Forgot your password?
typodupeerror
Bug Hardware

Nvidia 55nm Parts Are Bad Too 372

Posted by kdawson
from the or-so-they-say dept.
JagsLive sends in a story (in somewhat inflammatory prose) from The Inquirer, which links to many others; they have been following developments in the alleged NVidia quality "fiasco" for some time. "Hot on the heels of its denials that anything is wrong with the G92 and G94s comes another PCN [Product Change Notification] that shows the G92s and G92b are being changed for no reason. Yup, the problems that are plaguing G84 and G86 are the same that affect seemingly all 65nm and now 55nm NVidia parts ... It is hard to overstate how bad this is. Basically every 65nm and 55nm NVidia part appears to be defective ... We are hearing of early failure rates in the teens percent for 8800GTs and far higher for 9600GTs ... To make matters worse, NVidia has a mound of unsold defective parts that they are going to bleed out into the channel along side of the (hopefully) fixed parts. As a buyer, you have no way of knowing which one you are getting ... Until NVidia comes fully clean on this fiasco, lists all the defective parts, and orders boxes clearly marked, you can't say anything other than just avoid them. Then again, since doing the right thing would likely bankrupt them, we wouldn't hold your breath for it to happen."
This discussion has been archived. No new comments can be posted.

Nvidia 55nm Parts Are Bad Too

Comments Filter:
  • by bigtallmofo (695287) * on Friday August 29, 2008 @12:29PM (#24796513)
    Sure, the GPU might be faulty but the rest of the components on their graphics cards (cooling fan, PCI-Express connector) are not showing any issues.

    So let's not blow this out of proportion.
  • 8600GT? (Score:3, Interesting)

    by Fallingcow (213461) on Friday August 29, 2008 @12:30PM (#24796521) Homepage

    At risk or not?

    Also, this sounds like a class-action waiting to happen.

  • Pizza (Score:5, Funny)

    by Sockatume (732728) on Friday August 29, 2008 @12:30PM (#24796523)
    This is the kind of story that can only end with somebody being fired for making pizza in the silicon fab oven.
    • Re: (Score:3, Funny)

      by Yvan256 (722131)

      Actually it was Kramer warming up his clothes in the silicon fab oven.

    • Re:Pizza (Score:5, Informative)

      by Dogtanian (588974) on Friday August 29, 2008 @03:00PM (#24798847) Homepage
      I remember reading something not entirely dissimilar in Robert X Cringely's "Triumph of the Nerds". Might or might not be apocryphal; I don't have the book to hand. Apparently Intel (IIRC) were having problems. The amount of defective parts they were getting was going through the roof, and they were pulling their hair out trying to get to the root of the problem.

      Finally they traced it down to the guy responsible for receiving the deliveries of the silicon wafers. Apparently he was taking out the wafers and putting them down in his desk- quite dusty and very definitely *not* up to clean room standards!- to make sure Intel was getting what they'd paid for.
  • Fair and Balanced? (Score:4, Insightful)

    by wombatmobile (623057) on Friday August 29, 2008 @12:32PM (#24796551)

    "Then again, since doing the right thing would likely bankrupt them, we wouldn't hold your breath for it to happen"

    -5 Troll

    • Re: (Score:2, Funny)

      by Anonymous Coward

      Well, of course you wouldn't hold my breath, only I can do that.

    • by bogie (31020)

      Maybe so, but to an outsider they do have a lot of interesting statements about the changes Nvidia made to their hardware recently. Either the 8-K statement is business as usual in the chip market(I'm not expert on this) or it is an indication that larger problems loom. We'll know for sure if over the next month or so we start seeing massive amounts of failing gpus.

      http://biz.yahoo.com/e/080702/nvda8-k.html [yahoo.com]

    • by AcidPenguin9873 (911493) on Friday August 29, 2008 @12:58PM (#24797047)
      The person who submitted this story to Slashdot left out an important link [theinquirer.net] on that text from the original Inquirer article (linked again here [theinquirer.net] for your convenience). In the original story, that sentence reads:

      Then again, since doing the right thing would likely bankrupt them [theinquirer.net], we wouldn't hold your breath for it to happen.

