laptop Nvidia 8600 GT, are they faulty?

I know I've been posting about video cards lately but school is coming down close and I am just burning out from researching and looking for the best notebook for an architect student. My budget is under $1000 and but a little over $1100 is ok.

So far the MSI GX600 is really catching my eye since it has an 8600 GT, but my dad told me it that the 8400 and the 8600 are faulty chips?

is this true? Or does this apply to dell computers only? I know my friend recently bought a dell ms1330 and he had a video card problem, it was an 8400 and it broke or something..don't quite remember.

if the 8600 are faulty, what would a be good laptop for me to get for architectural purposes?

the Lenovo T61 doesnt let you upgrade to anything higher than a 128mb Nvidia Quadro.

 

Dell have actually been pretty good on the situation of the faulty/overheating GPUs, they've released a new BIOS to increase the speed of the GPU fan to make it cooler.

You should post in the "What Notebook Should I Buy?" Forum, people looking through there can assist you in determining what's best for your budget/needs

 

Theoretically all laptops with the Geforce 8400/8600 GPUs are affected, good possibility with the 9200/9300/9600 series as well. This is a defect on nVidia's end, not the laptop manufacturer's.

If you're concerned about getting a laptop with a decent midrange GPU for less than $1000, I would recommend you to look at the Gateway M-6862 and M-6864FX at your local Best Buy; they have the Radeon HD2600, which performs close to the 8600M GT.

 

there is nothing really defective... the cards work perfectly fine. that stupid heat cycle thing doesn't matter.. i'm really sick of these "defective nvida card" threads

 

^ then you shouldn't have clicked on this thread.

considering there is a softmod to the 8600 GT to make it into a quadro, I am definitely considering this out of all other cards.

I just wanted to have some clarification that I was confused on

 

I don't think it performs as well as a quadro when working on things the quadros excel at :S
I've seen the comparison somewhere on this forum, just don't remember where..lol

 

^ gaaaah

Let me see! I have the thread of soft-modding it from a different site.
I thought I posted this because I was focusing a lot on the video card. That's the main criteria and I wanted to see if people had a better opinion/suggestion for an architect student on a budget.

 

Nvidia sued over faulty GPU "cover-up"
Wednesday 10th September 2008

Nvidia is being sued on behalf of its shareholders for withholding information over faulty graphics chips.
Nvidia admitted earlier this summer that it had a problem with faulty GPUs that had affected laptop manufacturers including Dell and HP.

The lawsuit alleges that the company knew of the problem last year, but didn't disclose the issue to shareholders "in a series of false and misleading statements made to the investing public."

[...]
Speaking at its NVISION conference last month, an Nvidia spokesman said it had been prevented from coming clean over the scale of the problem by PC manufacturers. "The truth is," he explained, "our obligations to our partners limits what we can say. We need to leave announcements to our partners like Dell and HP.

[...]

Because they are perfectly fine, huh?

 

lol and that is why you should make a post in the "What Notebook Should I Buy?" Forum

 

Oh, I already did, but it was only a main focus of the video card

like, quadro nvs 140? 8600 GT? 9600 GT? etc. etc.

 

Even though many things are often hyped worse than they are, the nVidia manufacturing defect is not one of those. The issue is serious and nVidia is facing difficult times as a result. It wasn't completely their fault though, as it had to do with the TSMC fabrication process.

Personally, I won't touch an 8400/8600M series... The 9600M series seems to be good, or I doubt HP would be using them, since they were the ones that first reported the issue. I would think that if nVidia didn't fix it, they would not be using the 9600M in their new DV5Ts. That evidence is not enough to provide complete proof that the 9 series mobile cards are problem free, but good enough for me.

 

I know they mentioned only dell and hp, but could my 9500m gs from acer could be affected?

 

what is the fault with these cards so i can check it out for my self and others know what to look for.

 

I guess theres heating issues? I am not sure, i am still looking into it

 

No, I believe that the 9500s are the refresh of the 8600s which means that they were fixed, and run at much lower temperatures, but again, I have no proof since I don't own a book with a 9500M GS, but you can read around the forum, I'm sure it has been posted

Well, the fault for me would be that when I game, the temp has gone up to around 97C - 110C, and usually idles at around 75C

 

No, this issue really is extremely overhyped.

