[Sweclockers] EXTRA: Problems with Sand Force 2 - withdrawal in sight

[Ghostleader]

After numerous reports of failing SSDs with control circuits from the series SandForce 2 speaks of a total withdrawal of all the models.

With Intel's chipsets feldesignade for Sandy Bridge in recent memory, it looks like the computer world suffers another serious case of defective components. This time, the SSDs with the control circuits from the series Sand Force 2 who are in the spotlight after numerous reports of failing units.

Users in particular SweClockers forum talks about SandForce 2-based storage devices that suddenly stop working. Symptoms vary from Bluescreens in Windows that the device no longer detected by BIOS. The common denominator is that there seems to be no apparent solution to the problem.

Now, say sources SweClockers that SSDs with SandForce 2 sold topped and it can speak of a total withdrawal of all the models. Possible candidates are Corsair Force 3 and OCZ Vertex 3, but exactly which units these are is still unknown.

The news updated as more details become known.

11:45
Corsair Force 3 120 GB sold topped the Swedish stores. (The only Corsair unit delivered in large volumes)

13:46
As it stands now is the Corsair Force 3 120 GB which is withdrawn. OCZ "working with Sand Force" to correct similar problems with Vertex 3 of the firmware

16:16
To clarify: OCZ withdraws no SSDs.

Source

Translated source


EDIT: I just found this in the OCZ forum regarding BSOD and vertex3,agility3 or solid3

There are way to many goal posts moving around to nail this issue down. To date, 1st June 2011, SF have not been able to reproduce any of the issues you guys are seeing..reason? I think its setup not FW causing you all issues.

So...its time to fix the goal posts, we all test and report on our systems 1 way and 1 way only.

1 absolutely NO OVERCLOCKING..why? it turns off power management features and has documented problems with power management especially on P67 and higher chipsets.

2 You will be testing with a fresh install and your board needs to have been flashed to the newest UEFI or Bios available and it must have seen a thorough cmos clear.

3 You have to set up windows 7 exactly as I document it here and report info exactly as I ask you do..this info then will go back to OCZ and SF as needed.

I'm not interested in people who are imaging drives, don't moan at me, you want to test you test my way or you sit and watch, simple as that. If you overclock you MUST set systems to defaults, this includes ram and cpu and video...there are too many variables that get changed when the systems are overclocked and frankly we can not warrant a drive if its failing because the boards are doing something weird when overclocked...thats a problem for the motherboard manufacturer not OCZ.

So...what to do.

people with Intel based boards and vtx3/agility3 or solid3 and bsod look in here

[zalbard]

Wow... I knew a few people had issues, but I had no clue about the scale...

[Ao1]

Should have happened a while back.

[Ghostleader]

I just found this in the OCZ forum regarding BSOD and vertex3,agility3 or solid3

There are way to many goal posts moving around to nail this issue down. To date, 1st June 2011, SF have not been able to reproduce any of the issues you guys are seeing..reason? I think its setup not FW causing you all issues.

So...its time to fix the goal posts, we all test and report on our systems 1 way and 1 way only.

1 absolutely NO OVERCLOCKING..why? it turns off power management features and has documented problems with power management especially on P67 and higher chipsets.

2 You will be testing with a fresh install and your board needs to have been flashed to the newest UEFI or Bios available and it must have seen a thorough cmos clear.

3 You have to set up windows 7 exactly as I document it here and report info exactly as I ask you do..this info then will go back to OCZ and SF as needed.

I'm not interested in people who are imaging drives, don't moan at me, you want to test you test my way or you sit and watch, simple as that. If you overclock you MUST set systems to defaults, this includes ram and cpu and video...there are too many variables that get changed when the systems are overclocked and frankly we can not warrant a drive if its failing because the boards are doing something weird when overclocked...thats a problem for the motherboard manufacturer not OCZ.

