http://www.tgdaily.com/content/view/42472/118/
| Quote : The filing then goes into chapter and verse about Nvidia customers. "The chip claims arise out of allegedly defectively designed Nvidia chips G86, G86A2, G84, C51, G72, G72M, G73, G72A3, MCP67 and NV42." On August 6th last year, Nvidia contacted NUFI about a claim by Dell, demanding a defense and idemnity against claims against Dell.
|
and that just the ones they know about
Is your thread biased as in fanboyism not for nvidia?
I'm just saying this because the thread's title is somewhat agressive and "nearly" generalist!
I was just about to get a GTX285 2GB, is that defective too?
| maximiza wrote : I was just about to get a GTX285 2GB, is that defective too? |
Maybe, maybe not, the G92 & GT200's have not appeared on the list (yet) and as I have a nearly two year old 8400GS (G86) that is used each and every day and has yet to show any hardware faults, I'm not that bothered with this latest round of whining.
| maximiza wrote : I was just about to get a GTX285 2GB, is that defective too? |
No, but I would not spend extra cash on the extra 1gig, it is better spent in other areas. (You could get a factory OC card for instance).
Back on topic though. Will be interesting how this affects the companies image. Frankly, as big a deal this may be, it does not get much attention from reputable media. I'm not really sure how bad it actually is... But everyone has bad chips now and then.. Nvidia will probably pay the damages, avoid a larger spectacle, and focus on the future (if they are smart).
They obviously supplied more than an acceptable number of faulty parts to their partners, but I have not seen hard nubmers on how many that is, how many actualyl got through the customers QA and into consumers hands, and how many faults would be required to claim substantial damages.
Its only notebook GPUs not desktop GPUs, to attempt to imply that an issue with notebooks parts might effect all GPUs is stupid and trolling. Yes a large number of Nvidia notebook GPUs had an issue, but no where near the majority, mine is still chugging along fine and so are hundreds of thousands of others.
There are desktop cards that use G86, G84, etc GPUs as well, it's not just notebooks. It's just that notebooks generally have less than adequate cooling which exacerbates the problem.
| turboflame wrote : There are desktop cards that use G86, G84, etc GPUs as well, it's not just notebooks. It's just that notebooks generally have less than adequate cooling which exacerbates the problem. |
And more power up\down cycles.
Sorta like weak nichrome in a lightbulb. The more you turn it on and off, the more likely itll burn on the next on switch. They basically are all the same cards, to an extent, its just the usage that can make them fail, or, the type of usage
Has anyone seen hard nubmers on the failure rate for the consumers?
How many bad GPU's got through the QA of HP and the other manufacturers?
I'm not entirely sure if this is a larger company relations issue with them having to replace a lot of parts on the line.. or a loss of customer satisfaction due to the majority of the parts gettign to end users. I'd imagine they would claim damages for both.
Its more a long term testing situation, which the OEMs do little of, if any. This is a nVidia problem. Theyre the ones who should have caught it, and the OEMs are running off nVidias findings, which somehow, they failed at
Also Nvidia are still refusing to say what the problem is exactly, but they do know precisely how much it's going to cost to fix, at least that what their insurers are saying.
there's a lot of the 8800GTs braking
If I recall, the 8800GT's were the hottest running nVidia 8 series cards
still they should be designed to take the heat, if there not you could put that down at nvidia's door
They may be designed to take the heat but you may recall the G92's have a problem with the fan whereby it doesn't spin as fast as it should at higher temperatures. I have a 8800GT and over the last 5 months it's has started to artifact due to the damaged caused by heat not being dissipated by the fan. To get around the problem I use Rivatuner to override the on-board fan controller and manually set it to 90% before I load a 3D graphic intensive game.
^^is it heat or bad tech by nvidia, i would go with bad tech (bad under fill) but its not me thats buying crap its you
The notebook GPU issue is related to IC-Packaging. The GPU Packaging for notebook has a reliability-defect. The package can't dissipate the heat off the DIE resulting to delamination and or DIE-CRACKING.
If you have an affected notebook it will eventually fail specially if run application that make the GPU consume more power. Its like a death setence or cancer.
The OEM or PC-companies including NVIDIA recommended patches that speed-up the fans and turn-it on most of the time on affected notebook. The impact of this is..1) It will hide the problem and or extend the life of the GPU just beyond the typical warranty period of 1 year. 2) With the fan on more than what it should be then it drains the battery faster.
I returned 2 HP notebooks affected by this problem. I run 3d-Apps for extended period and one of my notebook started displaying vertical and horizontal lines.
Its a real issue and the best solution is RETURN YOUR NOTEBOOK and get a replacement with GPU not-affected by this problem.
DESKTOP GPU use a different packaging technology so in most cases its not affected. From my end i have not read information that the problem affecting desktop video cards. I can be wrong.
ITS NOT THE DIE that is BAD.... ITS THE PACKAGING THAT EVENTUALLY KILLS THE DIE.
I have a 8800GT OC 512 that i got since nov-2007.... Its still humming like a charm. Two months ago i got a GTX-280 OC... The best card that i had to date (HW/SW).
TECH-JARGONs:
DIE- The silicon integrated circuit, PACKAGE: The balgrid package that encapsulate the DIE . It provide electrical interface and isolate the DIE from the outside world.
We all know its the package, but what we dont know is whether its exclusively mobile only. The bumps are bad, manufactured during a particular time period, and yes, the bumps are bad, but who knows which ones were made this way, and which werent, and also the fact that heat (contraction/expansion) will/does play a part in this whole thing
Quote: "When I wrote the article, I had the suit and the two attachments. The suit has most of the relevant insurance clauses in it. Exhibit 1 and Exhibit 2 are the full insurance policies, scanned, including signatures"
http://forum.beyond3d.com/showthre [...] 43&page=29
Charlie, explaining that the insurance company has full reason not to pay for nVidias faulty gpus, including many types of cards
There are 634 identified and unidentified users. To see the list of identified users, Click here.
You are about to answer a thread that has been inactive for more than 6 months.
If you still wish to proceed, please ensure that your posting is original and does not duplicate or overlap any prior responses to this thread.

