3 ms·
I didn't include this in my post but here is the deal: the first time I ran the raw read / write test the results were pretty close to the manufacturer's claim.
by cowmixtoo 16y ago
I didn't include this in my post but here is the deal: the first time I ran the raw read / write test the results were pretty close to the manufacturer's claim. On the 2nd try (and the next 20 after) i got EXACTLY what is presented in the post.
Weird.
- AngryParsley 16y agoDoing a sequential write to the entire disk probably made the SSD think the disk was full of data. Unless you ran the ATA secure erase command after each benchmark or turned on TRIM, that will hurt performance. http://www.anandtech.com/print/2738 http://www.anandtech.com/print/2738 explains why, around page 6.
- cowmixtoo 16y agoThe other (or 'fast) SSD did not have this issue though. I re-ran the raw read / write test at least 15 times over two days and I got very consistent (and very good) results.
- AngryParsley 16y agoThe second SSD probably has better* firmware or flash or both. It's basically maxing out the SATA II bus no matter what you throw at it. * By "better" I mean it performs better in the benchmarks you used. A firmware that wins benchmarks often doesn't do correspondingly well in the real world. For example, Intel's SSDs are sub-par in sequential read/write benchmarks, but they perform much better in most application benchmarks.
- Andys 16y agoThe manufacturer's claim probably only applies to a fresh unused SSD.