Display driver has stopped responding. Warranty??

warvek · 6 Apr 2013 at 20:20

Hello Guys,

I wander if you ca help me please I've got a Gainward GeForce GTX 580 "GOOD Edition" 1536MB GDDR5 PCI-Express Graphics Card and for some games I get kicked back to the desktop with the following error: "Display driver has stopped responding"

I've read that you can fix this by increasing the cards voltage by 50mV BUT I've also read that doing so OR underclocking your card will void the warranty.

Taking this into account it makes me think that this model of card have a manufacture fault, where do I stand on getting them to replace it??

Thx in advance,
Jarren

SeraphX · 6 Apr 2013 at 20:25

Nonsense!!

slightly clocking your card will not invalidate warranty regardless of the stupid blurb you have read online.

obviously If you try to fry your card by overclocking with stupid volts being pushed through it, your warranty will be null and void, but a little adjusting cannot harm anything.

Gregster · 6 Apr 2013 at 20:51

Install EVGA precision X or Afterburner and where it says core voltage, add 0.10mV and leave everything else at stock. This should cure the problems and will NOT invalidate your warranty.

shankly1985 · 6 Apr 2013 at 21:39

Yes your warranty is safe, plus the fact they no proof you even touched it anyway, unless you fry it witch wont happen with a tiny adjustment.

dazzerd · 6 Apr 2013 at 21:58

At the end of the day if the card is not working as advertised, you should not have to over-volt it yourself to remedy the issue.

Ask Gainward if there is a bios flash available which will fix the issue. If you are not happy to over-volt it yourself then request a replacement.

shankly1985 · 6 Apr 2013 at 22:53

dazzerd said:
At the end of the day if the card is not working as advertised, you should not have to over-volt it yourself to remedy the issue.

Ask Gainward if there is a bios flash available which will fix the issue. If you are not happy to over-volt it yourself then request a replacement.

Happens a lot tbh
If you Overclock the card yourself you get Driver crash you know right away its not a stable Overclock so tiny bit more voltage fixes it.
Some times the cards come pre-overclocked and again not enough voltage.

This mainly happens to pre-overclocked cards and the non overclock work fine unless you fail at overclocking.

warvek · 6 Apr 2013 at 23:07

Thx for all the advise everyone but I'm with Dazzerd I shouldn't have to manually change the voltage of my card. I'm going to run it passed there support desk but I'm not holding my breath

Marine-RX179 · 6 Apr 2013 at 23:33

One thing though...can you give detail of the PSU you are using? Brand and model etc.

warvek · 7 Apr 2013 at 01:21

Marine-RX179 not sure on my PSU I'll have a check

Marine-RX179 · 7 Apr 2013 at 07:14

warvek said:
Marine-RX179 not sure on my PSU I'll have a check

Yea best you have a look at that first. The last thing you'd want is send the card back and then return back to you as "no fault found" and then you get charged testing fee or something for it. It would also be good if you can test the graphic card in a different PC (with a decent PSU) to see if the same fault develop.

warvek · 7 Apr 2013 at 09:05

It's a 750w so imo should be fine, I haven't got another PC to try it in :-(

but I'm sure it's the model of card as it only seems to do it with some games. I've done a load of googling now and there's a lot of ppl saying you can either underclock the card or raise the voltage a little but thanks for the heads up.

Marine-RX179 · 7 Apr 2013 at 09:14

warvek said:
It's a 750w so imo should be fine, I haven't got another PC to try it in but I'm sure it's the model of card as it only seems to do it with some games. I've done a load of googling now and there's a lot of ppl saying you can either underclock the card or raise the voltage a little but thanks for the heads up.

"750W" doesn't mean a lot...the important thing to know is whether it is of a decent brand and a quality unit or not. Lots of generic brand PSU (or some of us refer as the ticking time-bomb) have extremely poor output over the 12v rail (which the core PC components runs on i.e. CPU, graphic card etc) and won't even provide half of the "total output" claimed over that rail...and that's not even taking into the consideration of generic brand PSUs often don't even output as much as their own claim on their label.

