I'm one of the many that reported the failures and, as many of the voters, I believe, I failed to notice that I should have explained the failure. It was a BFG 8800 GT that died without explanation after only four months of use. As it is in a PC that I built for a friend, which lives in Brazil, getting a replacement through the warranty is going to be a nightmare.

I'm one of the many that reported the failures and, as many of the voters, I believe, I failed to notice that I should have explained the failure. It was a BFG 8800 GT that died without explanation after only four months of use. As it is in a PC that I built for a friend, which lives in Brazil, getting a replacement through the warranty is going to be a nightmare.

i have had MULTIPLE 7600 GO failures in an Alienware laptop.........

At one point the G8x desktop and laptop numbers were going up at the same rate. They could have at least been less obvious about it. O_0

I'm sure the entire staff over at the Inq voted multiple times. :-P

I actually have a 8800GT, which is now populating the wife's machine, and with which I've had no failure to report. I voted that way in the poll.

But really, why should ATi bother stuffing such a poll as this? I mean, the publicity generated by nVidia's own public announcements in that regard are far more detrimental to nVidia's fortunes, wouldn't you agree? I don't see how stuffing this poll might possibly improve on that, at least for ATi...;)

I myself reserve the word 'stuffing' for a person or group doing deliberate multiple votes to get results they want, the behaviour of people using 'funny' or random single voting in polls I would not call 'stuffing', that I call 'the freaking internet' ;)

Umm, if I remember correctly, The desktop chips were ahead of the laptop chips in regards to failure rate according to the poll. So if there was stuffing then all polls had to increase at least in the same pattern. But this poll here showed that the desktop chips were failing more often. So if nvidia aknowledged a problem with laptop chips then it's clear desktop chips also have a problem and there is no stuffing involved. I was one of the people who had problems with two 8800GT's out of 3 that I bought, was too lazy to make a post because I had to register. Both were XFX. First one had crashing problems. The second one had overheating problems and design flaws, like the mosfets on the wrong side. That card ended up frying in 2 weeks. Now I have one from EVGA and it's running ok.

I am totally against this conclusion about stuffing. Most people were just too lazy to post something cause of logging in.

I'm a little surprised at the conclusion you draw.

Sure, there is some stuffing, but there also is no basis for assuming that the stuffing has significantly canted the results, particularly since Nvidia has stated that there is a problem.

Look at the two largest categories: that the people with Nvidia cards had no problem, and that the people did not have those cards. If the poll was being dominated by box-stuffers, you would have had more votes about people with problems and very few about people who didn't have the cards.

what about the iPhone ?

why isnt here any news about is worlwide woes ?

Here's me thinking everything posted on the internet is reliable...

I don't think is stuffed, but I wouldn't call these polls "useful".

In a poll about a failing part, almost everybody (everybody that see the poll, I mean) with a faulty one will vote, because they're angry and need to do "something", but that's not the case if you have a good one. In that case, you'll probably not bother about it. Laziness, people..

In my experience (8800 GTS 512), It works ok, and I didn't vote but I wouldn't buy it again. When used as DVD player PQ is awesome but, sometimes, it displays red horizontal lines :-(.

Dude -- a POLL is not a SURVEY. You were trying to do a survey in the same space as a poll -- not going to work. If you were to announce the survey as a TR article and feature it on the front page, now that would get the attention it would need. :)

Somehow I missed this one... However, I did have a G92 - specifically a Palit 8800GT Sonic 1GB - fail completely in less than two weeks after purchase. The card in question was one, of two, in an SLI setup. At first I didn't realize it had failed. I did get a BSoD, but the system restarted so quickly I wasn't able to see the cause. After that my performance dropped but I was convinced that was due to "other factors". When I did realize that "SLI" was no longer functioning, I tried all kinds of things because I could not believe a part that new could just die.

There just aren't any conditions that would contribute to this failure. My power supply has adequate voltage, and amperage. It is also a high efficiency unit that is SLI Certified. Plus my case, a Coolmaster Stacker 830 Evo has plenty of cooling. Even for 2 hot video cards. And the system is attached to an APC Un-Interuptable power supply. So it wasn't a dip in the grid voltage that caused this part to die.

There is a possibility that I did vote in this poll. But I didn't have time to comment. That's also within the realm of possibility for others, don't you think? So even "if" you don't have a true "representative sample" (among TR readers in general). That doesn't mean there isn't a problem. The one occurrence that could be most telling is if several of Nvidia's partners start to discontinue product lines corresponding to those chips.

All someone like me can tell you is that: "Yes I had a G92 fail". I just hope the replacement, and the other original card continue to work.

I voted (once) for the "I don't have one " option.
Can't TR correlate the votes with IP addresses to see if there was stuffing?
I guess I'm one of the biggest ATI fans here but I seriously doubt that TR guys would try and stuff up the poll and there's hardly a plethora of die hard ATI fans here anyway.
Surely the best way would be to open a thread in the forums and require each vote to have an owner.
I tried to do this with the"So now you actually own Vista....how do you like it thread?" that I posted a few months back.
It seemed to work fairly well doing it that way and results seemed believable.

