what are the defective nvidia mobile gpu cards?

what are those?? all i know are 8400gs and 8600gt. i know there are more, but i can't find a complete list.

thanks.

 

nVidia was never very honest with this, therefore there is no complete list. Some rumors say it is only the 8400/8600 series are affected, other rumors state that all of nVidia's products are affected. Some rumors even state that a lot of the desktop cards are vulnerable as well.

So there is no real way to know.

 

G84/G86 core-based GPUs, based on the frequency of GPU failures, are confirmed to be defective across the whole line, but anything outside of that is just speculation. Some people claim that the problem extends all the way back to 6xxx to 7xxx series, but NVIDIA has been extremely tightlipped about the extent of its problems. Though one has to wonder if why they are so tightlipped...

If they were innocent, they would probably release records that prove their innocence, otherwise NVIDIA has something to hide. Possibly something that is such a big scandal that would bankrupt it instantly.

 

I have proof it extends back the 7xxx series, My go7200 failed within 6 months of ownership. All symptoms point to gpu failure, i managed to jump start it when i reflowed/reheat the gpu.

 

I had a 7400 go gpu that failed. I sent it in for repairs and it failed again within a month.

You're just going to have to do some thorough research or go with an ATI card that does not have this problem. Good luck with it.

 

8600m gs was one for sure.

 

I have heard this phrase several times, can you explain what it means.

 

This Inquirer article contains a long list of affected GPUs:

This is the best analysis I have read of the problem. The failure is caused by repeated temperature changes which is why many notebook manufacturers introduced BIOS updates which try to keep the GPUs cool by running the fan more and faster.

John

 

even the 7xxx series?

Go7800 in my Dell. 3.5 years and still going strong.

 

Isn't it so that heat cycles are always bad for a GPU? For example: idling for 1 hour @ 59 degrees, playing a game for about 1.5 hour @ 90 degrees, stop gaming and temperature drops to 59 degrees, after 2 hours you start to game again and the temperature raises again to 90 degrees.

I heard that heat cycles caused this big Nvidia problem, but every card is damaged by heat cycles.

 

Havent had any problems with any of these GPU's, have an 8600m gt for over a year now, (and friends that have had much longer) no problems at all, a 4-5 year 7600 go in my sony, still running - overclockable nicely, i have heard of issues but never yet experienced them.

 

I just purchased and returned a g50vt with a 9800m that had hardware failure while gaming. "The display driver has stopped responding and recovered" which is the same error a lot of the defective 8xxx series were getting IIRC

 

I get that when overclocking my hd4850, it keeps recovering itself :/ and other people can sucessfully overclock it paste the same values that i overclock at. I cant really say that 3dmark crashes while overclocking to actiate the warrenty though can i :/

 

is the 9300M G also a defect one?
coz from I've read...the chip were based on 84GS...it's just nV re-named the chip.

 

Some say all Nvidia card are defective, ranging from 6 to 9 series. In fact this is true, because all videocards get damaged by heatcycles. If you run a 'defective' 8600M GT 24/7 at full-load, nothing will hapen...but if you have a 9600M GT and let it heat up 2-3 times a day, changes are high that it will fail.

The 'defective' cards are more vulnerable to heatcycles, but every heatcycle is bad for every videocard

 

Hmm..I wonder also about the Quadro FX 1500 (G71 series?). I got an HP nw9440 workstation a few years ago which had random crashes out of the box. I eventually traced them to the graphics card overheating and insisted on an RMA and a replacement, which was fine. There were quite a few other nw9440 owners who had similar problems at around the same time, but I never made the possible nvidia connection till now.

 

Hmmm the gofx5700 in my qosmio g10 failed from the same issue... However it may have been due to the horrible thermal managment of the laptop at the time.

 

My dv9543cl has been in 3 times for graphics card failure, it has the 8600M gs

But now, HP is replacing it with a dv7

 

My old HP with the integrated 6150 Go has failed due to an overheating issue. It was sent into HP to be repaired. It's still bad because it regularly hits 90 degrees Celsius and it has gone as high as 115 degrees during normal use. Web surfing or watching videos.

