Usually when i test a hard drive, I put the drive into another computer already running Windows and then benchmark the new drive. This way I get the most consistent results without the drive doing anything.
Those with big spikes and dropouts are likely running their OS on the very drive they are testing with...thus the OS is doing things actively on the drive in the background and these things impact the testing.
Btw, there is faster than SSd. Considerably faster.
I splurged on yet another drive--identical 64GB SNV425-S2.
What's the deal with the rollercoaster ride in the HD Tune Bench?
Read speeds leave a lot to be desired...
Something must have been running in the system producing the rollercoster. What is more worrysome is the access time of 0.2ms which is double a single disk which is usually 0.1ms.
I don't think it buys much if you stripe the SSDs - especially if you use it only for the OS. The better data transfer rate does not really help because the system does not read large chunks of data, but the access time seems to suffer. And the access time is the determining factor.
One might think that the data transfer rate helps in such operations as imaging the whole OS off the striped SSDs. But then the receiving disk cannot keep up with it. So you can't win.
Performance-wise, the system is as fast and snappy as ever.
A bench is just a bench, so I'm not overly worried about what the results show. I am satisfied with system performance--so in the end, I guess that is all that matters.
Well, if you are happy, that's all what counts.
What AHCI/RAid driver are you using?
I just tried the new Intel Drivers for mine, and while it was passing TRIM which is good, my system just wasnt to happy with them.
For some reason it likes the generic MS drivers better.
Might be worth trying another, if that option is available. Just a thought.
Not that it really matters :) Those are good numbers as is.
AMD RAID driver supplied by Gigabyte
GA-MA790X-UD4P (rev. 1.0) - GIGABYTE - Support&Download - Motherboard - Driver
