Intel pulls TRIM firmware after data corruption complaints

Are Intel’s new 34-nm solid-state drives just naturally predisposed to eating your data? The chipmaker had to halt shipments this summer after a BIOS-password-related data corruption issue came to light, and today, Engadget writes that Intel has pulled its new TRIM-enabled firmware after some users who installed it complained of data loss.

Here’s what Intel told the gadget blog:

Yes, we have been contacted by users with issues with the firmware upgrade for our 34nm SSDs and we are investigating. We take all sightings and issues seriously and are working toward resolution. We have temporarily taken down the firmware link while we investigate.

A visit to the Intel support forums shows a five-page discussion thread about the latest corruption problem. The thread starter claims he successfully installed the new firmware, rebooted, let Windows 7 install new drivers, and attempted to reboot again, after which he could no longer start Windows. Other posts suggest affected drives may become unusable.

As we reported yesterday, the new Intel firmware is supposed to enhance write performance by taking advantage of Windows 7’s TRIM functionality. TRIM clears flash pages instead of marking them as available when users delete data, thus skirting the block-rewrite penalty in SSDs.

Comments closed
    • UberGerbil
    • 13 years ago

    Yeah — here’s a chance for him to be completely on-topic and even accurate, and he’s not here because he blew too many chances due to extreme irrationality earlier.

    • 5150
    • 13 years ago

    Garbage Collection has worked very well for me and has been around much longer than TRIM.

    • just brew it!
    • 13 years ago

    q[

    • OneArmedScissor
    • 13 years ago

    But they’re also one of the most popular SSDs, if not the most popular, ever, and they’ve been out of stock since they came out, a while back.

    There are a lot of them out there, and they are expensive. It’s kind of like saying it’s OK for a $100,000 car to have *[

    • crazybus
    • 13 years ago

    What’s wrong with intel?

    • l33t-g4m3r
    • 13 years ago

    [sarcasm]
    OMG! A firmware bug! We must all avoid intel storage products now!
    [/sarcasm]

    • wira020
    • 13 years ago

    I dont even have the money to buy em anyway 🙁

    • BailoutBenny
    • 13 years ago

    Looks like Intel is using Apple programmers.

    • MarkD
    • 13 years ago

    It should be sorted out before my SSD arrives, based on what Amazon is telling me. Intel still can’t ship enough G2 drives to meet the demand.

    • 5150
    • 13 years ago

    Thanks for the advice, unfortunatly I don’t have another model and a formatter board was crazy expensive so we ordered a new one. This printer has had issues for a while, in fact, everything from the LaserJet 3330 to this have been absolute trash. I’m going to try my luck with Oki for a while and see how they do.

    • zagortenay
    • 13 years ago

    I am glad I did not jump on the bandwagon and buy an SSD. SSDs are “still” too small in capacity, too expensive and full of problems. They just don’t make sense. Not yet…

    • radix
    • 13 years ago

    Except the Intel drives probably have a much smaller user-base, since they are a lot more expensive than the Seagate drives that had issues.

    • Mr Bill
    • 13 years ago

    If you have another printer of the same model, try a hot flash. This only works if the bios is socketed. Loosen the bios chip, boot the unit, remove the good chip, put in the bricked chip and then try flashing again. This has worked a couple time for me with motherboards.

    • 5150
    • 13 years ago

    I bricked a HP LaserJet M2727 yesterday doing a firmware update because the printer locked up in the middle of the update. Just wanted to share.

    • 5150
    • 13 years ago

    I’ve been flip flopping back and forth at work with the X25-M and the Vertex’s. I may swing more to the Vertex side, not because of the firmware issues, because they’ve both had them, but because I think the Vertex has better overall performance.

    • Flying Fox
    • 13 years ago

    Given the size of that discussion thread, I think there are quite a lot of people who did that already. The power of the internet is great these days.

    • Flying Fox
    • 13 years ago

    I can’t find those X18-M available anyways. 🙁

    • Flying Fox
    • 13 years ago

    A couple of guys at work already bought them.

    • Goty
    • 13 years ago

    Oh yes they will.

    • flip-mode
    • 13 years ago

    They’re band new workstation installs, so there isn’t any “data” on them yet, really. I installed them Sunday.

    • Kurlon
    • 13 years ago

    I love how everyone is assuming data corruption did occur due to the firmware… There is the possibility that this was user error.

    • MadManOriginal
    • 13 years ago

    I’d guess they are standard image drives so the data is probably not important.

    • AxMi-24
    • 13 years ago

    This sucks. I need a 1,8″ SSD for my XT2 from dell and ocz does not make those 🙁

    • Firestarter
    • 13 years ago

    I sure hope the data on those drives is considered expendable if you have such an aggressive update policy 😮

    • MadManOriginal
    • 13 years ago

    On the other hand Intel’s SSDs performed perfectly well in new or used states without TRIM or other software trickery like the wiper software.

    • geekl33tgamer
    • 13 years ago

    LOL!

    Unlike the whole Seagate firmware issue, it does look like Intel have stopped this problem before loads of users have downloaded the firmware.

