Defective Nvidia chipsets in your current Vaio?

Nvidia investors were just informed that the company is taking a $150 to $200 million dollar charge related to "higher than normal" failure rates in notebooks for it's (unspecified) "previous generation" GPU and/or it's thermal system.

Update:

The Nvidia G84, G84M, G86, and G86M chipsets ( GeForce 8 series, GeForce 9 series) have been identified as the ones that are defective. All of these chipsets (both mobile and desktop) use the same application-specific integrated circuit. It is possible that the G92, G92M, G94, and G94M chipsets are affected as well. The Inquirer has confirmed the issue also occurs in GPUs as recent as the new G96M chipsets.

See the list below for actual models that may apply to you. It appears to be specifically related to revision A2. The problem has not been reported with revision A3 and upwards.

This is a heat related issue with the bonding substrate used in the manufacturing process. The thermal tolerance for these materials was lower than what the specs required and can not handle the hot-cold-hot-cold states caused by power cycling the computer as well as voltage variations by the power management software. Notebook users are seeing failure sooner than their desktop counterparts specifically because of these reasons. ("usage patterns" mentioned in the 8k report)

Affected mobile models:
GeForce 8400M G (released May, 2007)
GeForce 8400M GS (released May, 2007)
GeForce 8400M GT (released May, 2007)
GeForce 8600M GS (released May, 2007)
GeForce 8600M GT (released May, 2007)
GeForce 8700M GT (released June, 2007)

The following models were identified by other manufacturers as well:
Quadro NVS 135M
Quadro NVS 135M
Quadro FX 360M
GeForce Go 7xxx (HP recent disclosure and 2007 class action settlement)

Affected desktop models:
GeForce 8300 GS (OEM) (released July, 2007)
GeForce 8400 GS (released June, 2007)
GeForce 8500 GT (released April, 2007)
GeForce 8600 GS (released April, 2007)
GeForce 8600 GT (released April, 2007)
GeForce 8600 GTS (released April, 2007)

Temperature Polls

Please take the temperature polls and answer a few questions about your notebook so we can try to determine the degree to which the Vaio models are affected.

GeForce 8400M GS ( SZ series)

GeForce 8400M GT ( FZ series)

GeForce 8600M GT ( AR series)

Timeline & Related Links

October 22, 2007
Brothers v. Hewlett-Packard Company class action (Power Plug & Graphics Card) lawsuit filed

November ?? 2007
HP identifies 24 models with faulty Nvidia chips, extends warranty to 2 years & free repair

June 22, 2008
Dell issues a bios update to fix thermal issues

June 23, 2008
175.19 drivers made available

June 26, 2008
BenchmarkReviews notices a substantial hit in performance with the new drivers.

June 27, 2008
175.19 drivers are pulled from the Nvidia website.

July 2, 2008
Nvidia admits GPUs failing at abnormal rates. Expects to take up to a $200 million charge for costs of repairs/returns.

NVIDIA Provides Second Quarter Fiscal 2009 Business Update

July 4, 2008
Foreign chip contractors claim defective chipsets date back to 2007

July 6-7, 2008
The 175.19 drivers re-appear on Nvidia's site.

July 7, 2008
Nvidia plays the meltdown blame game

July 9, 2008
All Nvidia G84 and G86s are bad

July 15, 2008
Answer Id 2214: Heat related issues with some NVIDIA notebook chips.

July 16, 2008
NVIDIA denies rumors of faulty chips, mass GPU failures

July 25, 2008
Dell identifies 10 models with faulty Nvidia chips..

Channel vendors demand card makers recall faulty Nvidia products

Jul 28, 2008
HP: Nvidia graphics defect an issue since November 2007

July 31, 2008
Figuring out which NVIDIA GPUs are defective -- it's a lot

August 12, 2008
GeForce Power Pack released (PhysX for all 8-series & 9-series cards)

Nvidia G92s and G94 reportedly failing

August 13, 2008
NVIDIA Second Quarter of Fiscal 2009 report

NVIDIA in the red last quarter due to manufacturing issue

August 14, 2008
Die Package Problem Might Affect Desktop Parts

August 18, 2008
Dell offers Limited Warranty Enhancement to All Affected Customers Worldwide

August 26, 2008
Nvidia's mouth shut by OEMS?

September 9, 2008
First Class Action Lawsuit filed against Nvidia over defective GPUs

September 10, 2008
NVIDIA sued over notebook GPU failures

October 09, 2008
Apple: some notebooks with NVIDIA graphics processor affected.

