Problems with 8800GT

[Hyperlite]

196.75 Alert!
NVIDIA is aware that some customers have reported fan speed issues after installing 196.75 drivers from NVIDIA's website. NVIDIA has removed these drivers and asked its partners to also remove the drivers. Any customers still using 196.75 drivers are asked to immediately discontinue use and either roll back to their previous driver or install the new 197.13 drivers. Driver roll back help can be found here. For those that experienced quality issues with 196.75 drivers, NVIDIA has setup this

Taken directly from the nvidia site. its on the "Download Drivers" Page on the menu.

I was thinking there was some issues with some drivers and i didnt know what ones they where, so i looked at the ones you had and went to the nvidia site to check for newer ones and that alert was up.

that was a couple weeks ago. He is using older drivers than that. These 8800xxx cards seem to be dropping like flies these days...

 

[7earitup]

196.75 Alert!
NVIDIA is aware that some customers have reported fan speed issues after installing 196.75 drivers from NVIDIA's website. NVIDIA has removed these drivers and asked its partners to also remove the drivers. Any customers still using 196.75 drivers are asked to immediately discontinue use and either roll back to their previous driver or install the new 197.13 drivers. Driver roll back help can be found here. For those that experienced quality issues with 196.75 drivers, NVIDIA has setup this

Taken directly from the nvidia site. its on the "Download Drivers" Page on the menu.

I was thinking there was some issues with some drivers and i didnt know what ones they where, so i looked at the ones you had and went to the nvidia site to check for newer ones and that alert was up.

Definitely old news - but thanks anyway.

I suspect that the 196.75 is what took out my original 8800GT, even though it never had fan issues nor high temperatures. I used the drivers before they ever posted that alert though.

EVGA recommends 196.21s, but they have not worked with any of my replacement 8800GTs. The only ones I could even get somewhat stable were the 191.07s and then they killed the card when trying to play Starcraft II.

I am at a loss here. I do this type of stuff for a living and it really hurts one's ego when they cannot figure out an issue like this.

 

[SinxarKnights]

I recommend you don't use what EVGA recommends. And remove any and all drivers and related components (you said you used a fan speed program or something in a previous post) and use the ones from windows update. If you have the time, I would also try a fresh windows install and dont install any drivers other than what comes from windows update and try it like that. No PhysX but if those still give you problems it has to be the card itself.

you could use the free imaging software for your drive to save some time if it doesnt help. Goto your drive manufacturers page and look under downloads for your drive and do an image backup, and just restore the image later if needed.

 

[7earitup]

Well the card I have in now is toast anyway - there is an RMA replacement on the way for it now. So really there is no use to do anymore testing.

 

[AndroidVageta]

Maybe its a bad PCI-e slot? In my opinion, something like bad drivers could be the result of artifacting/BSOD while in Windows/playing games...but for it to artifact while booting and in BIOS? Definitely sounds like a heating issue to me...as, like you said, even bad drivers will have ZERO effects on the the boot screen and/or BIOS screen.

Since your case seems to be up to par on the cooling part and what not (Ive known 8800GTs to be in stock cases from Dell and Gateway with 1 80mm fan with no problems) it might be that during shipping the heatsinks may have gotten loose or not seated properly...but again, these arent push pin heatsinks, theyre screwed on and cannot fit any other way than the right way...so even a heat issue is kinda out of the question.

My only thought is that it might be a bad motherboard...or a bad PCI-e slot to be more specific...dont know how common this is or what not, but nothing else seems to fit a single profile related to solely to your video car (ie. drivers, heat issues, etc.).

 

[SinxarKnights]

If anyone has any other suggestions as to what may be killing video card after video card, I would absolutely LOVE to hear them! Thanks!

Ok sorry for trying to help.

 

[7earitup]

Ok sorry for trying to help.

It was not directed at you specifically. I am just trying to formulate some ideas on what else it could be. I mean, what are the chances that I have burned through 3 8800GTs in the past 2 months?

So on to another issue, what are the chances that the PCI-E slot is bad? I mean, every time I get another 8800GT it runs fine for a day sometimes or a week or two other times. Can a bad PCI-E slot have these effects on a video card?

 

[AndroidVageta]

It was not directed at you specifically. I am just trying to formulate some ideas on what else it could be. I mean, what are the chances that I have burned through 3 8800GTs in the past 2 months?

So on to another issue, what are the chances that the PCI-E slot is bad? I mean, every time I get another 8800GT it runs fine for a day sometimes or a week or two other times. Can a bad PCI-E slot have these effects on a video card?

Im sure the 8800GT isnt the problem...unless you have on the slim chance gotten 3 bad 8800GTs in a row then it sounds like a motherboard issue to me...I mean, honestly, that seems to be the only logical problem.

 

[7earitup]

Is there anyway to test this theory? Or is my string of dead cards test enough?

 

[AndroidVageta]

I say wait on another replacement just to be sure...100%...if the problem still exists, then I would say its time to check out a new motherboard.

Another user here said it might be the hard drives...I doubt that...that might cause Windows to crash, but, just like with bad drivers, should in no way effect the BIOS/boot screen.

 

[7earitup]

I say wait on another replacement just to be sure...100%...if the problem still exists, then I would say its time to check out a new motherboard.

