Sign in with
Sign up | Sign in

SandForce Random BSOD Firmware Bug Identified, Fixed

By - Source: AnandTech

SandForce looks to have finally provided a firmware update for its infamous BSOD/disconnect issue that have plagued SF-2200-based SSDs from the start.

OCZ and SandForce don't provide many details on the root cause of the issue, except to state is was a bug in the firmware that caused "a rare condition that may cause Windows Blue Screen." OCZ has been testing the new firmware for several weeks and has finally released it for public use (look for other manufacturers to quickly follow with their firmware updates). The new firmware is version 2.15 for OCZ drives and 3.3.2 for drives that utilize SandForce's standard numbering. As with any firmware update, users should back up all vital data before proceeding with the update.

OCZ sent along its official statement: 

OCZ is pleased to announce that the cause of a BSOD issue experienced by some SF-2000-based drive owners has been identified by OCZ and SandForce. A new firmware update which directly addresses this BSOD occurrence related to SF-2000 based SSDs is available here. All newly manufactured OCZ SF-2000 based SSDs will feature the new 2.15 firmware revision (which is based on SandForce firmware version 3.3.2.) We highly recommend that any customers that have experienced the BSOD issue update their firmware to 2.15.

We sincerely appreciate the support from our customers, and if any customers have any questions or require additional support please do not hesitate to contact a customer service representative and we will be happy to address any questions or concerns.


Source: OCZ and Release Notes.


Does this firmware update bring more confidence to those who have been afraid to purchase a SandForce SF-2200 controller or are you going to take a wait-and-see approach? 

There are 41 Comments. B
Other Comments
  • 6 Ð
    anonymous@guest , October 18, 2011 9:05 PM
    Finally! If Indeed this solves this LONG standing issue, I would happily purchase a Sandforce based drive!
  • 9 Ð
    nikorr , October 18, 2011 9:20 PM
    Now people can sleep at night : )
  • -3 Ð
    gogogadgetliver , October 18, 2011 9:22 PM
    How could they fix this if they already fixed this? Sick of Sandforce and frankly I no longer trust them. I RMA'd.

  • 5 Ð
    nikorr , October 18, 2011 9:26 PM
    gogogadgetliverHow could they fix this if they already fixed this? Sick of Sandforce and frankly I no longer trust them. I RMA'd.

    Don't be so paranoid : ) Look what happened to Sandy Bridge, its piece of electronic. Happens all the time.
  • 5 Ð
    philharmonik , October 18, 2011 9:33 PM
    I've owned a SF2000 Corsair Force 3 120 since June. I was relieved to see they finally have fixed the issue. My BSODS' and freezes weren't very frequent, but they did happen. Nice to know I can update the FW and not have to worry about this anymore! About freakin time!
  • 0 Ð
    gogogadgetliver , October 18, 2011 9:34 PM
    nikorrDon't be so paranoid : ) Look what happened to Sandy Bridge, its piece of electronic. Happens all the time.


    If this was a one-off incident I would possibly be guilty of paranoia.

    If I've experienced data loss, multiple return hassles, and multiple promises that things are fixed and I continue to have faith then I would be guilty of being a fool.

    They are fast drives and I hope everyone enjoys them without problems. I'll sit out this round and skip the SF-2281. I'll revisit Sandforce once they have a proven track record.
  • 0 Ð
    timaeus , October 18, 2011 9:53 PM
    I'll be happy if this actually fixes it. Luckily my BSODs haven't been too often--only gotten 2 in the 3 or so months I've owned it. But it will be nice to not have to worry about it.
  • 1 Ð
    __Miguel_ , October 18, 2011 10:02 PM
    Finally! I've been reading about those dreaded bugs for months now... I have been holding on getting an SSD partially for that reason...

    So, the big questions here are:

    1) Has anyone already tested this? If so, how did the drive fare?

    2) Do SF-based SSDs finally replace the current "go with m4/Intel/Samsung 830" general rule of thumb people have been telling me for a few months already? Or are there other problems these SSDs have that might not grant them the top spot on price/performance/reliability?

    Miguel
  • 2 Ð
    anonymous@guest , October 18, 2011 10:13 PM
    I would wait and see if it actually fixes the problem, and not just hopefully fixes the problem, since it's not as rare as they try to make it sound.
  • 3 Ð
    moonshell , October 18, 2011 10:17 PM
    i juat see some people at ocz forum still having problems after installing the new firmware...
  • 0 Ð
    cknobman , October 18, 2011 11:01 PM
    Had my Corsair Force 3 120gb SSD drive with Sandforce 2200 controller for about a month now in my Core i5 520 laptop and have yet to see a blue screen.