October 14, 2008
HP confirms 38 desktop models have faultly GPUs

December 9, 2008
First hard evidence that new G96 chipsets not immune to manufacturing problem

December 26, 2008
Nvidia's Solution To Defective GPUs : Buy Our New Chips

August 3, 2009
Sony: Distorted Video, Duplicate Images or Blank Screen Issues [with Nvidia notebooks]

August 4, 2009
Important Notice for VAIO® Computers with NVIDIA Graphic Chip

VAIOパーソナルコンピューターに搭載のNVIDIA社製一部グラフィックス
プロセッシング ユニットに関する無償保証期間延長のお知らせ

August 6, 2009
Nvidia takes [additional] charge for faulty graphics chips

 

Is there any release from Sony on this. Will we get a free replacement of mother board or only if there is a failure, replacement will be free.

 

All good questions. I don't know. In the interim they're reported to be considering tweaking their drivers to increase fan output earlier (as stress starts to build) rather then later. They're stock plunged when this came out; I'm sure you'll be seeing a lot of websites chime in tomorrow.

 

from the link of the website it does put the 8 at the end of it. i hope it isn't true because iam ordering a sony with a 8000 series intergrated with it...

 

that 8K is the link looks like some type of SEC id.

 

SZ750 here. I checked out the thread the Dell 1550/30 XPS chaps are running. I haven't experience any issues on my pretty much brand new laptop. I've played some Dawn of War, 5 hour tracts of Guild Wars, and some C&C Generals so far without issues. Really the only issue was finding updated drivers for my laptop since nVidia didn't bother to put any out. (X-Treme G drivers currently).

This does worry me a bit as cramming a t8100 and a 8400 GS into the same 13.3" case sounds dangerous. I suppose I'll have to download a temperature monitor to keep tabs on it. On the plus side, I'll be using the "stamina mode" more often since it runs the x3100 Intel graphics instead of the nvidia chip.

 

It is. That is a standard SEC form number. There is NOTHING in the article to indicate which chip family is affected.

Gary

 

Other than the quote " previous generation". Current generation chipset would be the 9xxx series no? That quote is most likely directed at the 8xxx series although it could also include some of the 7xxx series as well.

Hmm. Their public relations guy stated that they won't disclose which chipsets were affected. Probably because they fear a class action lawsuit. I'm certainly going to be monitoring the temperature a lot more closely when using the Speed mode in my SZ.

 

I have seen some issues on Asus notebooks and other brands, but nothing with sony.

I guess sony likes to take the real quality, not like asus, who likes to takes the cheapest parts available. I am sure manufacturers knows for each line of product, which are better quality and which arent.

 

Nvidia pulled a driver that was available just before the SEC filing - does anyone remember what it was? (It was higher than 175.16 - possibly 175.80?) One of the hardware sites (still trying to find the link) noticed a major hit in performance. I wonder if this is related. Would they deliberately throttle performance for the cards in question (in addition to increasing the fan output?). If so that would be a fairly easy way to identify the chipsets.

Edit:
It was the 175.19 drivers. Finally found the article. It was released June 23rd. BetaNews also seems to think there might be a connection. The 175.19 drivers may be throttling the core and memory clock speeds - can anyone confirm this with one of the suspected 8xxx series chipsets?

Also - some OEMs have released updated drivers as early as Wednesday. The updated drivers are targeting the 8400M & 8600M chipsets. No response yet from Sony.

It's worrying that they are not releasing details on the specific chipsets affected; it may indicate that they've only just become aware of the issue or that the problem is more wide spread than they'd like to report.

 

I fear its all the 8xxx series.

 

I'm wondering if the 8400M GT is affected by this... all I see are good things about it, and the temps are cool, even when overclocked. The FZ is on my list, along with a Gateway w/ HD 2600 and the SZ 730 w/ 8400M GS & X3100.

 

DAMN. It's the G84, G84M, G86, and G86M chipsets - and it's ALL of them - both mobile and desktop platforms!

Affected mobile models:
GeForce 8400M G (released May, 2007)
GeForce 8400M GS (released May, 2007)
GeForce 8400M GT (released May, 2007)
GeForce 8600M GS (released May, 2007)
GeForce 8600M GT (released May, 2007)
GeForce 8700M GT (released June, 2007)

Affected desktop models:
GeForce 8300 GS (OEM) (released July, 2007)
GeForce 8400 GS (released June, 2007)
GeForce 8500 GT (released April, 2007)
GeForce 8600 GS (released April, 2007)
GeForce 8600 GT (released April, 2007)
GeForce 8600 GTS (released April, 2007)

All Nvidia G84 and G86s are bad

I'm at a loss for words. This is really bad.