In fact, the whole HP dv2xxx series just seems to run hot.

 

@ roninmedia: even if it wasn't one of the defective nvidia mobile gpu cards, with temps like that, chances are it's going to fail very soon. I'm not sure what you can do about it, but you should fix up the thermal paste, or send it in or something..

 

Well, the only exception in terms of defective cards are the G92M and G200M-based GPUs....otherwise:

8400M-G
8400M-GS
8400M-GT
8600M-GS
8600M-GT
8700M-GT

This also applies to each respective Quadro Mobile card- an example being the G84M (8600M-GT) based Quadro FX 570M that bricked my T61p.

 

I just do not buy nVidia now. While they did fix *something* in the GPU cores with the most recent products, only time will tell if they really fixed it. Makes it much easier.

 

Yup...I sure didn't know the 8600m gs was faulty until a year later.

I just hope the G94 series and onward is actually fixed. They'd have to do something, they can't just keep releasing bad chips.

 

I have the hottest running 8600M GT in all the forums and mine is still working strong After 2 years.

I seriously hope they fixed it. But since it was silicone, they probably just changed it.

 

There's nothing concrete to suggest they are all like that. Just a large batch of cards in which the company was cutting corners and using a poor solder component. They got caught and probably are being especially careful now. It doesn't necessarily mean all products dating back from Geforce GT 5xxx parts and up are bad.

Also, many people here are complaining about their Nvidia part in a HP laptop. Coincidence? I don't know but I have heard laptop sellers who also repair complain about fixing HP laptops. It's too tight in there and they are a total pain to fix. Could that have something to do with it? If the cooling element is really poor in many of these HPs because HP just wanted to cram everything in there would mean that would share some of the blame. Time will tell because HP has moved to ATI cards when it comes to discrete video cards.

 

I'd say, punish nvidia for their greed and get ati ^^

 

No coincidence. Dell and HP both released BIOS fixes to get the fan running earlier to prevent said damage fairly soon after the news was released. Unfortunately, part of the problem also lies in users. The 8600M GT/GS and 8400M GS lines were often used in low-mid consumer models, and purchased by consumers who may not know enough to clean out their vents frequently or all the little things to help keep your laptop running. This applies to both Dell and HP. If you head to the Dell forums, you'll see a large number of complaints about the M1530/M1330 lines and their GPU failures as well. Less now, since most units are either dead or working fine, but if you go back over the last year or so, you'll see it showing up.

However, HP did stonewall some owners. http://www.tomshardware.com/news/HP-Having-Nvidia-Chipset-Issues,7669.html

That said, I do think that I have noticed more complaints regarding the 8600M GS in the HPs. I'm not sure why that's the case, or if I'm merely remembering the HP cases more than I do the Dell cases.

Interesting Nvidia response.
http://www.breakitdownblog.com/repo...-being-bad-65nm-and-55nm-parts/#comment-10867

 

@dtwn those nvidias in acers (which manage to overheat even witouth the gpu) also get pretty hot. Mine shuted down frequently during gaming.

 

Oh, I know. If you like I can toss in the FX570M in the Lenovo T61p that died in the same way. There are a few users here whose T61p died due to GPU. Have a look at signatures.

 

I don't really think there's a coincidence, although HP's consumer line does tend to run hot. HP outsold Dell for the past few years, so I think there are just more HP laptops with faulty graphics chips out there. In addition, Dell did a relatively good job handling the issue, while HP took longer and rejected some customers (who probably went online and complained).

If you do look around the forum, though, you will see quite a few Thinkpad, Sony Vaio, Acer, etc users whose machines have died - and many of which have not had a favorable experience getting compensation.

 

how do you prevent the gpu from overheating? i have a nvidia card on my hp g60 laptop and the laptop gets really hot even when im not even doing gaming

do i have to take apart the laptop and put thermal paste on it?

 

What's the GPU on your g60?

