OCZ Vertex 2E crystal mark results

You would need to bench it outside the OS you are using, as this will skew the results.
The OS will still be accessing the drive when benching.

I have two identical drives (not in raid0) so I can actually do this, assuming i've understood you correctly - bench mark the second disk whilst i'm booted off the OS installed on the first?

Is it likely to make the sort of difference i'm seeing between those review benches and my own results? I'm not usually obsessive or anal about new hardware - benchmarks performed in reviews are often in idealised circumstances. However, 50 meg a second seems like a rather significant difference. It's more than have as much again.
 
Update:

OK, i've done a bit of reading on the ocz forums, and an explanation for the lack of coherence between my own benchmarking results and those found in on-line reviews has been forthcoming. Basically, it's a direct consequence of the shift to 25nm process for the NAND chips.

A bit of a stink, to put it mildly, has been created about this on-line. I think the story emerged a couple of days ago and has attracted a fair amount of criticism to OCZ, not least for leaving the model number and branding completely unchanged, despite the change to the internals and resulting reduction in benchmark performance. A lot of people are ****ed at having paid the same price a few months later for a 25nm drive with far worse performance.

So i'm in two minds about what to do now. Now that the C300 firmware has been fixed to improve trim support, that drive pretty much blows the ocz right out of the water in most benchmark results, apart from IOPs (but what difference do they make if the practical benchmarks like data rates and random access times are better?). I'm tempted to try and send these drives back and order a replacement c300 for almost exactly the same price (slightly more). The drives have obviously been very lightly used though, so this isn't a straightforward swap. I didn't order them from OCUK, because i'm in the same town as another big computer hardware retailer and wanted to drive down and pick them up the same day. The retailer in question has the drives advertised as winning a ***** performance award (which of course, pertains to the old 34nm drives, not the new 25nm ones) so i'm thinking this is the route to take if they get awkward.

Anyway, the mystery is solved. I'm not dissatisfied with the performance in a day-to-day context, but i feel i should have paid substantially less for it. That's really the whole point about 25nm in SSDs - cost reduction. It doesn't offer the same performance benefits as in gfx for example, simply because the application is totally different. The 25nm NAND also have significantly reduced cycle life times, to make matters worse.
 
Oh and benchmarks after installing latest intel RST drivers:

jz8wfd.jpg
 
Back
Top Bottom