OCZ Vertex 3 MAX IOPS & Patriot Wildfire SSDs Reviewed
by Anand Lal Shimpi on June 23, 2011 4:35 AM ESTPatriot's Wildfire
OCZ isn't the only company experimenting with SF-2281 drives and 32nm NAND. Patriot sent me the Wildfire, its first SF-2281 SSD. After a quick run through some performance tests I grew suspicious as it performed a lot like the Vertex 3 MAX IOPS I just tested. Cracking open the case I found the reason:
The Wildfire also uses 32nm Toshiba NAND. The 120GB drive Patriot sent has a 16 chip, 2 die per package configuration - compared to 8 chips with 4 die per package in the Vertex 3 MI. I don't see an advantage for one approach vs. the other. Patriot is targeting a $299 MSRP for the Wildfire, which would put it lower than the Vertex 3 MAX IOPS - although anything is possible once the drive goes on sale on the 28th of this month.
Patriot confirmed that the Wildfire would only use 32nm Toshiba NAND and that any NAND vendor changes would take place in other lines or subsets of the Wildfire brand.
Patriot uses a different PCB layout from OCZ. I don't know that there's an advantage to either layout, they are just different.
The unit I have here shipped with SF firmware revision 3.19, which is equivalent to OCZ's revision 2.08 firmware as far as I know. No word on when we'll see a 2.09 equivalent.
Performance of the Wildfire (as you'll see from the tests that follow) is identical to the Vertex 3 MAX IOPS. To keep the charts more manageable I've only included 6Gbps results from the Wildfire in most areas. Performance on a 3Gbps interface is identical to a 3Gbps Vertex 3 MAX IOPS as you'd expect.
The Test
| CPU | Intel Core i7 965 running at 3.2GHz (Turbo & EIST Disabled) Intel Core i7 2600K running at 3.4GHz (Turbo & EIST Disabled) - for AT SB 2011, AS SSD & ATTO |
| Motherboard: | Intel DX58SO (Intel X58) Intel H67 Motherboard |
| Chipset: | Intel X58 + Marvell SATA 6Gbps PCIe Intel H67 |
| Chipset Drivers: | Intel 9.1.1.1015 + Intel IMSM 8.9 Intel 9.1.1.1015 + Intel RST 10.2 |
| Memory: | Qimonda DDR3-1333 4 x 1GB (7-7-7-20) |
| Video Card: | eVGA GeForce GTX 285 |
| Video Drivers: | NVIDIA ForceWare 190.38 64-bit |
| Desktop Resolution: | 1920 x 1200 |
| OS: | Windows 7 x64 |
Facebook
Twitter
RSS
112 Comments
View All Comments
alpha754293 - Friday, June 24, 2011 - link
Can you run h2benchw on the drives and post the results? Thanks.doylecc - Friday, June 24, 2011 - link
Anand,What write amplification did you get during your use of the SSD 510?
Thanks, and good review.
bse8128 - Friday, June 24, 2011 - link
I was wondering for a moment how (128GB-120GB)/128GB can be 13%, but then I noticed that it's really 120GB but 128 GiB. It's a bit confusing to call both 10^9 and 2^30 just "GB".jwilliams4200 - Friday, June 24, 2011 - link
Yes, I really wish Anand would keep his GB and GiB units straight. It makes his articles very difficult to follow sometimes.Marian666 - Saturday, June 25, 2011 - link
Who the hell asks for qd32?? Like there werent enough of such tests on internet, and anand was like the only one giving us qd3 4k read test....And whats with "depth of 32 instead of 3" ?? How hard it is to test drives in both queue depths
ARGH!!!!
/ragequit
MamiyaOtaru - Saturday, June 25, 2011 - link
this site used to be like my bible for SSDs. This continued pushing of OCZ in spite of manifest and multiple failures in their drives has soured me on the whole thing a bit.That aside, I went with Intel for a Macbook (with Snow Leopard) that had seen a couple hard drives die. It doesn't seem faster at all. I'm not willing to trade reliability for a few more percentage points, so some other drive is not an option. And if an SSD can't improve on the performance offered by a laptop drive I can't imagine what motivation I'd have to put one in my desktop.
somedude1234 - Saturday, June 25, 2011 - link
I had the exact opposite experience. I replaced a 7.2K laptop HDD with an Intel 80GB G2 SSD in my Dell D810 (running XP at the time) and have since migrated that same SSD to a Dell E6400 running Win7. The difference in overall system performance after moving to the SSD was absolutely clear in both XP and Win7, across both laptops.Granted, you're working in a Mac environment, but I will never again willingly deal with a workstation that isn't running an SSD for the OS drive.
I'm currently running my G2 with less than 5GB free, so it feels a bit slower than it did when there was > 20 GB free, but it's still night and day vs any HDD.
The system is used every day for productivity apps (primarily outlook/word/excel) as well as SAP client, putty, remote desktop, etc.
Movieman420 - Saturday, June 25, 2011 - link
Over the last few days, there has been a spark that has brought on a 'meeting of the minds' in this thread:http://www.ocztechnologyforum.com/forum/showthread...
Be warned, this is a deeply technical discussion...I only thought I was up to speed on SSDs...lol
cactusdog - Sunday, June 26, 2011 - link
That is just another theory in a long line of theories. It doesnt explain why people have issues on other boards without IME.OCZ have tried to blame everything from sata cables to install methods.
If its only 1% with issues i dont know why OCZ are putting so much effort into it. It would be better for them to just give those 1% a refund to move to another drive.
As it stands, the OCZ forum and staff is preoccupied with this issue that "only affects 1%". It looks much worse than that and no doubt some people will be put off by all the discussion about BSOD's
If it is only 1% with issues, OCZ are handling the situation badly.
mcg75 - Sunday, June 26, 2011 - link
I was getting the bsod so I was watching their forums waiting for the result. OCZ said system was setup wrong. Then there are issues with secure erase in parted magic not working properly. So I went through the hassle of doing it all over again according to Tony's guide with no rst loaded. Still got bsod. Now IME is corrupting cmos. Told Tony that IME wasn't loaded when I got bsod. He only replied that I didn't follow his guide by not installing IME. Later in the thread in response to a post, Tony said we could run without IME using MS ahci which is exactly what I was doing.I've been setting up win7 the exact same way for years. First on a X-25m and no issues. Next a C300 with no issues. Now I setup the same way on a V3 and get bsod and it's all my fault. All OCZ has to do is look around at other forums and see there are far more than 1% being effected by this and it's cross platform with the Sandforce controller being the only constant.
They said they were able to recreate the same problem on other competitors ssd beside V3. When asked, Tony pointed to stuttering experienced by c300 users that was taken care of by firmware. That was his only example, no others and no c300 bsod either.
Now with firmware that reduces performance to get rid of bsod, we're back to the same old story that none of OCZ computers are showing the slowdown just like none of their computers would do the bsod. I dropped 50 points in as-ssd after new firmware was put in then secure erase and fresh install Win7. Obviously, I must be doing something wrong again.
Never again OCZ, never again.