 

the only GPU I see available on the HP site for the G60 is a 4500m and 4500MHD

 

help!! i put my laptop in the oven and my screen sort of melt a little and now its stuck against the keyboard but i think it cooled

i have ths one:
http://h10025.www1.hp.com/ewfrf/wc/...link=ot_we/prodlink/en_us/3795258/loc:0&cc=us

except the candian version

edit: the reason why i did that because i read somewhere that if i put it in the oven it melts the gpu back and fixes the problem

 

all of them

 

This is a case of knowing just enough to get yourself in trouble. You're supposed to be putting the GPU, or in some cases, the whole motherboard, with some method of protecting the areas you don't want overheated.

1) you're not supposed to put the WHOLE laptop into the oven.

2) It only applies to certain GPUs, most if not all, belonging to certain Nvidia lines. Unfortunately, I cannot be certain if your model is included.

3) I do believe you're S.O.L.

This is more along the lines of what you should have been doing.

http://www.dykab.com/showthread.php?11746-Laptop-motherboard-bake

 

xD i really hope that guy isn't serious, but i actually think he is. sorry buddy.

 

I don't know whether you guys have followed the massive slew of M1730 failures. Many of the folks moving into the Alienware M17x forums are originally M1730 users who had dead GPUs.

Thermal cycling seemed to be the common problem, and Dell did little to fix this problem. Fans kick in too late. The 2000 RPM mark kicks in at around 75 degrees, and the maximum 3000+ RPM mark kicks in at 86, which I think is too high. These cards should have the fans blowing full blast by around 78 so as to prevent them from ever exceeding the 85 degree mark.

On my original M1730, I had 6 GPU failures, and the last one died without any real gaming. All this within a 2 month window! With my new machine, I simply got an X9000, which I run overclocked, which in turn defaults all system fans to run at maximum speed. This has helped prevent that extreme cycling that is known to kill the GPUs. No problems with the card! Good solution, but Dell could've done better by releasing a newer BIOS for these laptops. Not everyone has the X9000 option.

 

It's actually a little strange how the cascade of M1730 failures happened almost overnight.

 

>.> The clevo/sagers with 8800m gtx's seem to be following this pattern recently as well.

 

You noticed, too, huh? I was wondering about the same.

It's quite strange how sudden it has been.

 

Hard to say if its defective gpu's. The other part to the pattern is the same people overclocked their gpu memory. It seems like most of the people that did not overclock or at least left the memory alone are doing just fine.

 

Oh, trust me on the whole paranoia factor. Even I found it a bit odd, but with my own history I don't like to get too sure. Perhaps one side or the other is the exception to the rule.

Note: The Dell GPU is built on a propriety single board with a propriety slot. So perhaps it isn't the exact same problem across all brands using 8-series and 9-series cards.

We have older M1730 owners, who have overclocked--and continue to overclock--but their cards are still running strong. On the other hand, what I have noticed is that with the Dell cards, there is another issue. If and when a replacement is offered, the card seems to be of a low refurbished standard. After the original GPU in my first XPS died out--with no overclocking whatsoever--the follow-up cards always produced a faint 'buzzing' noise and often stuttered.

Certain members have blamed the cheap ceramic thermal paste, but others have blamed the hardware. Like I said, I know of many older M1730 users with the first batch of 8800s still running strong and freely overclocking their cards (memory included). Although it should be noted that many of these people also take care of their machines - raising the back by half-an-inch and cleaning out the vents makes quite a bit of difference.

My current machine, for example, is also an older M1730. I intentionally went out of my way to track down one of these with an on-going premium warranty, and this one has had no issues.

 

I just sent in my Sager NP5793 for repair after getting a yellow checkerboard screen mid-movie followed by a black screen of death and failure to boot. My vidcard is a 9800M GT. I was advised to post on here from my original thread. If I decide to pay Sager to repair it with a new 9800M GT as they claim they must for $600 then it will be the last Nvidia card I get. When I get it back and either have Sager repair it or a third party repair it then I will install whatever old update I need to in order to let me run the fans on maximum all the time.

 

i have a 8600m gs in mine that's still running fine (2.5 years). I've been using a notebook cooler too after about 1.5 years. I'm hoping that this thing doesn't die after the warranty I have on it, which lasts for three years. I went the owners thread for this model of the notebook I have (ASUS S96S) and there are actually people in that owners thread who have had their 8600m GS fail on them. Thanks nvidia!