So...what to do.

people with Intel based boards and vtx3/agility3 or solid3 and bsod look in here



No OC with a Intel board and a OCZ Sandforce controlled SSD

[Brahmzy]

Again, this all has to do with firmware 2.06. I have had Vertex 3's running absolutely rock-solid and fast for 2 months now @ 4.8GHz. The key being I never "upgraded" to fw 2.06.

The 2.06 firmware is flawed in it's hit n miss, flaky compatability. I'm not downplaying this at all - it's of monumental embarrassment for SF and any vendors that sell SF 2xxx SSDs.

Sandforce is to blame for the broken firmware. How the problems are being handled can be blamed on the vendors (OCZ, Corsair) that have marketed, sold and supported the products.

It's going to take SF's reputation years to recover from this one - especially since their last gen products weren't exactly problem-free.

[Anvil]

Terrible story for all parties, it's not news exactly though...

Almost all shops in Norway have pulled Force 3, Vertex 3 is still in stock though.

I guess Corsair was hit the hardest as I expect all or most of their drives were at FW 2.06.

[Andrea deluxe]

crucial/intel rulez

i have an M4 128gb and with a fresh install and the latest intel storage driver i do not have a single issues.....

this afterall a trotthled agility and vertex3 120gb....

crucial is the winner of these bad ssd situation....

[Ao1]

It certainly seems that 2.02 is more stable BUT, if it was that stable why has back flashing been excluded as an interim solution?

Its also hard to understand how these issues did not get picked up by any of the SF vendors that are supposed to test/valid the FW before release. Looking on the Corsair forum it is occurring on SATA 2 & 3 and a wide range of motherboards.

[One_Hertz]

Sandforce has been a mess from the very beginning. Too small of a company trying to do too much without proper testing.

[Anvil]

It certainly seems that 2.02 is more stable BUT, if it was that stable why has back flashing been excluded as an interim solution?

Its also hard to understand how these issues did not get picked up by any of the SF vendors that are supposed to test/valid the FW before release. Looking on the Corsair forum it is occurring on SATA 2 & 3 and a wide range of motherboards.

Beats me

All my Vertex 3's on 2.02 are doing just fine and the one I did flash to 2.06 looks fine as well.
The Agility 3 that arrived with 2.06 is flaky at best on P67/Z68, X58 looks fine though, strange...

Anyways, they cant expect ordinary people to perform any tasks other than the normal ones before installing the OS onto their brand new SSDs.

[RyderOCZ]

I just found this in the OCZ forum regarding BSOD and vertex3,agility3 or solid3
No OC with a Intel board and a OCZ Sandforce controlled SSD

No OC during testing, if the method works for you, go back to overclocking.

Please don't read into things that we are saying you can't overclock with an SSD.
Just trying to get EVERYONE on a level playing field to find out where the issue lies.
If XYZ (who doesn't overclock) has no stability on board PDQ, but then ABC with same board does have issues, it makes it hard to troubleshoot. Stock speeds makes for equal testing, then move on to restoring overclock and see if issue appears again, if not, then move on to next item.

There really is something fishy going on with drivers and chipsets. We have already had cases showing this.

[Ao1]

Doesn't seem to be a problem with Intel/Micron/Samsung/Toshiba etc

[Ao1]

Whoa........

CORSAIR FORCE SERIES 3 UPDATE, JUNE 7, 2011

Over the past several days, we have analyzed issues associated with the stability of our recently released 120GB Force Series 3 SSD. Our review has identified that a significant percentage of these drive do not perform to specifications. The solution will require changes to both the SSD firmware and the hardware components of the SSD itself.

We have stopped shipments of the affected drive, Corsair part number CSSD-F120GB3-BK, effective immediately and have implemented the following remediation plan:

(1) ALL users should cease using the CSSD-F120GBG3-BK drive immediately, even if you have not experienced issues of stability.

