PCPer File Copy Test
Our custom PCPer-FC test does some fairly simple file creation and copy routines in order to test the storage system for speed. The script creates a set of files of varying sizes, times the creation process, then copies the same files to another partition on the same hard drive and times the copy process as well. There are four file patterns that we used to try and find any strong or weak points in the hardware: 10 files @ 1000 MB each, 100 files @ 100 MB each, 500 files @ 10 MB each and 1000 files at 1 MB each.
Octane busted out the gate and promptly smacked down everyone else on the field, including even the best offerings from SandForce.
Octane did not fare as well in batch copying of larger files. Since this test copies the files to/from the drive, it forces the controller to simultaneously read and write. For drives with larger caches (like the Octane), more juggling must be done as the controller occasionally purges the RAM cache out to flash memory. Even with a large cache, performance here boils down to how that particular controller was tuned, so there is room for improvement with a bit of tweaking to those configs.
Great review and an
Great review and an interesting product. I am finally tempted to put an SSD in my Laptop. To bad about the compatibility issue. There is no way I would do a fresh install.
also, on the pc per File copy test page, you haev a small typo :
”Even with a laege cache”
Thanks, fixed!
Thanks, fixed!
When making screenshot
When making screenshot :
Please do not using JPEG , use PNG
It looks a lot better with PNG.
Took me about 1/2 sec to spot the JPEG artifacts …
delete this lame^^
delete this lame^^
blocked and cleaning now
blocked and cleaning now
It is still made by OCZ and
It is still made by OCZ and that alone is reason enough to go no where near this product.
Times are changing… keep
Times are changing… keep your eye on them.
My vertex recently failed and
My vertex recently failed and it took over a month to get a replacement from them. In the mean time, I had to buy another ssd for my system. I got the intel 320 because I was more interested in reliability than having the fastest drive out there. Really two issues come up here: one is that the vertex failed; and two is that it took over a month to get a replacement. I too will stay clear of OCZ for the foreseeable future. I just hope my vertex 3 max io edition that I bought previous to my vertex debacle hangs in there.
Your perception is based on a
Your perception is based on a first gen product. Look at the reviews of the Vertex 3 (or all SF-based drives) and you’ll see that they’ve become a lot more reliable. I’ve had a Vertex 3 since launch and never had a single issue.
Now at OCZ is developing their own products (own controller), they’ll be able to respond much quicker to compatibility issues. Keep in mind they’re a relatively small company. As they grow their support and subsequent reliability will improve.
Allyn,
Can you say anything
Allyn,
Can you say anything as to the effectiveness of trim on the drive?
It’s too bad that OCZ handed
It’s too bad that OCZ handed out all 512GB drives, considering the price, that’s probably more than what the majority of buyers are going to consider as viable option. I’d love to see a review of something smaller since the specs are worse, I’m not sure it would keep up to any SF-2281’s or other competitors.
Rumor has it OCZ is being
Rumor has it OCZ is being qualified at a major OEM who wants to put SSD’s in every laptop across their product line, either all SSD or as part of a hybrid system. This is part a response to the recent HDD shortage. I’d assume that’s where the 128GB and 256GB drives are being shipped at the moment.
I just read a relevant review
I just read a relevant review on tom’s hardware. Honestly, based on my knowledge of SSDs, I think their test approach is more complete, especially if your are evaluating a new controller.
The worst case scenario and steady performance they pointed out is orthogonal to what you’ve tested here (sequential/random I/O). I don’t see many other websites doing the same thing, and it would be great if you can try similar methodology if possible. Thanks.
First – I am the original
First – I am the original source for review sites conducting this type of testing:
https://pcper.com/reviews/Storage/Long-term-performance-analysis-Intel-Mainstream-SSDs (dated Feb of *2009*).
The part of the Tom’s article you speak of:
http://www.tomshardware.com/reviews/octane-sata-indilinx-benchmark-performance,3081-8.html
…is a dated method that is no longer relevant when using an SSD with an OS that continually issues TRIM commands to the drive. Benchmarks can not accurately test for this as they are unable to issue TRIM commands directly to the drive while doing all of those random writes. The closest you can get is to run something that fragments the drive, but then to partition and format the drive under Windows 7, *then* run the HDTach pass and see what happened. I do this to all drives as part of my testing, and the Octane behaved as all other modern SSDs do – performance had returned to normal. This was actually noted below the basing portion on that page of the Tom’s piece.
That said, I’ll revive my “Performance Over Time And TRIM” page for future pieces.
you will see that they’ve
you will see that they’ve become a lot more reliable. I’ve had a Vertex 3 since launch and never had a single issue.
It’s too bad that OCZ handed
It’s too bad that OCZ handed out all 512GB drives, considering the price, that’s probably more than what the majority of buyers are going to consider as viable option. I’d love to see a review of something smaller since the specs are worse, I’m not sure it would keep up to any SF-2281’s or other competitors.