Updated first post with listing and more info.

 

soo what does this mean for people with these laptops????!!!

i bought a FZ21S at the end of last year and whenever i play games on it..the keyboard is soo hot!!

I personally hope theres a new free replacement..

anyone know anything?? i Just hope mine doesnt catch fire like those dodgy batteries did!!

 

please keep us updated! i bought a FZ last summer with the 8400 GT and that mofo gets hot as hell... I hope sony does something about this... I would even accept a refund or store credit (so i could buy the newer models...which hopefully won't be affected)

 

holy crap mine is manufactured on 05/08.. should i worry?

its a sz740

 

8400m GS... hmm....

 

same!!!!!!!!!!!!!!!!!!1

 

My SZ was manufactured 4/8 (also a 2008 Santa Rosa model) and it's a G84M revision A2. My guess if that you probably have a A2 as well. (check it with GPU-Z)

While we wait for an announcement/lawsuit/whatever I'd suggest:

- use the integrated graphics whenever possible now
- monitor the GPU temp if the Nvidia card is in use (tray and possibly OSD)
- disable anything that auto-adjusts the voltage on the chipset (thinking specifically of Powermizer - other power management apps may also apply. The idea is to minimize the temperature variation as much as possible.)
- avoid using hot drivers from LaptopVideo2Go/NBF. stay with the stable ones that other users have vetted
- it goes without saying that overclocking these chipsets would be a really, really bad idea.
- email Sony and complain about this!

It's sad because mobile users are essentially screwed; any "solution" to this will almost surely reduce the battery life while trying to keep the GPU alive. It's not about keeping the chipset from overheating. Any solution has to focus on reducing the gap between the lowest and highest temperatures because the materials they used just can't handle it. Unfortunately powercycles are unavoidable; think about how many times a notebook user puts their computer on standby or hibernate.

It's just not a pretty situation.

 

If the Inq is right, it looks like my gf's SZ680N is affected too. Manufacture date 02/08. That sucks.



I thought G86M would be an 8600 part...

 

Yeah, it's a bit confusing. The technical briefs listed on the official 8400M page are actually for the GeForce 8800 GPU. The 8800's chipset name is G80 (desktop) and G92 (mobile). Maybe we should interpret this as all 8xxx cards are based off the 8800 architecture and they (for whatever reason) named them in reverse order? Either that or the whole internet is wrong.

Anyway some more info:

GeForce 8 Series

NVIDIA GeForce 8400M G

 

Has anyone had any problems with their AR series? I just (as in today) bought an VGN-AR770 with a 8600M GT brand new off an ebay store and am wondering if I should forfeit the sale!

 

Hmm...

Let me try to sum this up. The problem is with overheating of the GPU, which could lead down to meltdown of it? If this is true...we all (that have such GPUs) should experience problems with heat?

Well...I do have a passive cooled 8500GT in my desktop and the baby is very cool. Is there perhaps a way to test it? Like run some benchs and check the temp, which should go up drasticly?

 

As I understand it, the issue is not so much with heat as it is with heating and cooling cycles. Of course the wider the range of this cycle, the worse the outcome.

Gary

 

Yup. @ScuderiaConchiglia is spot on. The problem is that the physical material used to construct the chipset can not withstand the constant heating, then cooling, then heating, etc.. Given enough cycles the material will fail/crack/rupture and at that point no amount of cooling (from the fan) will keep it from failing. For notebook users it's a major issue; there's no way to swap out a video board if it's been integrated into the main motherboard. The whole motherboard has to be replaced.

Let me explain. No, there is too much. Let me sum up. ;p

It started to surface on the HP and Dell forums. The $hit really hit the fan when Nvidia filed a 8k report with the SEC stating that they would take $150M-$200M charge to "repair, return" (and replace) their defective products that were "failing in the field at higher than normal rates". In the same report Nividia lists the following as factors that contributed to the chipset failure:

- "weak material set of die/package" (their suppliers)
- "system thermal management designs" (the OEMs)
- "customer use patterns" (us, the users)

This report was written with their investors in mind; Nvidia essentially places blame on everyone but themselves. It *could* be true that the suppliers screwed up. It's *possible* that the OEMs screwed up as well. The users (especially mobile users) are the proverbial kitchen sink; it's unrealistic to expect notebook users to avoid suspending/hibernating and/or taking advantage of power management software on a mobile platform.