      At that link, you'll find The Inquirer's (however flimsy and speculative) financial analysis of a full-scale Nvidia recall of the bad parts.

      The Inquirer doesn't and has never claimed to be a fair and balanced news source, so they are free to put these sorts of quips on their stories. People there are pretty knowledgeable, and appear to have connections and sources in the industry, which is why people keep reading The Inquirer and don't really complain about stuff like that.

      • Re: (Score:3, Insightful)

        by Keith Russell (4440)

        People there are pretty knowledgeable, and appear to have connections and sources in the industry, which is why people keep reading The Inquirer and don't really complain about stuff like that.

        *cough* Rydermark! [wikipedia.org] *cough*

      • Re: (Score:3, Interesting)

        by darkwhite (139802)

        I don't understand why a full recall would be needed. Most cards in desktops are completely stationary, which drastically reduces the likelihood of failure due to interconnect stress, and many are not even thermal cycled often (e.g., computer always on, no stressful games), which is the other big source of physical stress on the interconnects. If the failure rates are in the teens, the reasonable solution is to extend the warranty to cover all instances of interconnect failure e.g. 5 years from sale, which

    • Re: (Score:2, Insightful)

      by Anonymous Coward

      I don't think anyone, including The Inquirer, has ever claimed The Inquirer was "Fair and Balanced".

      And seeing a comment like that on Slashdot makes me think we have pot kettle situation here :)

  • Intel... (Score:5, Insightful)

    by faragon (789704) on Friday August 29, 2008 @12:33PM (#24796575) Homepage
    ...to buy Nvidia? Problem solved.
    • by faragon (789704)
      It is paradoxical, I said almost the same four months ago (@20080417 [slashdot.org])... and partially taken also as "funny", however, I was also speaking seriously.
  • by Sj0 (472011) on Friday August 29, 2008 @12:33PM (#24796579) Homepage Journal

    If you're a betting man, now's a good time to pick up on Nvidia stock.

    The question is, do you feel lucky, punk?

    • Re: (Score:3, Funny)

      by martinw89 (1229324)

      I'm not exactly going to pick up NVDA [yahoo.com], but I may sell it shorter than the 55nm process used on these chips!

    • by Cheeko (165493)

      Or you could short the stock if you think word will eventually get out and they'll be forced to take action.

    • by szquirrel (140575)

      If you're a betting man, now's a good time to pick up on Nvidia stock.

      The question is, do you feel lucky, punk?

      Absolutely. Nvidia is getting hammered lately but they aren't stupid and they aren't poor. They have $1.6 billion on hand to weather this storm.

      I wouldn't be surprised if they sold off their chipset business to refocus on high-end, high-margin GPUs, but they aren't done yet by a damn sight.

    • Re: (Score:3, Informative)

      by kesuki (321456)

      actually, wall street hasn't yet factored in the possibility of 20% of nvidia's high to mid end chips being totally reject chips yet.

      so a betting man would watch the stock closely for the next few weeks, then when it bottoms buy massive quantities of stock.

      this is the kind of a massive chip recall scenario which makes nvidia a likely buyout target by say Intel (everyone likes buying a company at a fraction of the value of the company, which is why M$ worked so hard to try and take over yahoo)

      for those sayin

  • I've always secretly been an ATI fanboy... and a traitor since the 6800GT.

    Now, I've got ATI again but recommended everyone I know (up until 48XX by ATI) buy the 8800 or 9600....

    I wanted ATI to regain some track to even the market... but this is a little much. Complete flops are not good for competition either.
    • by Anonymous Brave Guy (457657) on Friday August 29, 2008 @01:04PM (#24797161)

      IMNSHO, the quality control at both companies has been terrible for several years now. What's the point of paying a premium for a good graphics card, if hardware problems make your system unstable as soon as a demanding game is loaded or the drivers take out your operating system at random intervals? It's not like this has happened only on bleeding edge cards with new drivers, either: several entire models have had basic incompatibilities with other common system components, and sometimes drivers have been unacceptably poor for the entire useful lifetime of a gamer's card.