  • 0 Ð
    JohnnyLucky , October 18, 2011 11:52 PM
    I read several other news reports. It is my understanding it is for a specific firmware bug. There is a possibility that not all BSOD's are caused by the same firmware bug. It is going to be another situation where it may or may not work. It just depends whether or not a user has an ssd with the correct firmware bug.
  • 0 Ð
    clonazepam , October 18, 2011 11:58 PM
    I wonder if this resolves the bug I experienced and reproduced 3+ times. On a nvidia 750i chipset board, any attempt to update the firmware on the ocz agility 3 and ocz vertex 3 max iops produced a BSOD until the nvidia graphics driver was uninstalled and using the generic vga driver. I really dont feel like risking the data on either SSD to find out though. SSDs updated fine w/o any issues on the Z68 based board.

    Edit: I experienced the bug I mentioned above on firmwares 2.09 and 2.11. So far I haven't had cause to seek more recent updates and haven't looked into them since. I will tread lightly when considering the update.
  • 0 Ð
    compton , October 19, 2011 12:05 AM
    There are several different causes, but this should address the two main ones. I don't understand what the hell sort of nonsense SF and it's partners have been talking up, but on my system and workload, the problems are repeatable, frequent, and evenly spaced.
  • 0 Ð
    ikefu , October 19, 2011 12:11 AM
    Glad that got a fix... but its too late. I couldn't wait any longer and bought a Crucial M4 128Gb 2 months ago. Stability is critical on my laptop which is a combo business/personal rig and this fix has been a LONG time coming.

    I was all ready to jump on the SF bandwagon, but now that I went Crucial I don't see any reason to change with as great as the M4 is.
  • 0 Ð
    buzznut , October 19, 2011 12:36 AM
    clonazepamI wonder if this resolves the bug I experienced and reproduced 3+ times. On a nvidia 750i chipset board, any attempt to update the firmware on the ocz agility 3 and ocz vertex 3 max iops produced a BSOD until the nvidia graphics driver was uninstalled and using the generic vga driver. I really dont feel like risking the data on either SSD to find out though. SSDs updated fine w/o any issues on the Z68 based board.Edit: I experienced the bug I mentioned above on firmwares 2.09 and 2.11. So far I haven't had cause to seek more recent updates and haven't looked into them since. I will tread lightly when considering the update.


    Me too. My drive is working fine and I won't update until I'm sure it's safe to do so. I have had a couple of blue screens and some odd things at startup that scared the crap outta me like messages about not being able to find the boot record. But this is a typical windows experience for me, weird stuff happens from time to time.
  • 0 Ð
    excidium , October 19, 2011 12:44 AM
    clonazepamI wonder if this resolves the bug I experienced and reproduced 3+ times. On a nvidia 750i chipset board, any attempt to update the firmware on the ocz agility 3 and ocz vertex 3 max iops produced a BSOD until the nvidia graphics driver was uninstalled and using the generic vga driver. I really dont feel like risking the data on either SSD to find out though. SSDs updated fine w/o any issues on the Z68 based board.Edit: I experienced the bug I mentioned above on firmwares 2.09 and 2.11. So far I haven't had cause to seek more recent updates and haven't looked into them since. I will tread lightly when considering the update.


    Have you tried using the Linux tool ( easy Linux boot iso, just need a small USB drive to boot from )? Works wonderfully.
  • 0 Ð
    clonazepam , October 19, 2011 12:51 AM
    Nah. I only experiment with Linux on an aging laptop. Once I figured out to un-install the nVidia graphics driver, the update was nearly painless. Now, if needed, I just do the firmware updates on the Z68 board and don't have to mess with that nonsense (that board is currently being outfitted with wc setup so I can't play with the new firmware just yet).
  • 0 Ð
    excidium , October 19, 2011 1:26 AM
    clonazepamNah. I only experiment with Linux on an aging laptop. Once I figured out to un-install the nVidia graphics driver, the update was nearly painless. Now, if needed, I just do the firmware updates on the Z68 board and don't have to mess with that nonsense (that board is currently being outfitted with wc setup so I can't play with the new firmware just yet).


    You dont have to install Linux... you realize that right? Its a bootable Linux image that exists on the USB key, and boots into a "live" Linux OS. You dont need to uninstall anything or swap drives around. You plug your USB key in and reboot your machine, catch it at boot-time and boot from the USB key. At the Linux desktop you run a script. Done. Reboot your machine, unplug your USB key and boot into Windows.

    Mind you, thats not to say they should not be fixing the issue with the i5 750 setup, just saying there are alternatives that are easier to manage while waiting for a fix to that issue.
  • -1 Ð
    techguy378 , October 19, 2011 1:43 AM
    This issue is not critical by any stretch of the imagination. On a Sandforce-based SSD that is always under very heavy use 24/7 this BSOD issue happens maybe once every 6 months at most. Sandforce controllers have always been more reliable than any other brand since their introduction. Why in the world would anyone not want a Sandforce-based SSD?
Display more comments