What part did Nvidia play in this? That's the real question and we probably won't have all the details until a lawsuit is filed; they are in damage control and are refusing to talk about it. There's compelling evidence however that Nvidia knew about the defect prior to the SEC filing and possibly before OEMs placed the orders for the parts. The voltage difference from the revision A2 to A3 is particularly damning. The removal of the underperforming 175.19 drivers just prior to the announcement is suspect as it appears they deliberately tried to underclock the cards in question. The driver has since been made available again to the public as of 7/7. As of this writing BechmarkReviews does not endorse it.

What happens next? Well, HP has released a bios update that specifically addresses the issue. Some Dell users are already on their second and third replacement boards. It's clear that simple replacement does not solve the problem. Increasing the airflow only prolongs the inevitable (while reducing battery life in the process). The chipsets need to be re-manufactured. Someone has to foot that bill. Notebook OEMs are in the middle of a platform shift; the motherboards (which the defective chipsets are integrated into) are more than likely no longer being manufactured. Sony has not replied to my requests for information and there's nothing in their knowledgebase that addresses this issue.

The problem (for users) is that by the time this works it's way through the courts most users will probably be out of warranty anyway.

 

Well...if users laptops will get defective because of this, they will change the mobo, thats for sure, no matter if the law suit is over or not.

What nvidia is probably trying to do is to stall the process. So they would avoid the problem is changing everybody's mobos and chips and only to replace the ones that will break down.

But my questions remains...can we somehow "reproduce" the error" or what? As you said it yourself, HP and Dell are working on it, and it seems Sony does not even have an issue with it..since there is nobody on the forum complaining about the damn thing. Pehaps sony have so much better cooling and materials so it does not effect their products or whats the catch?

As I understand it..this is an issues that is waiting to blow into our faces...it sits there and waiting to blow...it may happen today in a year or never...correct?

 

Yes, by alternating stress loads on the GPU. (Going from idle state to high load back to idle/standby/hibernate state.) Repeat. (I don't recommend anyone doing this on purpose.)

They might have a better thermal cooling design than other OEMs but they order the same chips like everyone else. The key issue here is that those chips are manufactured with the defect. A superior thermal cooling design is like chemo-therapy with certain forms of cancer; you're buying time at best. You'll still be shutting off your computer (or put it on standby/hibernate) which reduces the GPU to it's lowest temperature. Presumably you're also using some form of power management as well to ramp up/down performance while you're working. Reaching higher temperatures and immediately dropping the temperature down to idle or (standby/hibernate/off) will accelerate the breakdown of the materials faster than a gradual controlled decrease in temperature.

Run a game/high load app and close the window. What happens? You immediately see a drop of 10-20 degrees right? My guess is that the driver "fix" they were talking about in the 8k filing keeps the fan running longer after you close the window, gradually dropping the temperature slower than before. They probably also start the fan at full blast when you launch a game/high load app and gradually decrease the output to the levels we have now.

Yes. We face this with any mechanical hardware like hard drives, fans, etc.. They typically have an estimated manufacturer's life expectancy. The Nvidia chipsets don't.

 

DAMNIT my 1 year warranty is nearly over. i want a non defective mobo, NAO!

 

Mine expired two days ago, of course! This could get REAL interesting. It seems odd that we have not heard of any issues with Sony machines, yet the reports are coming in for Dell and HP machines. Could there be ANY truth to Nvidia's claim that the issue is with chips delivered to HP and Dell? Seems odd since the substrates OUGHT to be the same across the entire line. Unless the issue is restricted to a particular FAB location.

Gary

 

Might just got to do with a fact that HP and Dell have a poor built quality.

Do this effects all HP and Dell models, or only the "cheap" ones?

 

One FAB location is a possibility. If that was the case however why didn't Nvidia identify the plant specifically? I also agree with you that the materials would have affected the entire product line; this wouldn't be model specific but the symptoms *might* appear earlier in models with less conservative cooling. Figure the power cycling would be about the same but the temperature variance would be greater than their counterparts.

Don't forget Apple. The Macbook Pro uses a G84M chipset (GT version no less). It's already been all over the news because of the misapplication of thermal grease. That's probably compounding the problem and may help to explain why we're seeing failures sooner from certain OEMs.

