Are all SandForce based SSDs equal?

[therat]

i am looking at upgrading my SSD from an OCZ Vertex 2 to either a SandForce based 22xx drive or a Crucial M4. I like the speed of the SF based drives but am worried about their reliability especially in view of the recent BSOD bug.

I have been reading reviews and users comments of the OWC Mercury Extreme Pro 6G SSD which is a SF 22xx based drive and I can't find any mention of them suffering from the BSOD bug.

I also read on their website the following:

"If you have an OWC SSD, though, you don’t need TRIM. The SandForce controller in our SSDs takes care of this “garbage collection” as well as performs various other tasks that keep your drive running at optimal speed, without the drop-off that you see with other brands. Especially note page two of this performance testing expert’s report where he feels so strongly about TRIM’s inefficiency that he calls call it “half-baked”…and that’s the kid friendly version of the phrase.

In fact, enabling TRIM could actually hurt the performance and reliability of your OWC SSD, rather than help it."

i know that applies to a MAC (am still waiting to hear back from OWC if it applies to Win 7) but it had me thinking "Are all SF 22xx based drives the same?". OWC offer a 5 year warranty which is all well and good but I am not interested in having to continually RMA my drive if it fails.

Is it possible that OWC have produced a better, more reliable SF22xx SSD or is it just hype and if one brand suffers from a certain problem then all SF22xx SSDs will regardless of make?

If I can be sure the OWC SSD will be as reliable as the M4 then I will go for it otherwise I'll have to settle for the M4.

[CedricFP]

Are OWC saying by this phrase: "without the drop-off that you see with other brands" that they've disabled the throttling on SF 2 based drives?

[johnw]

If it is true that enabling TRIM hurts performance, then somehow they have really screwed up their TRIM implementation. All TRIM does is inform the SSD of which sectors are no longer needed by the OS/filesystem. It is just information, that the SSD firmware can either make use of or not. It would be pathetic if providing some information to the SSD actually hurt performance.

[therat]

I think the TRIM being referred to is on the MAC which prior to Lion, did not support TRIM and the only TRIM support was provided by a third-party utility. As far as TRIM implemented by windows, I'm waiting for a definite answer from OWC support.

[Tony]

LOL you have to love OWC's bogus marketing...they use reference design PCB and the same firmware as everyone else, the ONLY thing you can do different is set warranty period to what ever you want and toggle a few other options in the MPtool.

Do NOT get suckered by them.

[therat]

Can they turn throttling off (they seem to be suggesting they have)?

[CedricFP]

Can they turn throttling off (they seem to be suggesting they have)?

This is what I thought, and from what I've read, I suspect it is possible for an end-user to so... they certainly could?

[Ao1]

^ In theory yes, throttling is a vendor selectable option. It does not mean that they have switched it off.

Vendor selectable options via the MPTool (AFAIK)
• Initial credit duration
• LTT warranty duration
• Burst speed throttling
• Power throttling
• RAISE - on or off

@ Tony, how about a statement on RAISE? Is it activated on OCZ 60, 120 & 240GB drives?

According to Storage Review:
Incidentally, RAISE can now be turned on or off by SSD manufacturers, which is especially important for lower capacity drives. Imagine a 64GB SSD using 8 8GB NAND chips. With RAISE enabled, one of those 8GB NAND chips gets used for drive endurance management. Without, the user gets full access to the NAND, and actually gives up little in terms of endurance for those capacity levels (1029 vs. 1016 error rate) in practical terms. Drives with RAISE will see a performance gain though (10-20%), so we would expect drives in the 128GB range to be the cutoff for the disabling of RAISE. It's important to clarify this point as consumers will likely see drives of similar capacity side-by-side (120GB and 128GB for instance).

EDIT: According to Kingston:

RAISE is a fundamental feature of SandForce drive architecture and so we are pleased to confirm that this feature is also present on our SH100S3/120G and SH100S3/240G HyerX SSD drives.

[therat]

OWC have confirmed that their Mercury Extreme Pro 6G 120GB SSD does have RAISE turned ON. They also confirmed that the SandForce controller that OWC SSDs are based upon does support TRIM. It works just fine under Windows 7.

i wonder if i ask whether OWC will comment on whether throttling is disabled or not on their drives?

[Ao1]

Anandtech did not get back to me. Kinston were not so clear on their statement once I started to dig further. Storage Review state that RAISE improves performance, which I somewhat doubt.

Buying a SF drive is like buying a pig in a poke, you never really know what you are getting.

[Tony]

Forwarded your concerns to our head of marketing, I could tell you now but its best for all it comes direct from OCZ corporate...

[Ao1]

Thanks Tony, much appreciated.

[Computurd]

buying a pig in a poke

what if you poke the pig first

[therat]

As expected the OWC SF based drives do suffer from throttling as do all other SF based drives. Pity it wasn't disabled.

[Tony]

you guys want nothing but speed...remember 0.00001% of the market does not have much say when it comes to getting features added, the bigger picture is money has to be made and the larger market is where the money is. If you want fast go pcie and go more than 1 drive...sorry to say its the only way.

The technology is working against you im afraid, as die shrink comes in more constraints will come in along with it although outright speed will increase overall.

