August 12th, 2008
Are NVIDIA G92s and G94s failing prematurely too?
The Inquirer is reporting that in addition to the NVIDIA G84 and G86 GPUs being bad and failing due to thermal issues, the G92s and G94s are also affected.
If you haven’t been following the story, the short version, up till now, is that all G84 and G86 chips are bad. Nvidia is blaming everyone under the sun, but denying they have any hand in the failures. While this may sound plausible, technical analyses by people intimately involved in the requisite semiconductor technologies tell The INQ that it is a bunch of bull: NV simply screwed up. Badly. If it was a problem with the suppliers, NV would not be paying out more than the chip cost, much less gagging OEMs: it would simply be passed along.
…
A little digging revealed what this, and more, is all about, and it’s far uglier than just the ‘notebook’ version. It seems that four board partners are seeing G92 and G94 chips going bad in the field at high rates. If you know what failures look like statistically, they follow a Poisson distribution, aka a bell curve. The failures start out small, and ramp up quickly - very quickly. If you know what you are looking for, you can catch the signs early on. From the sound of the backchannel grumblings, the failures have been flagged already, and NV isn’t playing nice with their partners.
Hmmm … G92 GPUs are used in 8800GT, 8800GTS, 8800GS, and some mobile versions of the 8800, while the G94 is the GPU powering the 9600GT. These, in addition to the G84s and G86s, amount to a LOT of GPUs that are potentially affected here.
What I’m seeing is stonewalling all round. Beyond the statements that have already been made, companies aren’t willing to go any further to help clarify things. I’ve had a number of people come to me off the record but proof, in the way of documentation or hard numbers, isn’t forthcoming. However, this is to be expected. Contacts inside the chip industry I’ve spoken to say that it might take NVIDIA months to get to the bottom of this problem and isolate the true cause of any failures.
However, if you prowl around support forums and tech help sites (as I do), you can’t help but notice a huge increase in what seems to be GPU-related issues (screwed-up display, jumbled-up display, lines and other artifacts on the screen, or no display at all). There’s also more than a hint over on Apple’s support forums that the MacBook Pro, which has a 8600M GT GPU, is also affected by this thermal issue.
I think that the biggest issue here is that there hasn’t been a clear, concise statement from anyone as to which GPUs are affected, and that is making owners uncertain. Over the past few weeks I’ve had dozens of emails from people which go something like this:
I own an XYZ notebook which I think is fitted with a 123456 GPU. Is my system going to do a China Syndrome on me?
Truth is, I don’t know. One thing is for sure, customers are getting tired of being kept in the dark and aren’t too happy about being offered a software patch to fix what seems like a hardware issue, and to be honest, I can understand that. If I’d shelled out $1,000+ on a notebook and discovered that the GPU was potentially dodgy and it could go dead at any time, and all I was being offered was a BIOS patch, I think it’d get pretty emotional about it.
I feel that this story has a long way to go before it plays out to a conclusion.
Background to this story here, here, here, here, and here.
Adrian is a technology journalist and author who has devoted over a decade to helping users get the most from technology. He also runs a popular blog called The PC Doctor. See his full profile and disclosure of his industry affiliations
Want to get in touch? Got a tip? Feel free to drop me a note! I ALWAYS respect anonymity. I'm also on Twitter (@the_pc_doc)
Right to Reply: Should any industry representatives wish to comment on any posts on Hardware 2.0, I will be happy to publish their reply verbatim on this blog.
Subscribe to Hardware 2.0 via Email alerts or RSS.
| 08/13/08