So in answer to your question It's not limited to 'cheaper' models; the problem exists across the board (no pun intended). There's probably a greater chance that a cheaply made model (or rather, a model with a poorly designed thermal system) will fail sooner than others because higher temperatures accelerate the process. Any rock will crack with alternating extreme temperatures. The issue at hand is that the materials Nvidia used during manufacturing have a much lower tolerance than what is needed. The lowest extreme (off state) is constant amoung all models. The variables which are determining which models fail first are related to the idle temperature (not everyone idles the same) and highest extreme (full load) - it's the process of going back and forth between those extremes that is causing deterioration in the material. Once it ruptures/cracks/whatever the temperature (for idles/high extreme) will climb causing the whole process to accelerate until it eventually fails.

 

has anyone talked to sony about this? I emailed bestbuy since I paid for a warranty through them for my laptop. I am not happy as I expect most here arent about putting down this amount of coin for a laptop that is born to fail befor its time.

 

I'll post their replies when I receive them. I've sent three so far with no response.

 

Is there any confirmed news from Nvidia?
I believe I saw somewhere that the 8400GS would be fine, but then that person could have been misinformed...

And I myself have effectively no knowledge about graphics cards.

(I was fine with an intel chipset in a laptop (same laptop) for 3 years, that was till the end of this June)

Now in my SZ I do prefer the Intel chipset just as somne other people have said because its cooler, but if I want to play Anno 1701 or Age of Empires I need the Nvidia card...

Yet I don't need to play these games (its just that I got tired of books, how many hundred pages can one read every day? And I read a lot...)

So should I make sure I minimize use of the NVidia chipset or not? Or should I try using it as often as possible (no overclkocking or so, just a lot of normal usage) to get a replacement under warranty?

What is the current official line?

And just wondering, when did they notice this? I got my SZ7 only this June and its the high end model, so could it have a card in which the issue was resolved already so I'm fine?

....

Any information is greatly appreciated.
Thank you for the information posted already.

 

High temperatures = bad. Low temperatures = good. That pretty much is all you'll ever need to know about graphic cards. This situation is a little different. The deciding factor appears to be how often you change states (high, low temperatures) within a certain time period and the speed that you reach the highest and lowest thresholds. This is what Nvidia was referring to when they described the customers "usage patterns" in the SEC filing.

The physical materials used in manufacturing the Nvidia chipset can't take it; they start to develop tiny fractures which in turn raises the temperature. This process repeats itself over an over again slowly raising the highest temperature until it exceeds what the system is capable of. At this point usually the notebook will shut down. Do it enough times and it may never boot up.

Definitely no overclocking. The X3100 is actually quite capable for office tasks; I use it during the whole workweek. If it works for you then you might as well take advantage of it. The problem with your replacement theory is that you're likely to get another defective board. Not one user has reported that their SZ has a revision A3 in it; it's possible that the SZs were only manufactured with the A2s.

The public became aware of this on July 2nd, 2008 through their 8k SEC filing. There is however some evidence to suggest that they (Nvidia) knew about it prior to their report. It's very likely that the OEMs are interested to know when Nvidia knew about as well (e.g. was it before they placed their orders for the parts).

You most likely have one of the Santa Rosa Refresh models which has a G86M chipset in it. It's probably a revision A2; you can confirm this with GPU-Z or another program of your choice. Sorry. Wish I had better news for you.

From Sony? Nothing on their knowledgebase and no replies to (my) emails about the issue. I've been encouraging other people to email them and report back with what they hear. It's possible Sony just found out (in early July) like the rest of us.

 

i don't mean to take over this thread.. but is mine defective? and how can you tell?

And whats everyones temps at? my GPU-Z says mine is at around 67C


I also gave sony an E-mail about it.. hopefully if they get to many emails by us they will give a damn about the problem

 

iisdev,
thanks a lot to summing everything up and explaining/writing in a way I (and popssibly some other people) undestand in one go, so one knows what is going on and where we (as customers) stand.
(Read the high/low temp bit...)

Now, yes I have an A2 Nvidia GPU, does this mean I'm affected?
I don't quite get the A2, A3 bit. Sorry.
All I undestand these are two versions of the same graphics card manufactured a different times, but which one is affected?

And again, thanks a lot!

All the best!

Vash the Stampede mine is currently running at 56°C/58°C...

(But except for games I'm fine with the Intel Chipset...)

 

Yes, you're in the same boat as the rest of us. @Vash, you too.

The A2 is the chip revision number. It should also be physically printed on the die of the chip. Example pic below (this is a 8400M GS from a Dell XPS courtesy of @Trave11er):



The Macbook Pro users on their second motherboards (mostly G84M chipsets) have an A2 (possibly A1 - it's not clear if OSX is reporting that incorrectly). The Dell users (some of which are on their third boards) have revision A2 (the replacement boards also have A2 as well). HP (at least the Pavillion models) were A2. The A3 had a slight voltage change and appears to be immune.