On a side note here, I run fancycache with my revo x2 on the workstation, 4GB cache size with deferred writes....its damn snappy

[th3pwn3r]

you guys want nothing but speed...remember 0.00001% of the market does not have much say when it comes to getting features added, the bigger picture is money has to be made and the larger market is where the money is. If you want fast go pcie and go more than 1 drive...sorry to say its the only way.

The technology is working against you im afraid, as die shrink comes in more constraints will come in along with it although outright speed will increase overall.

On a side note here, I run fancycache with my revo x2 on the workstation, 4GB cache size with deferred writes....its damn snappy

Did you notice any increases in temperature when you started running the Revo as opposed to not? Idle power consumption is way up there and my system already produces an annoying amount of heat as is, maybe my PSU just sucks and creates heat itself but I'd like to keep things cooler in my room if possible, eh, I guess in the Winter time it's kind of nice but now during Summer no thanks.

[Ao1]

you guys want nothing but speed...

So should one assume that RAISE is disabled on the 60, 120 and 240 GB drives, or is it just disabled on the 60 & 120 drives?

[RyderOCZ]

All "3" series 60GB drives have RAISE off, all other drives have RAISE on.

RAISE is on for ALL the "2" series drives. Hope that helps.

[Computurd]

they did recently add a heatsink to the revo on another revision of the same product. that would lead me to conclude that it is kicking off some heat, possibly more than it was able to dissipate effectively without the heatsink.

[Ao1]

LOL you have to love OWC's bogus marketing...they use reference design PCB and the same firmware as everyone else, the ONLY thing you can do different is set warranty period to what ever you want and toggle a few other options in the MPtool.

Do NOT get suckered by them.

More bogus marketing?

RE: Does anyone else care... by OWC Grant on Friday, August 12, 2011

Glad you noticed that warranty term because it's somewhat related to topic of this article. I've been in direct contact with Anand on this as the tone of article is all-encompassing and I wanted to shed some light on that from our perspective.

While many SF based SSDs share firmware, not all hardware is the same. Our SSDs have subtle design and/or component differences which is what we feel reduces or eliminates our products susceptibility to the BSOD issue.

The honest truth is we have not been able to create a BSOD issue here with our SSDs using the same procedures that caused other brands' SSDs to experience BSOD. Nor have we received or read one direct report of such an occurrence using our drives.

And while we cut our teeth so to speak in the Mac industry, PLENTY of PC users have our SSDs in their systems...as well as that we do extensive testing on a variety of motherboards/system configs to ensure long term reliable operation.

More supportive perhaps is the fact that we've had other brand users who experienced BSOD, but after buying our SSD, they reported back that it eliminated any issues they were experiencing.

Page 6 of comments on the Anandtech review: http://www.anandtech.com/Show/Index/...pared#comments

[Wendy]

More bogus marketing?

RE: Does anyone else care... by OWC Grant on Friday, August 12, 2011

Glad you noticed that warranty term because it's somewhat related to topic of this article. I've been in direct contact with Anand on this as the tone of article is all-encompassing and I wanted to shed some light on that from our perspective.

While many SF based SSDs share firmware, not all hardware is the same. Our SSDs have subtle design and/or component differences which is what we feel reduces or eliminates our products susceptibility to the BSOD issue.

The honest truth is we have not been able to create a BSOD issue here with our SSDs using the same procedures that caused other brands' SSDs to experience BSOD. Nor have we received or read one direct report of such an occurrence using our drives.

And while we cut our teeth so to speak in the Mac industry, PLENTY of PC users have our SSDs in their systems...as well as that we do extensive testing on a variety of motherboards/system configs to ensure long term reliable operation.

More supportive perhaps is the fact that we've had other brand users who experienced BSOD, but after buying our SSD, they reported back that it eliminated any issues they were experiencing.

Page 6 of comments on the Anandtech review: http://www.anandtech.com/Show/Index/...pared#comments

Cerntainly not from an engineer, I'd say a marketing rep.
OWC appear to use a SF reference design, I guess it's a blue PCB, but I doubt that would make it BSOD proof.

Fact:
The core logic on all SF2281 is the same, they all share the same interface to the outside world via SATA, and all share the same interface to the NAND. So they all stand or fall by that same core logic.

IMO, if they haven't witnessed the problem with their SSDs, then it's more likely that they haven't sold that many of them.

[Computurd]

i do in fact know a few users of OWC SSDs, as i have used some myself and did an article on them. strange thing is not one complaint, ever.
what thats worth, i dunno.
im just saying this, i haven't seen anyone complain of issues with OWC drives.

[Wendy]

i do in fact know a few users of OWC SSDs, as i have used some myself and did an article on them. strange thing is not one complaint, ever.
what thats worth, i dunno.
im just saying this, i haven't seen anyone complain of issues with OWC drives.

Most folks have no problem with SF2281 based SSDs, including OCZ SSDs.
I just don't see how OWC could do something so fundamentally different with their SF2281 SSD to be totally immune of BSOD.

[edit]
Apparently they aren't immune.
http://forums.anandtech.com/showpost...1&postcount=51