My 6800GT is dead

G

Gargoyle_Hunter

Limp Gawd
Joined
Feb 9, 2002
Messages
305
R.I.P. BFG 6800GT OC
I got it tuesday and installed it. I ran it through 3dmark03 a couple times, played some unreal2k4 then overclocked it to 415/1.1 using that coolbits2 crap. The autodetect was showing 417/1.09 so i thought I would be safe. It ran fine and my 3dmark score went up 1000 points. I left it running overnight (as always) and in the morning was greeted with a locked up box. Rebooted and was surprised to see some artifacts all over, but in nice square patterns. I've tried to do everything I could to bring it back (clocked it down to stock, uninstalled, re-installed an old card and installed again, etc...) but every time I tried to get into windows, it kept getting worse. Now it won't even boot. BFG doesn't even know when they will be able to get me an RMA and Best Buy is worthless to get info from so far. And Doom 3 is almost a week away... Dammit. Here's a gory pic for the jackals ;)
6800screen.JPG
 
dude, that blows. nice monitor though. maybe let it rest for a day - but i think it's hosed from the OC. did you run benchmarking overnight?
 
tranCendenZ said:
So wait you OC'd your card and now its permanently stuck playing Kingdom of Kroz? ;)
Click to expand...

Guess I did get a free game out of it :p

I'm half tempted to return it to BB and take my money to eVGA. Anyone know any good deals right now?
 
That is basically what my screen looked like with my defective BFG 6800GT OC. But I never had it in my computer for more than 10 mins and the entire screen wasn't like that, only where the mouse cursor was over.

I'm waiting on my RMA from them myself. They told me today that they had no 6800GT OCs in stock to send but yesterday they told me mine would ship out today. A supervisor is supposed to call me by Monday with more information as they seem to be the ones who handle the RMAs and the calltakers have no information about them really.
 
If you live near a frys that has the BFG 6800GT in stock. Buy it and return the bad card to best buy. I got a bad card from BBY. Talked to BFG and they said they didn't have any cards on hand for an RMA for at least a week. Took matters into my own hand and bought another card and returned the fubared card. If you got in on the 299 deal you could be evil and return the bad card back to frys.
 
Damn. I'm a little worried about running my PNY GT at 421/1.15. It doesn't seem to be running too hot, though. Question about the temperatures in the driver control panel. What is the "Ambient Temp" refering to?
 
Why is it that people feel it to be absolutely necessary to OC? I mean, I could see doing it when games start to chug down to 30fps, but other than that you aren't getting all that much gain out of OCing, especially with a high end card. I guess its the reasoning that if one CAN overclock, why the hell not. *shrugs* oh well, i'm not having any problems with my GT OC @ default.

Good luck finding a replacement card and be careful when and if you overclock it.
 
planetary said:
What is the "Ambient Temp" refering to?
Click to expand...
There's a temperature sensor on the card itself. As the case and card warms up, that ambient temperature goes up. It will be higher than the case temperature reported by your motherboard.
 
Agent4054 said:
Why is it that people feel it to be absolutely necessary to OC?
Click to expand...
Because we are [H]ard. Otherwise we'd be talking about how well the Sim's plays on Dell's integrated graphics over AOL.

Due to how fast this thing degenerated once the problem showed up, I really think it was a problem with the card. Some of the pads on the ramsinks weren't even covering the entire chip. I'm wondering how much that coolbits2 crap was involved with this also. It seemed to accellerate the problem once I entered that in.
 
There a lot of people OC'ing their GT's and this only the 2nd time I've seen this, both with BFG's. Since they take off the reference cards when they get them from Nvidia (you guys know ALL these GT's come from the same place) to put on the AS5, who knows if they damaged anything during this process. It could be a bunch of low-wage factory workers doing the HSF R&R.

I see everyone bragging about BFG support, but that is incredibly weak that they don't have stock set aside for RMA's.

eVGA keeps stock set aside for RMA's. True it's only a 1 year warranty, but at least you can get your card replaced immediately within that timeframe.
 
Well, you take a risk when you overclock. Immediately declaring that the card is "bad" and trying to return it is a little presumptuous, because for all you know you caused the problem with your overclock and your overnight 3DMark loop. Since the BFG 6800 GT is already overclocked out of the box, maybe it's really not meant to go beyond that, or BFG would've overclocked it to the speeds you reported.
 
Coolbits 2..

Eats...