      Contrary to the marketroid reports, it is not in any way unavoidable that new cards with new drivers have to crash a significant fraction of the hottest games at release time. It's not like these kinds of problems are subtle and might be missed during a decent period of testing, and it's not like the card vendors couldn't co-operate with the game vendors on a beta test programme. This happens because commercially, it makes more sense for them to race to market with inadequately tested hardware and poorly engineered driver software and hope they can patch up any widespread problems later with a minimal PR hit. As long as both the big names are as bad as each other, consumers in the target market are pretty much screwed anyway.

      It's about time something like this happened and one of the companies took a major financial hit as a consequence. Perhaps then we'll move back towards supplying hardware and drivers that actually, you know, work. Gamers the world over (other than those currently suffering from these problems, of course) should probably be happy about this, because it might be serious enough this time to make a difference to future quality control, which is much better than a significant fraction of people being disappointed with each new model but never enough of a critical mass to really punish the company that supplied substandard kit.

      • by Fred_A (10934)

        IMNSHO, the quality control at both companies has been terrible for several years now. What's the point of paying a premium for a good graphics card, if hardware problems make your system unstable as soon as a demanding game is loaded or the drivers take out your operating system at random intervals? It's not like this has happened only on bleeding edge cards with new drivers, either: several entire models have had basic incompatibilities with other common system components, and sometimes drivers have been unacceptably poor for the entire useful lifetime of a gamer's card.

        If they stopped putting out new hardware like mad every other week in order to outbest one another, they could just possibly have time to fix their hardware and drivers.

        Just a thought.

      • by Splab (574204) on Friday August 29, 2008 @02:13PM (#24798213)

        No, you are just most likely American. The problem for you guys is as a customer you have next to no rights, very limited warranties removes the need to make proper products.

        In EU however, most places now require 2 year warranty, that means all the good hardware is being shipped here because they don't want to be stuck with a customers problem 1,5 years down the road.

      • Contrary to your belief that 'these kinds of problems are subtle and might be missed during a decent period of testing' it can be EXTREMELY difficult to find these kinds of problems. Beyond your wildest imaginings difficult.

        Having worked on high performance hardware/software systems as an engineer I can tell you from first hand experience that the situation is more like there are 999,999,998 ways for things to go wrong and about 2 ways you can get it right, and those 2 ways are not AT ALL obvious. Usually t

  • Basically, the meat packing industry's favourite tactic is that when some contamination, like salmonella is found in the meat they will just sit on it. If the media gets wind of the story, they'll voluntarily recall a small fraction of the meat unfit for human consumption.

    At least Nvidia by doing the same thing is not directly endangering human lives...
  • by RetiefUnwound (472931) on Friday August 29, 2008 @12:39PM (#24796689)

    I've got a two 8800 series cards (one 8800GT, one 8800GTS), and I live in a place with no air conditioning. If these cards were subject to heat failure the way the Inquirer has been hollering about - one or both would have died by now. Particularly the one in my wife's computer - it's a Shuttle box, which runs toasty. It's been rock solid, running 24/7 for more than a year now.

    I'm not suggesting there is NO problem - but the Inquirer has been talking about this like all of these cards are just waiting to die. With no A/C, and temps in the house above 90F during the summer, they should be dead if the Inq is to be believed. Perhaps I'm just lucky, but I still aint buying the story.

    • by neo8750 (566137)
      I to own a 8800GTS and a 8800GS neither of which have any problems and both of the boxes are on 24/7
    • by mandark1967 (630856) on Friday August 29, 2008 @12:47PM (#24796849) Homepage Journal

      Actually, if your wife's computer runs 24/7 it would be less susceptible to the problem since the problem they are experiencing is exacerbated by thermal cycling. fewer cycles of heating up and cooling down = less of a chance to fail.

    • by Curien (267780) on Friday August 29, 2008 @12:50PM (#24796915)

      They say failure rates are "in the teens percent". Figure 20%, just for kicks. That means your chances of either card failing is 1 - (1 - .2)(1 - .2) = 36%.

      For some reason that I don't understand, the vast majority of people have innate misconceptions of the rules of probability.

    • Re: (Score:3, Informative)

      by Cheeko (165493)

      Well for starters they said its a failure rate in the teens. The odds even with 2 cards that 1 would fail is still less likely than not.