Another user here said it might be the hard drives...I doubt that...that might cause Windows to crash, but, just like with bad drivers, should in no way effect the BIOS/boot screen.

One of my online friends just recently upgraded from socket 775 to socket 1156. I am going to see if he will let me have his old motherboard for cheap if this problem continues.

I really do not know how many more cards EVGA will send me before they throw in the towel.

 

[AndroidVageta]

Those are my thoughts...either they need to completely check the card before sending or send you a brand new/different model.

But yes, do ask your friend about his old motherboard...or at least use it as a test platform to see if the PCI-e slot is indeed the problem, which is what Im putting my money on.

 

[7earitup]

Those are my thoughts...either they need to completely check the card before sending or send you a brand new/different model.

But yes, do ask your friend about his old motherboard...or at least use it as a test platform to see if the PCI-e slot is indeed the problem, which is what Im putting my money on.

Thanks for your help.

 

[tweakboy]

Well when a video card gets Artifacts it is because of the BGA memory chips on the most cards have 8 memory chips.

Its not your GPU fault, something is up with the memory maybe you should also try to increase your fan to 100 percent,, I bet the artifacts might dissapear.. gl

 

[Puffnstuff]

Just how did you check your voltages? If you didn't use a digital multimeter on the actual rails then you've checked nothing. A failing ps will start to reduce its output as it heats up. All ps's loose output as they heat up but one that is already marginal will show itself sooner when the heat is on.

 

[AndroidVageta]

Well when a video card gets Artifacts it is because of the BGA memory chips on the most cards have 8 memory chips.

Its not your GPU fault, something is up with the memory maybe you should also try to increase your fan to 100 percent,, I bet the artifacts might dissapear.. gl

I highly doubt that memory could be the cause...with 3 different cards...if so its something on EVGAs side and nothing that he can prevent or make better...plus, I dont see the memory overheating that quickly. There are plenty of people out there with 8800GT's that are as computer illiterate as blind deaf people are to watching TV...none of them have had this many problems with their cards...especially not 3 in a row. Now Im not saying that no one ever has issues...but I highly doubt that if they do they are this many in number. Just saying...

Just how did you check your voltages? If you didn't use a digital multimeter on the actual rails then you've checked nothing. A failing ps will start to reduce its output as it heats up. All ps's loose output as they heat up but one that is already marginal will show itself sooner when the heat is on.

Im sure that even if it were the PSU he computer would either lock or freeze...hes got a good unit and not nearly enough running off it to even make the PSU break a sweat. Even if there is no multimeter testing done, the BIOS or Windows application will be close enough to tell if theres something seriously wrong with the PSU...if it were reading 14v on the 12v line, even with Windows programs/BIOS being as inaccurate as they are wouldnt be THAT far off.

 

[VirtualLarry]

I'm surprised that no-one mentioned this, but your PSU is an FSP platform, that is know to have high ripple (parts killing ripple?) under heavy load. I would replace the PSU, before you go through too many other cards.

 

[7earitup]

FSP platform? What do you mean?

 

[faxon]

he means it is made by FSP group and that FSP group is known for making a lot of the shitty PSUs sold in markets all over the world, as well as being one of the companies which supplies a lot of OEMs with cheap shitty PSUs as well

 

[7earitup]

Ah - Well, there have been nothing but good reviews on the OCZ line I have.

 

[Puffnstuff]

My son is running an ocz powerstream 700 and it has done fine but the enermax liberty 620 it replaced took a dive and took the mb with it. Enermax replaced it with an infinity 650 which has never been opened. The ocz went in and has worked fine ever since. I still say that the rails need to be checked using a reputable tool to see if they are sagging when the pc comes under load.

 

[Axon]

I remain convinced it's a combination of that particular game - notorious for killing rigs - the card being a touch older, and the drivers.

So what's your status now, another card shit the bed? Christ. BFBC2 BSODs me like it's LeBron James and I'm, well, me. And we're playing basketball, not like, SC2 or something, in which case I would dominate that fool.

 

[7earitup]

BC2 had caused issues with newer drivers. Starcraft II caused it to die with 191.07s.

Supposedly each PCI-E wire is on a separate rail on my power supply. I have a digital multimeter - how can I test the one that is feeding the graphics card while it is plugged in?

This point is kind of mute right now anyways because I cannot load any NVIDIA drivers without it crashing. I can only load Windows 7 when Driver Sweeper has been used and the video driver is not loading in the device manager.

 

[Axon]

Yeah SC2 is a bad little boy as well, though my reference to it was limited to the fact that I would destroy LeBron James and his weak Terran turtle strat. 3 zealot rush, gg

Nonetheless, I have not heard of as many issues with a popular game ever as with BFBC2. I can understand why you'd think it's the PSU, but I have a feeling it's not. It's worth the effort I suppose. However, three failures with the same game on the same card? Is it still crashing/artifacting during BFBC2 or is it just FUBAR now?

 

[7earitup]

The one in the desktop now is FUBAR. It has artifacts in POST, BIOS, and the Windows loading screen. Ironically, as long as the drivers are not loaded, I can view my desktop fine - I just cannot use Aero or use anything that needs 3D acceleration.