(2) Users may then either RETURN the drive to Corsair, at Corsair's expense, to be exchanged with a replacement drive that does meet specifications OR

(3) Return the drive to their place of purchase.


Corsair is committed to product quality and we want to assure our customers that this remediation plan will be implemented at no cost to you.


To return your Force Series 3 120GB SSD to Corsair, contact Corsair to ARRANGE YOUR REPLACEMENT . Corsair will pay for shipping your drive back to Corsair and will replace your drive free of charge. Estimated turnaround time is ten business days.


http://forum.corsair.com/v3/showthread.php?t=95825

[Anvil]

So, nothing on the 240GB.

It does leave a lot of questions in the air though, anyways, Kudos for taking action.

[One_Hertz]

Kudos for taking action.

+1 Kudos for Corsair taking responsibility.

[Ao1]

No choice I guess if there is a hardware issue involved as well. Boy this sucks for the whole SSD industry.

[Ghostleader]

@Ao1

Nice find, I update the OP with it if you don´t mind.

[thebad]

This is the very reason why I have been preaching Intel/Micron/Samsung/Toshiba everywhere I have posted. The Indilinx/OCZ/Sandforce/Jmicron do not have the testing capability the big boys have and with the number of different OS motherboards configurations, at least early adopters are bound to do the beta testing of the drives for the smaller companies until they fix their firmwares/hardware. This is the same story over and over again. The Indilinx Barefoot (Vertex), Jmicron, Sandforce (Vertex 2) and now Sandforce 2 (Vertex 3) all have had issues for months after market release with FW disappearing/ destructive reflash and what not required. If you care about your data, stick with the big name controllers until the small guys get their stuff figured out. Heck even the Sandforce/IMFT 25nm combination had problems just a couple of months ago and OCZ even snuck these drives on the market without P/N change.

[Computurd]

if there is a hardware issue then OCZ will probably follow suit with a recall as well. man that sucks for both companies. bad thing is some of these people buying SSD now for the first time, now that they are hitting somewhat of a mainstream status, are going to be turned off after this.
Looks like the issue lies with SF imo.

[Brahmzy]

LOL!! The poop is hitting the fan for sure!!

[Anvil]

An official statement from OCZ

Link http://www.ocztechnologyforum.com/fo...-manufacturers

[Brahmzy]

Good for OCZ that they don't have to recall due to hardware problems... That'll help soften the blow.

[zalbard]

An official statement from OCZ

Link http://www.ocztechnologyforum.com/fo...-manufacturers

Good news for OCZ users. Hope the fix is coming soon.

[One_Hertz]

I, for one, do not believe that Corsair somehow messed up their SSDs and OCZ did not. It makes no sense because 95% of the complexity of the SSDs is in the controller and the NAND. The PCB itself (which is all the manufacturers have control over) is pretty damn straight forward and difficult to mess up. Additionally, Corsair and OCZ ssds are experiencing exactly the same issue. Seems pretty clear that the issue is in the controller and not in the PCB. Of course Corsair could have made a bad mistake I suppose, it is just unlikely.

Just my opinion.

[Brahmzy]

I, for one, do not believe that Corsair somehow messed up their SSDs and OCZ did not. It makes no sense because 95% of the complexity of the SSDs is in the controller and the NAND. The PCB itself (which is all the manufacturers have control over) is pretty damn straight forward and difficult to mess up. Additionally, Corsair and OCZ ssds are experiencing exactly the same issue. Seems pretty clear that the issue is in the controller and not in the PCB. Of course Corsair could have made a bad mistake I suppose, it is just unlikely.

Just my opinion.

Corsair is saying this hardware problem is ONLY affecting the 120GB version of their Force drive - their other Force/Force GT SF 2xxx SSDs don't have the same hardware problem. Yet, their other drives experience the BSOD issue, just like their OCZ counterparts. I think we're confusing a few things here, or rather, Corsair's announcement has introduced some further confusion into this mess.