      Also the 8800 cards have been out for a while. The impression I get is that this is a newer issue with the cards, so initial 8800 cards might not be an issue.

    • by Tridus (79566) on Friday August 29, 2008 @01:06PM (#24797207) Homepage

      This sounds like what Xbox fanboys were saying when word of the 360 being defective reached a fever pitch.

      "Well, MINE works fine, you all must be using it wrong!"

      Course a week later Microsoft admitted to it and shelled out $1 billion.

  • A Grain of Salt (Score:5, Interesting)

    by Colonel Korn (1258968) on Friday August 29, 2008 @12:40PM (#24796701)

    I don't get people who show any sort of devotion to a GPU manufacturer. I just don't. The author of this article is one of them. That doesn't mean it's not true, but he's written a number of articles that later proved to be completely false in the past, for instance saying that the 8800 series would doom nV because of low performance and high power usage compared to the 1900 or 2900, whatever ATI was releasing at about the same time. I'd suggest you not take any article written by Charlie seriously until it's been confirmed (not just repeated, as often happens) elsewhere.

    • Re: (Score:3, Interesting)

      by bughunter (10093)

      Yea, but a lot of people show devotion to their notebook computer manufacturers, who hardwire these chips onto the mobos.

      Like, say, the Apple MacBook Pro and its nVidia GeForce 8600M GT [lowendmac.com], for instance.

    • by eebra82 (907996)

      I don't get people who show any sort of devotion to a GPU manufacturer. I just don't.

      It's just like buying a car, being satisfied with the purchase and put the car maker in favor of others when you consider your next purchase. I hate car analogies, but this one fits.

      On the other hand, there are different levels of devotion. Some people explicitly hate one of the companies and favor their products even when it is not maintaining a lead. That kind of devotion is just silly, unless you have an uncle employed there.

  • Charlie Demerjian (Score:5, Informative)

    by Qhartb (1311541) on Friday August 29, 2008 @12:40PM (#24796709)
    I stopped reading when I got to "By Charlie Demerjian."

    Seriously, this guy is to NVIDIA as Jack Thompson is to video games. It's just not as common knowledge that you shouldn't take him seriously.
    • by unsigned integer (721338) on Friday August 29, 2008 @01:14PM (#24797335)

      Could someone explain just what exactly the chip on his shoulder is?

      Did a nvidia graphics card molest him as a child?

    • by Sockatume (732728)
      He also has a huge chip on his shoulder about Vista and a strange tick of exclusively referring to it as "Windows MeII", as in "Windows Millenium 2". Except in their font it is indistinguishable from "Windows Mell", as in Windows M[e]llenium. He also refers to Windows 7 as "Windows Mell SP1a", from his insider knowledge that "WINDOWS 7, AKA Me II SP1a, is [...] simply a warmed over Vista", furthering confusion as he discusses the inevitable failure of the first service pack of an operating system from almos
  • by nimbius (983462) on Friday August 29, 2008 @12:41PM (#24796715) Homepage
    has been proposed:

    buy ati.
    • buy ati

      Yeah, someday they might actually release the Glaze3D [wikipedia.org] video cards now that they own BitBoys Oy. Hopefully just in time to be able to play DNF on them.

  • by Animats (122034) on Friday August 29, 2008 @12:42PM (#24796743) Homepage

    Why is NVidia using lead-based solders at this late date? The European RoHS deadline for lead-free components was back in 2005. The NForce and 8800 parts were RoHS compliant years ago. Are these NVidia parts even exportable to Europe?

    • by XanC (644172) on Friday August 29, 2008 @12:58PM (#24797035)

      I would say it's because lead-based solder actually works properly, but according to this story that doesn't seem likely to be their motivation.

    • rohs has exceptions for very fine pitch stuff iirc.