    Wonder why it never come up when Intel tested it tho, before release?

    • Farting Bob
    • 13 years ago

    Thats not disgusting. Its just a little disapointing. But its not like SSD’s do not work with w7 unless you have TRIM support. They’ll work exactly as they did before. If you think that is disgusting then you really need to calm down and maybe venture out the house once in a while.

    If i was going to upgrade my SSD firmware, id be taking a mirror image of the dsic first. At the size of disc we’re talking its not a complete pain in the ass day long operation like backing up a 1TB HDD would be.

    • Fighterpilot
    • 13 years ago

    Better they find these bugs while the product is still new and cutting edge…rather than in a few months when I buy one.

    • UberGerbil
    • 13 years ago

    Well, I think some level of Trim support has been in the Linux kernel since 2.6.28, so that would be another way to determine where the problem is. But Win7 has been frozen and effectively shipping for a while now (not in retail, but certainly the RTM code has been in Intel’s hands for many weeks). So Intel had a stable target to test against; if their QA had turned up a problem and they determined it was Microsoft’s fault, then either they work around it in the firmware or they hold off on releasing the firmware until MS has pushed an update to Win7. But the crucial point is that, regardless of whose fault it is, /[

    • Meadows
    • 13 years ago

    My heart weeps.

    • just brew it!
    • 13 years ago

    What are the odds that this was purely Intel’s fault? We’re talking about new firmware which is supposed to work together with a brand-new OS feature. This could just as easily be MS’s fault, or a case where they both dropped the ball on testing…

    • mako
    • 13 years ago

    Gosh, Intel, I’m not feeling so confident about the G2 I bought the other week. On the bright side, I haven’t gotten a chance to install it yet, so there’s no data to lose.

    • OneArmedScissor
    • 13 years ago

    *que ATHF movie intro song*

    WE’VE ALREADY GOT YOUR F***ING MONEY!

    • flip-mode
    • 13 years ago

    Ouch. Glad I didn’t apply the update on the machines at work today. I meant to, but laziness combined with other things to do conspired in my favor.

    • BiffStroganoffsky
    • 13 years ago

    Cyril’s recall of the complaints leaves out a little detail. The reboot prompts are all from the automated firmware update process and not of the user’s own doing. They confirm but do not instigate the reboots.

    • UberGerbil
    • 13 years ago

    Intel has pulled the update, which is the prudent thing to do, but do we have any data that shows with certainty that this was a systematic problem? I haven’t waded through the five page thread (or looked elsewhere) — are we sure the problem was actually Intel’s firmware, or something else — drivers, users not following directions, or whatever?

    Presumably Intel’s QA would’ve caught any obvious and common problem, so at the very least there has to be a confounding factor at work. It’s oem thing (and bad news for Intel) if the factor was, say, a certain chipset; it’s quite another if it’s only happening on (say) heavily overclocked systems or to users who didn’t reboot when they should have, etc.

    • Jigar
    • 13 years ago

    That’s what documentations are for. To save your ass, if you’ve made something, that can break things.

    • Kurotetsu
    • 13 years ago

    <Enthusiasts> Documentation? What’s that?

    • adisor19
    • 13 years ago

    So much for that Intel reliability that was due to them being INTEL. Uhuh, so far it looks like only the Indilinx controller manged to get TRIM done right. I don’t know about you but the price premium Intel charges for their SSDs looks kinda bad now..

    Adi

    • tejas84
    • 13 years ago

    Well well well… who would have thought it eh? Intel and data corruption.

    It seems the Nvidia haters who continually moan about chipset data corruption have seen their precious Intel stumble into the same position. (obv tho Nvidia’s problem was chipset rather than storage but still both are crucial faults)

    OCZ and Indilinix based SSDs will get a boost from this!

    • Kurkotain
    • 13 years ago

    actually its more like lots of enthusiasts pulled the trigger faster than what they should have…

    • Sunburn74
    • 13 years ago

    Serves em right for screwing over the g1 users so. Ha ha no one will buy their g2 drives now!

    • Krogoth
    • 13 years ago

    I sense a butthurt storm coming from people who forget why it is called “bleeding edge”.

    • indeego
    • 13 years ago

    Releasing HDD firmware[s] is fairly lose/lose in the pro/consumer space. If it works, great, but if it doesn’t work, even for users that do everything right, you catch hell for it and your reputation suffers greatly.

    My guess is the users reboot before it was done loading the device drivers, or some form of interruption of the process/timeout… Either way Intel is in for a world of paing{<. <}g

    • Shining Arcanine
    • 13 years ago

    I decided to post something more productive in #4.

    • Shining Arcanine
    • 13 years ago

    Weren’t there tons of warnings of possible data loss in the documentation for this tool?

    • OneArmedScissor
    • 13 years ago

    …yet.

    DUN DUN DUN DUN!

    • Jigar
    • 13 years ago

    Someone at Intel pulled the trigger faster than he should have…

    • Shining Arcanine
    • 13 years ago

    It wasn’t a problem for me.

Pin It on Pinterest

Share This

Share this post with your friends!