I remember ever since I got my notebook and started coming to NBR pretty much everyone was being recommended the 8600 for gaming and 8400 for light gaming. It was a solid card, great performance, etc. etc. and no one ever mentioned anything about its failure rates or any heating issues they have had that may be an anomaly.
Fast foward to today and we know the GPUs high lead solder bumps tend to fail at temperatures above x degrees celsius and suddenly everyone's card is a piece of crap and everyone has a story about a friend/third cousin twice removed who's card exploded as he was typing a paper in powersaver mode with a notebook cooler (exaggerated but you get the point).
That's not to say I am taking NVIDIA's side either; they shouldn't have hidden the problem and they should replace any cards that fail with good cards (Later batches have the problem fixed, I believe). Unfortunately, this will be a lot more costly for NVIDIA than this, as many uninformed consumers (such as Mr.Third Cousin) will want their cards replaced despite the fact that they probably aren't ever going to reach temps high enough for the fault to be an issue.

 

Lol, true. True.

Despite me owning an 8600 that's not faulty [yet, hopefully never], I'd say stay away from them. The DV5t is a nice choice, and I'm quite sure the problem doesn't carry over to the 9-series cards, even the ones that were "refresh" cards [9500 GS, 9300 G]

 

Has it occurred to you that a failure needs time in order for the number of thermal cycles to build up. Not hours, not days, not even weeks. Months and years.

Every single card with that defect IS a piece of crap, since its lifespan is considerably shorter that it ought to be, and since the notebook designers expected a card able to withstand a certain kind of thermal cycling. It's just like putting a 486 fan cooler on a Pentium4 at 3.6 GHZ.

There is a sort of statistical bias here: the great majority of the people who ran the extra money to buy a dedicated video card did not do that to type a paper in powersaving mode. Did that because they expected to use the full power of the card.

Now, more than one year later, all the people who, in that period of time, had their card (or worse their mobos) fried are stepping out, pointing their finger at DELL (Should I mention _which_ finger?)

Unfortunately nvidia did not disclose that kind of information.
Moreover, changing a material in an ASIC fabrication process is not like using olive oil instead of butter. Something needs to be adjusted, and money has to be spent. I an not sure that nvidia (that so much has done to cover this up, and still is not coming clean) had the will to run more money for that.

A card of that type won't reach 70°C???
C'mon. Most system under stress have 'em run at 90°C, some even at 110°C.
But in a way you are right: mr. Third Cousing would not build up enough thermal cycles to have the card fail under the warranty period.
After that, he's on his own with a lemon.

And, what's worse, he knew it and saw it coming (or better yet: he heard it coming, since the bios update had the fans run all the time, eating battery life).

 

I have yet to see my stock-clocked GPU go over 65 degrees yet (temp readings from I8kfanGUI), it's my CPU that can go well over 90 degrees instead. My GPU is supposedly unaffected because it's a really early version of the 8600, I'm not worried and don't expect to be suing Nvidia anytime soon for screwing up their production process.

In short, anyone who wants to trumpet the stupidity of the company in making defective chips has a vested interest in litigation.

To the OP: Another reason not to touch the 8600GT is that better more efficient cards are available with better performance for the same budget.

 

can we trust the temps programs shows us? Because I know my cpu can have 2 different temp depending on the program I use(one 15°C higher that makes more senses).

 

To be honest, I'm not entirely convinced that the DDR2 cards are even affected. I called up Micro Express (the company I bought my Compal JFL92 with an 8600M GT from) and they said that they have not heard of any problems with this. Also, something that goes wrong on the scale of years is not that big a deal -- it's not like the rest of the laptop is perfect and will last forever.

 

I would hate to get in an internet argument, but I also don't want to offend you by leaving your post unanswered.

Yes.

Not really. Once again, the defect only occurs as you said, with enough thermal cycles (above the critical temp). A considerably large number of consumers such as myself have sufficient cooling systems to not be affected (my temps have never gone above 67C) and another considerable number will use their card lightly enough that the defect does not affect them either. The affected userbase is gamers (especially that overclock) and users with notebooks that have very high operating temps.

lol, sorry about that. I was exaggerating to make a point: The userbase that wouldn't be affected by the fault are now under the assumption that the GPUs have/will fail at times when the fault doesn't affect them. Everyone is pointing fingers at the fault for all their GPU troubles.

I remember a topic on here where someone said that cards after july were fixed. I don't remember if they had a source or not.

Some notebooks had enough cooling that they don't. Mine doesn't. A few others on here as well. I recall the Dell Blog post where they stated they would cover affected notebooks for an additional year of warranty and my system wasn't listed there, so I suppose that further proves my point.

Once again, you are assuming Mr. Third Cousin is going past the critical point to build up these thermal cycles. He isn't in the userbase that builds up temps high enough to be affected.

I fully agreed with your implied point here. I have said it before (probably in another topic) that manufacturers should replace the cards on a failure and request basis. Even if someone doesn't reach those critical temps, they will still be adversly affected by the fans quick fix.