      • by Tycho (11893) on Friday August 29, 2008 @04:15PM (#24800157)

        Yes there is an exception in RoHS for lead solder that has a high melting point. However, the official RoHS rule is that while lead solders in general are prohibited, there is an exception allowing for the use of lead solder that contains at least 90% lead. The idea being that solder with at least 90% lead melted at a higher temperature and was at least somewhat safer if disposed of improperly. Otherwise, potentially there may also have been no replacements for high lead content solders that performed as well when the first RoHS directives were drawn up in 2003. Currently (2008), however, there are lead-free solders that would work, but the lead free solders are more expensive than lead based solders (by roughly three times). Using a lead-free solder with a significantly different composition may also require a new packaging design and another extensive round of qualification, too. I am not totally sure how this would be done.

        It get worse, the new solder to be used by nVidia mentioned in this Inq article states that it will only contain 63% lead and 37% tin, making nVidia based cards with this solder not saleable to consumers in the EU according to RoHS directives. The replacement 63Pb/37Sn solder has a somewhat better tensile strength and a lower coefficient of thermal expansion than the older 95Pb/5Sn solder, which may be why nVidia chose this route to fix the problem. Whether nV will be selling very many products in the EU with this fix and whether this will correct the problems, is another issue.

    • by gweihir (88907)

      Are these NVidia parts even exportable to Europe?

      Very likely not.

  • by Ecuador (740021)

    Even though I am an ATI fan (obviously I want nVidia around to drive competition), and I have seen the Inquirer pull off great reports in the past, I still take this with a grain of salt.
    First of all, I am not aware of any panic about failed products in the various fora. In cases such as the Red Ring of Death, the Deathstar etc you could not visit a tech forum without having hundreds of people complaining. So maybe the problem is not that big.
    Also, the article centers around the fact that they switched from

  • Yeah, because the Inquirer is such a steady and accurate news source.

    I'll believe this when I see more proof.

  • Okay, i hear about supposedly deffective nvidia GPUs all the time now, but why are we not seeing forums crowded with people with these failed graphics cards? i believe this issue is being overblown substantionally out of it's actual proportions.
  • by Ungrounded Lightning (62228) on Friday August 29, 2008 @01:01PM (#24797109) Journal

    To make matters worse, NVidia has a mound of unsold defective parts that they are going to bleed out into the channel along side of the (hopefully) fixed parts.

    This sounds very similar to what finally took down Weitek, back when there were a bunch of graphics chip companies competing hotly and being shaken out if they screwed up.

    Weitek had built a very fast and powerful chip. But they had goofed: While it had the mandatory basic VGA mode for acquiring the Microsoft certification, there was a bug in it.

    QA told management that the bug was there and would fail them. But Software told them a driver could work around it and people would want the chip because it was so fast on graphics rendering. (Of course it could not - because to get the cert it had to work with the stock bootstrap stuff, before a custom driver could be loaded.)

    So they went to production with the bug. And the customers got their prototypes, found the bug, and demanded a fix. Eventually they did a fixed version - but had maybe a couple million of the buggy ones on hand and wouldn't sell the fixed ones unless the customer bought some buggy ones, too. So nobody bought and the company folded.

  • SHOULD bankrupt them.

  • by mooingyak (720677) on Friday August 29, 2008 @01:17PM (#24797365)

    It is hard to overstate how bad this is.

    This will end all life on earth.

    That wasn't hard.

  • Not widespread (Score:2, Interesting)

    by cgadd (65348)

    Based on personal experience with the 8800 GT boards, I think this problem is vastly overstated... Running 4 of them in my house, and three friends also running them in SLI config, and no failures. That's with the boards overclocked by a bit.

    Additionally, failure rates based on NewEgg reviews seem very low (and we all know people love to post a nasty review if they get a bad one).

    The cards do run nasty hot, at least until you set the fan to turn on at something under 180F.... who the hell came up with tha

    • by Knara (9377)

      The cards do run nasty hot, at least until you set the fan to turn on at something under 180F.... who the hell came up with that turn-on temp?

      Where can you set that particular variable?

  • by mlwmohawk (801821) on Friday August 29, 2008 @01:27PM (#24797509)

    The Intel 486SX was a defective 486DX who's numeric processor was dead or not working.

    Most very very large scale integrated chips have defects. Depending on the nature of the defect, they simply categorize the part differently.

    A chip is not fast enough for a high speed gaming system? Us it in an embedded device.

