other being unused OR
3) just natural variation between drives?

Is it worth my time to clone drive #1 to drive #2 and return drive #1? Obviously, as a control I should do that anyway, then load the OS from drive #2 and repeat the testing, but I'm trying to avoid figuring out how to clone the 100Mb unmarked system partition that Windows 7 creates. Should I not bother and just return drive #2? I'd appreciate any opinions on the matter.

[ame]

both drives seem to be within tolerance range of the same specs.

I don't think any of the two is defective. If anything one of them is (ever so) slightly better than it should be.

http://www.tomshardware.com/reviews/hit ... 017-5.html

[Kisakuku]

http://www.tomshardware.com/reviews/hit ... 017-5.html

I'm only seeing F1 Spinpoints in that comparison, no F3's.

[mkk]

A variance as small as 3% is no cause for alarm and the cause could be anything. I've seen larger differences between drives of identical models.

[ame]

http://www.tomshardware.com/reviews/hit ... 017-5.html

I'm only seeing F1 Spinpoints in that comparison, no F3's.

Google is your freind

[Kisakuku]

A variance as small as 3% is no cause for alarm and the cause could be anything. I've seen larger differences between drives of identical models.

Thanks for the reassurance. I guess I'll chalk it up to drive variance within spec and return the 2nd 3%-faster one. I'm just waiting for the new SSDs to come out, so the first drive will end up being a mostly media drive soon anyway.

Now if somebody could help me figure out what to do about the high-pitched whine of my AX850 PSU when the GPU is under load....

[ame]

About the PSU - you will get a better answer if you post a new thread in The proper section