Ass.

biohazard.jpg



It removes SO many options it's not even funny. One of the important function that it removes is 16xAF..

EDIT: And you don't want a beta program in total control of your video card fans (which Coolbits 2 does have control over).

coolpis.jpg
 
So, you buy a card, overclock it, kill it, and, now want to RMA it? :rolleyes:

Can I buy a car, and crash it into a wall, and exchange it also? :confused:
 
sc0tty8 said:
So, you buy a card, overclock it, kill it, and, now want to RMA it? :rolleyes:
Click to expand...
I would agree with that if you were talking about a CPU since it's easy to kill it through overvolting and overclocking, but the card shouldn't have died through unmodified overclocking. The card has protection on it to prevent overheating damage. It obviously didn't work or some external factor (PSU or motherboard) killed the card.

Your car crash analogy fits better with *physical* damage. If you buy a car and drive it hard, within the way it was designed and it breaks, who's fault is it?
 
Gargoyle_Hunter said:
Because we are [H]ard. Otherwise we'd be talking about how well the Sim's plays on Dell's integrated graphics over AOL.
Click to expand...

Too true. What in the hell was I thinking... *hits self with a zalman heatsink* Ahhh much better. I keep forgetting about the 5 Vantec Tornados I have in me case. Factory defaults are for wimps!
 
pxc said:
I would agree with that if you were talking about a CPU since it's easy to kill it through overvolting and overclocking, but the card shouldn't have died through unmodified overclocking. The card has protection on it to prevent overheating damage. It obviously didn't work or some external factor (PSU or motherboard) killed the card.

Your car crash analogy fits better with *physical* damage. If you buy a car and drive it hard, within the way it was designed and it breaks, who's fault is it?
Click to expand...

I think if we want to use the car example, you could basically use this analogy. It's like buying a brand new car with a manual transmission, and then driving it at 9,000rpm instead of shifting when you are supposed to.

The graphics card in question wasn't driven hard within the way it was designed, it was pushed beyond that.

Either way, you really shouldn't be able to RMA that card.
 
DigitalEmperor said:
The graphics card in question wasn't driven hard within the way it was designed, it was pushed beyond that.
Click to expand...
I disagree. The card should clock down when it's running too hot like it was designed to.

If a car would cut fuel when the engine revs too high, then you could compare the two situations. Otherwise it's still not a similar situation.

If it's the memory that's fried, i'll give you that point.
 
I need to clear a little confusion up. I did not have 3Dmark running all night. I meant the computer was running all night. The machine was just idling when it locked up the first time. Sorry about making that clear in the first post.

That being said, I have killed enough cards in my time to know when it was something I did. Did I do something that was out of spec, yes. But from the way this acts, there seems to be more than just some corruption of the memory from overclocking. If it was just from the overclock, the card would not get worse when it was installed in a different machine at stock specs. That screen shot was taken during POST, before the drivers and anything would have a chance to affect the card. I do believe I accellerated a problem, but I dont think I caused it.
 
pxc said:
I would agree with that if you were talking about a CPU since it's easy to kill it through overvolting and overclocking, but the card shouldn't have died through unmodified overclocking. The card has protection on it to prevent overheating damage. It obviously didn't work or some external factor (PSU or motherboard) killed the card.

Your car crash analogy fits better with *physical* damage. If you buy a car and drive it hard, within the way it was designed and it breaks, who's fault is it?
Click to expand...

I have killed many a card with out modding it to be over clocked, I have simlply added a orb and as3 back in the day, tell that to the kyro2's, radeon 8500's and the gf3's I killed, that, I do not have to mod them to kill them :D All I did was apply some new thermal paste, and bumped up the mhz and I got the same thing the thread starter got. Did I rma then? No. I simply put them on the shlef, and, every time I went to OC a video card again, I looked at them, and decided not to:D
 
Since ur 6800gt died, and u can't get another, why don't just get the refund and wait for the fall refresh, PCI-X and all that? ;)
 
I got the BFG 6800Gt 256 OC last week.. and left it on for a good couple days straight.. my case idle temp is a lil high.. but i haven't noticed any problems with the card like u did..
My old 9700pro that did die makin me get the 6800gt had heating problems like u but its not the gpu like some have said its the ram overclocked too high and perm burned
This screen would pop up any time i started up Pandora Tomorrow and just idle'ing in windows for about 20mins would show it too..
Garble.jpg
 
You must log in or register to reply here.
Back
Top