    Buy it, if it fails, return it. Just because nVidia has issues you know about, don't think for an instant that ATI doesn't.

  • Changing solder does not prove a chip is faulty. The parts obviously work to the point that everyone testing them had one that functioned correctly.

    There are plenty of reasons (e.g. cost, RoHS) that the change could be made.

  • by sirwired (27582) on Friday August 29, 2008 @01:35PM (#24797647)

    I am neither an NVidia or ATI fanboy (heck, my current GPU is an integrated Intel), but this article is a steaming pile of crap.

    Somehow, he takes a report of a routine running change to the production process (a new kind of solder), and magically turns this into some wild tale of how NVidia is shipping thousands of defective parts that will remain in the field.

    Completely lacking is how he corresponds the running change to some defect...

    SirWired

  • This is /. for crying out loud. We're all supposed to be geeks, right? This sounds like a solder problem; so, just re-solder the damn chip if it fails. Problem solved... or do we need to revoke a few geek cards?
  • A wolf! A wolf! (Score:4, Informative)

    by ozbird (127571) on Friday August 29, 2008 @02:35PM (#24798507)
    Charlie at The Inquirer has no credibility when it comes to nVidia.

    From TFA, nVidia is changing from high lead to eutectic (tin) solder - for RoHS compliance - and has issues a PCN to that effect. Charlie has latched onto this as "proof" of his claim that all nVidia chips are faulty and overheat.

    What Charlie doesn't explain is how switching from high-lead solder (5/95 Sn/Pb) to eutectic solder (63/37 Sn/Pb) - which has the lowest melting point of all tin-lead solders - is supposed to help if the chips are overheating. Nor does he explain how changing the solder material has any relationship to changing the underfill material on some mobile chips (other than they were both PCNs.) But hey, why let facts get in the way of a conspiracy theory/page hits?
    • Re: (Score:3, Informative)

      by Jay L (74152) *

      Good.. I thought it was just me. And I'm definitely NOT a hardware guy. But I can't see, from his description of the PCN, how switching from high-lead to tin solder could be seen as a response to, well, anything except "let's use less lead".

      I know that 63/37 has a lower melting point than 60/40, and a "sharper" one (no pasty phase), which is why I use it for audio repairs and cabling; I'm a klutz, and anything that makes my solder joints more stable is good. But I can't imagine that this matters as much

  • top banana - fdiv (Score:3, Interesting)

    by Sebastopol (189276) on Friday August 29, 2008 @03:41PM (#24799439) Homepage

    nVidia needs to take a page from Intel's FDIV days (ca. 1993) and just do a no-questions-asked recall and replace.

  • by rahvin112 (446269) on Friday August 29, 2008 @05:49PM (#24802143)

    I own two notebooks with Nvidia Chipsets in them. Both HP notebooks, one contains an 8400M the other an 8800m GTS. The 8400M notebook's cover broke at the hinge conection (a problem that was in no way related to circuit boards) last week and was sent back, just got it back today and checked on the repair slip was a note that they replaced the outside cover but they had also replaced the video circuit board. Surprise!

    Just last week the Laptop with the 8800GTS started blue screening windows with a video subsystem problem before the login prompt. Ubuntu booted without error but would freeze every 30 seconds for 15 seconds or so if you moved the cursor on the screen. HP concluded the graphics system was malfunctioning and off to repair it went. I'll know in a couple weeks what was replaced but I bet the 8800GTS gets replaced.

    This is a BIG deal people. Charlie is being a sensationalist but it's a BIG deal if HP extends the warranty on every laptop with the chips in them for an additional year. HP wouldn't do that unless they feared loss of customers or a class action lawsuit because the warranty extension costs them serious dollars. And I would also bet HP isn't going to eat every dollar. Nvidia will share the cost at a minimum. Even 10% bad parts could cost Nvidia hundreds of millions.

    Charlie might go overboard in his complaints about Nvidia but he's right about this issue, it's really really big and Nvidia will eventually talk about it because of stories like this. Without Charlie's stories Nvidia would probably try to bury the issue and pretend it wasn't happening and if I was invested in NVDA I would want to know this information because it's a harbinger of a profit warning by NVDA.

Pause for storage relocation.

Working...