Oh, as for the requested totals, I had 4 8600GT go through RMA and one G92 show up DOA, but the current G92 is running strong and the last RMA'ed 8600GT was given as a gift and seems to be mostly OK, though now starting to cause display driver lockups and odd corruption/failures. I think it might drop again now that the 1 year warranty is over.

Thanks for these wonderful 8600 cards, NV. They were slow and cheap when you made them, and now they gleefully die young. I kind of expected it due to the Ti4200 meltdown extravaganza, though. NV's low end = OMFG FIRE.

I meant 8600 GTS .. all I proved was that NVIDIA's naming convention has destroyed my brain.

Yes, NVIDIA, it is all your fault ... and not the fault of my lovely friends, the bottles of beer.

It was a Sparkle 8600GTS btw. It worked for a week, then that's it, no picture, no nothing, it didn't even show up as attached hardware (luckily the motherboard had integrated graphics as well).

I have a 8800 gts and have been running it nearly nonstop since I got it (about a year ago) with no problems.

108% of the 36,189,185 ATi nvidiox G200 4D graphics adapters I've purchased since September 1863 have failed because of the G84, G86, G92, or G94 manufacturing defects, whatever that means (I didn't bother reading about any of this, I just like typing in boxes that appear on the internet).

The majority of these failures were fictional or imagined but I posted in the poll anyway based on my religious beliefs.

I voted on the poll.

I'm in the process of RMAing my GeForce 9600GT. (G94)
It consistently locks up during gameplay for brief intervals.

Note: completely unrelated to "stuttering" from loading graphics assets for the first time.

99% of the 8600GT cards we've sold turned out to be faulty. . .

I've been running a 6800 GT for 3 years OCed with stock fan and no complaints. My friend has had two 8800 GTX for about six months without a single problem or with dual 7900 in his Alienware laptop. So I haven't seen any particular failure on Nvidia parts.

What would be more useful to know is if mobile and desktop GPU dies are manufactured on to the same package then separated in to mobile or desktop, or if they are selected pre-packaging for their destined role. In this case the packaging is what failed, apparently after heat-cycling stress which would take longer to show up on desktop parts, not the GPU itself so if I could get this answer out of NV I would.

Well now we know where some of the votes came from...
Quote from Beyond3d member "weeds"
I voted 6 times just by switching between IE and Firefox
http://forum.beyond3d.com/showthread.php?t=48843&page=15
..... and more from Chris Ray
Edit* Heh I just tested it. I was able to vote 5x. And probably could do more if I felt compelled too. Pretty easy thing to manipulate. __________________ Nzone/SLIZone: SLI Forum Support Mod

Shouldn't the new poll in this article have a "totally lame" option below the ATI choice?(which is pretty lame in itself imo)

Can't you limit it to FORUM MEMBERS only?

Just out of curiosity, does TR have any actual evidence that the ballot has been stuffed, or is TR simply unhappy with the lack of commentary?

ATI! (393 votes)

I see what you did thar

AWESOME ! I got mentioned on the front page, AGAIN :D
(This time in a more constructive manner :))

OK now onto serious business. The first time the XPS1330 failed, the video output would become garbled randomly up to a point where the screen was just black or the laptop refused to boot. The mobo got changed and after a few months, a similar thing started happening only this time the screen was showing corruption with pink line everywhere. Dell replaced the mobo once again and right now it's running fine.

Adi

You want details on Nivida issues?

8800gtx here. Vista X64 Ultimate. Dual monitors. When watching videos or playing games, both monitors would go black and eventually go into standby mode as if the computer had been turned off. All while the sound would still play for a few minutes. That is until the whole system would freeze and sometime reboot. Upon restart, Vista always informed me that Nvidia driver caused the issue.

Video card nor system was overheating. Video card was factory overclocked though I didn't do anything more. A friend with the same exact 8800gtx experienced the same exact symptoms on his system when doing anything video related. Doing massive amounts of research indicated that for some reason, a new feature in Vista (TDR) would be tripped. Although it's supposed to bring you back and in our case, it would just cause the whole computer to reboot.

If you want to know about nvidia issues, just go ahead and google "TDR" and "Nvidia" and you'll be filled in on what you are looking for.

I thought this was going to be the result just as soon as the poll went up. TR style polls show the results immediately after voting. Any half-decent fanboy will occasionally see reality diverging from his chosen company's prescribed version and will then strive mightily to bring that pesky reality back into line.

In other words, any NVidiots would see any significant number of failure votes as a problem and begin wildly stuffing the 'NO FAILURES HERE KK' option as fast as their anonymizer of choice could refresh. Likewise the ATIdiots (AMDrones just doesn't sound as good) would want those failure rates to exceed 101% and thus start stuffing the 'OMFG FIRE' option.

There was no way this could work out well. I'm betting vote totals would have exceeded the number of unique IPs hitting TR inside a week.