TRIM Performance

In our Vertex 3 preview I mentioned a bug/performance condition/funnythingthathappens with SF-1200 based drives. If you write incompressible data to all LBAs on the drive (e.g. fill the drive up with H.264 videos) and fill the spare area with incompressible data (do it again without TRIMing the drive) you'll actually put your SF-1200 based SSD into a performance condition that it can't TRIM its way out of. Completely TRIM the drive and you'll notice that while compressible writes are nice and speedy, incompressible writes happen at a max of 70 - 80MB/s. In our Vertex 3 Pro preview I mentioned that it seemed as if SandForce had nearly fixed the issue. The worst I ever recorded performance on the 240GB drive after my aforementioned fill procedure was 198MB/s - a pretty healthy level.

The 120GB drive doesn't mask the drop nearly as well. The same process I described above drops performance to the 100 - 130MB/s range. This is better than what we saw with the Vertex 2, but still a valid concern if you plan on storing/manipulating a lot of highly compressed data (e.g. H.264 video) on your SSD.

The other major change since the preview? The 120GB drive can definitely get into a pretty fragmented state (again only if you pepper it with incompressible data). I filled the drive with incompressible data, ran a 4KB (100% LBA space, QD32) random write test with incompressible data for 20 minutes, and then ran AS-SSD (another incompressible data test) to see how low performance could get:

OCZ Vertex 3 120GB - Resiliency - AS SSD Sequential Write Speed - 6Gbps
  Clean After Torture After TRIM
OCZ Vertex 3 120GB 162.1 MB/s 38.3 MB/s 101.5 MB/s

Note that the Vertex 3 does recover pretty well after you write to it sequentially. A second AS-SSD pass shot performance up to 132MB/s. As I mentioned above, after TRIMing the whole drive I saw performance in the 100 - 130MB/s range.

This is truly the worst case scenario for any SF based drive. Unless you deal in a lot of truly random data or plan on storing/manipulating a lot of highly compressed files (e.g. compressed JPEGs, H.264 videos, etc...), I wouldn't be too concerned about this worst-case scenario performance. What does bother me however is how much lower the 120GB drive's worst case is vs. the 240GB.

Power Consumption

Unusually high idle power consumption was a bug in the early Vertex 3 firmware - that seems to have been fixed with the latest firmware revision. Overall power consumption seems pretty good for the 120GB drive, it's in line with other current generation SSDs we've seen although we admittedly haven't tested many similar capacity drives this year yet.

Idle Power - Idle at Desktop

Load Power - 128KB Sequential Write

Load Power - 4KB Random Write, QD=32

AnandTech Storage Bench 2010 Final Words
POST A COMMENT

153 Comments

View All Comments

  • sleepeeg3 - Thursday, April 07, 2011 - link

    Thanks for the info, but IMO the bottom line is if it works reliably and it allows them to deliver something at a lower price, I am all for it. If it backfires on them and they get massive failure rates, consumers will respond by buying someone else's product. That's the beauty of capitalism. Reply
  • sor - Thursday, April 07, 2011 - link

    Oh sure, I agree that the bottom line is whether or not it still works, that's why they do the binning and have grades of product within that brand. If OCZ can use cheaper flash and the controller takes care of the increased failures, or the users never reach the failure threshold, then who cares, as long as the product works the same?

    I can't speak for the testing procedures within SpekTek or their tolerances, as I only worked for a facility that tested parts for Micron, and in the process generated the bad parts and did some of the binning before sending them to SpekTek. Much of the stuff that went to them failed our tests but was otherwise not physically damaged.

    There's a reason why those parts are sold under the SpecTek brand at a discount, it shows that even the manufacturer doesn't trust them to be sold under the good brand after testing.
    Reply
  • Panlion - Thursday, April 07, 2011 - link

    I wonder if OCZ will produce a 7mm 2.5 inch drive. The newer notebooks from Lenovo are starting to demand that format it'll be nice if I can have some option other than Intel SSD. Reply
  • sleepeeg3 - Thursday, April 07, 2011 - link

    Maintaining integrity while sticking out for the little guy, instead of bending over backward to write glowing articles for every vendor sponsor. That's what has made this site succeed.

    I wish you could also take OCZ to task on SandForce's controller strange tendency to lock up and vanish from a system, due to built in encryption. They are in complete denial that it is an issue, despite dozens of reports on their user forums.
    Reply
  • edfcmc - Thursday, April 07, 2011 - link

    Thank you Anand for this very informative and in-depth review of the OCZ issue and their latest 120gb vertex 3 product; especially since the 120gb products are within my price range and the size I am looking to purchase. On a side note, I have been reading your reviews since your review of the FIC PA-2007 many years ago and I love the evolution of this site and your dedication to keeping us consumers informed.

    p.s. Please consider asking asus/Nvidia to update the Nvidia driver on their ULV80 series as nothing new has been updated since I purchased the UL80vt based on this site's recommendation. Asus/Nvida seem to be a little non-responsive to us folks who have been requesting an update for quite some time.
    Reply
  • ekerazha - Thursday, April 07, 2011 - link

    Can't wait for reviews of SSDs (Intel G3, Crucial m4) with comparable size (120 GB). Reply
  • Chloiber - Thursday, April 07, 2011 - link

    Anand:
    Just a quick note. In the newest SF-firmware, there is also still a bug with Hynid Flash. You can see it here, under "Open Issues":
    http://www.ocztechnologyforum.de/forum/showthread....

    "Under benchmarking scenarios with IOMETER 2006, 60GB drives that use Hynix32nm MLC (1024 blocks, 8KB pages) can impose long latencies"

    Just FYI.
    Reply
  • MarcHFR - Thursday, April 07, 2011 - link

    Dear OCZ, Dear Anand,

    In the past, it was simple :

    Vertex : always the same NAND
    Agility : NAND could change

    I know that Vertex name is a best seller for OCZ, but i think it will be simplier to back to this
    Reply
  • strikeback03 - Friday, April 08, 2011 - link

    That is what I was wondering, I thought the point of the Agility line was that they would use the good controller but possibly cheaper NAND. Reply
  • Adul - Thursday, April 07, 2011 - link

    Why not make use of QOR codes so a shopper can just scan the code to be taken to a page with more detail information. Reply

Log in

Don't have an account? Sign up now