Are you getting those temperatures with the X3100 or 8400M? My 8400M idles around 68°-70° which is high (compared to the Mac users) and about average with what the Dell users have reported. It might be interesting if we had our own poll.

 

I don't get a temperature for the Chipset, so my 57°C/58°C is for the 8400GS

 

true Cool Fan for the CPU, it has 2 temps there. on mine its around 40's




So is there guides or something on how to crack open this thing and fix the problem ourselfs? like using article silver or some kind of thermal greese on it like that.

EDIT:

heres my sony reply about the email towards the defective cards and if mine was effected:

 

Huh. They must be catching up on their inboxes.

Here's the response I received when I asked them to confirm that my model (SZ780NW) uses the G86M Nvidia chipset. Pretty easy question; I was expecting a simple yes or no.

"Thank you for contacting Sony Support.

I'm sorry for the inconvenience. The information you have requested is an unpublished specification. The hardware (e.g. motherboard, hard drive, video graphic chipset) used in our VAIO products are custom designed and built specifically for Sony as OEM hardware (Original Equipment Manufacturer) from various selected manufacturers. They do not have a direct retail counterpart model from which you can ascertain additional information. All product support and available hardware documentation published for your system is provided exclusively through Sony. Should we decide to release any additional documentation, or firmware upgrades for the hardware of your system, it will be posted on your support site.

Thank you for your time.

The Sony Email Response Team
CP3M
Eric"

This was one of several questions I sent. It was a control question (one where I already knew the answer) to better gauge their response.

Eric was wrong; the information I requested was published on page four of the service manual (available to anyone for purchase):



@Vash, no that's a picture from the Dell XPS heatsink mod currently circulating in the Dell forum. Other than the 128MB of video RAM (in the XPS) it's the same chipset. I plan on swapping out the thermal paste on my SZ in the near future in an attempt to bring the idling temperature under control. I'll take some pictures then.

Ars (I love that site) ran a story today about the issue.
NVIDIA denies rumors of faulty chips, mass GPU failures

They ultimately concluded that there's no way to know if Nvidia is lying until they release their fourth quarter financial filings. I'm not exactly happy with the way they're discounting reports from actual users.

@Vash, I'm curious why your response from Sony referenced the video driver (not an issue) and did not mention the physical chipset (the issue). Maybe the service rep that answered was confused?

 

so, if we have one of these chips.. then what do we do?

sit and wait for our notebooks to fail on us? :[

i have a 8400m gt revision a2 also :/

 

Well, all components fail, just when. Even if your video card is not defective, you are still practically sit and wait for it to die.

So far, there are very few facts and explanations from nVidia. Just based on personal experience as an engineer that had dealt with this kind of "fleet-wide" problems and the consequent root cause analysis and the eventual liquated damages, I would not be surprised if anyday now nVidia annouces more notebook video card are affected.

If you are worried and still in the warranty, you could buy an extension right now. Other than that, there is really nothing you can do except boycotting nVidia in your future purchases -- yeah, class-action lawsuit is one way, but your chance of success is very low because, like I said earlier, all components die, just when, so it's almost impossible to pin nVidia on "defective design". Been there. Done that.

 

does sony come with a 1 year warranty right off the bat?

i bought this notebook may 2008, so i should be okay, right?

 

Depends on where you buy it, some places come with 3 year, but all brand new Sony VAIO should have at least one year. You will probably want to check your GPU temperature early next year and decide whether you want to extend your warranty.

 

i have the fz CTO, so i have the 1 year warranty...

how grave is the situation with the chipsets?

right now my gpu is running about 62 C but right now its night time and cooler. in the afternoon it was higher.. around 67+

 

The short answer is that we're not sure.

The long answer is that so far the majority of replies have been from SZ users who may (or may not) have been alternating between the Intel X3100 integrated chipset and Nvidia chipset. Since you're one of our first FZ users would you mind taking the 8400M GT poll?. The more data we can collect the better.

 

You guys might be better off if you insure the laptop, not extend warraty...some insurance policy comes with anfi thief clouses too and they are genery more wider than just warranty...and in a lot of cases its cheaper to insure a laptop, since sony vaio have a bit expensive warranty extentions

 

And how will that cover the defective Nvidia chipset issue?

Gary

 

If it covers theft..i am sure it covers faulty nvidia cards too?