You really need to check the brand and model of your PSU.

shankly1985 · 7 Apr 2013 at 12:10

Sorry but you leading him away from the problem. If the problem was with his psu he would have worst problem than just a driver crash, the hole pc would lock up.
I know this from experience with a faulty psu. His problem is with not enough voltage to a pre factory Overclock.

woody79 · 7 Apr 2013 at 12:16

i have the same card and was getting the same problems, formated about 6 times but kept on doing it then i decided to muck around with the psu wires in my case thinking i had a loose connection somewhere, stripped all me hds out to bare bones and all started working again then put it all back together and the problem went away been stable for 3 weeks now lol

Marine-RX179 · 7 Apr 2013 at 12:26

shankly1985 said:
Sorry but you leading him away from the problem. If the problem was with his psu he would have worst problem than just a driver crash, the hole pc would lock up.
I know this from experience with a faulty psu. His problem is with not enough voltage to a pre factory Overclock.

I don't see how asking him to check what PSU he's using (just opening the side-panel of his PC) can be too much of an hassle, or stating it would be best if he could test the card in a different PC (if possible) before RMAing being irrelevant.

Also, it's not just about whether or not the PSU have enough output for the card...but also how stable is the 12v rail as well- jumping up and down between 11v and 14v certainly won't help with stability.

shankly1985 · 7 Apr 2013 at 12:57

Marine-RX179 said:
I don't see how asking him to check what PSU he's using (just opening the side-panel of his PC) can be too much of an hassle, or stating it would be best if he could test the card in a different PC (if possible) before RMAing being irrelevant.

Also, it's not just about whether or not the PSU have enough output for the card...but also how stable is the 12v rail as well- jumping up and down between 11v and 14v certainly won't help with stability.

Dont take my last post wrong way, I not meaning to be horrible here.

Its that change between 11v and 14v like you said will just make a PC black screen and not recover. A failing PSU is a good sign if your PC just keeps locking up and black screen and only means of recover is to power off and on.

OP issue is GPU voltage, because he is able to recover after the driver as failed.

Marine-RX179 · 7 Apr 2013 at 13:16

shankly1985 said:
Dont take my last post wrong way, I not meaning to be horrible here.

Its that change between 11v and 14v like you said will just make a PC black screen and not recover. A failing PSU is a good sign if your PC just keeps locking up and black screen and only means of recover is to power off and on.

OP issue is GPU voltage, because he is able to recover after the driver as failed.

I'm not disagreeing with you that it is likely to be a GPU voltage, but in addressing what you said, I have to point out that back in my "didn't know better" days of using generic brand PSU, the 12v rail will hit as low as 11.xx and as high as 12.80 according to HW Monitor, despite no crash occurred.

And as I said I don't see any harm for OP to get to know the hardware he's got in his PC. Take the side-panel off and check the label on the PSU would take like what...1-2 minutes? It's not like I'm asking him to do some more hassle and time-consuming like removing the CPU cooler and reseating it and the memory as well etc...

Let's say if OP open the side-panel and found a...say "WinPower 750W" PSU inside, will you still be confident to say it is problem with the card not the PSU? (Don't take this wrongly, I'm just saying).

One thing I have learned from troubleshooting PC problem is not to jump to conclusion, and gather as much information as possible (particularly if from sources that require minimal effort). For all we know, it might not even be a graphic card voltage problem, but a simple unstable driver issue. Come to think of it, why hasn't any of us recommended trying a different version driver?

Marine-RX179 · 7 Apr 2013 at 13:20

Ignore.

shankly1985 · 7 Apr 2013 at 13:22

Marine-RX179 said:
I'm not disagreeing with you that it is likely to be a GPU voltage, but in addressing what you said, I have to point out that back in my "didn't know better" days of using generic brand PSU, the 12v rail will hit as low as 11.xx and as high as 12.80 according to HW Monitor, despite no crash occurred.

And as I said I don't see any harm for OP to get to know the hardware he's got in his PC. Take the side-panel off and check the label on the PSU would take like what...1-2 minutes? It's not like I'm asking him to do some more hassle and time-consuming like removing the CPU cooler and reseating it and the memory as well etc...

Let's say if OP open the side-panel and found a...say "WinPower 750W" PSU inside, will you still be confident to say it is problem with the card not the PSU?

One thing I have learned from troubleshooting PC problem is NEVER to jump to conclusion, and gather as much information as possible (particularly if from sources that require minimal effort). For all we know, it might not even be a graphic card voltage problem, but a simple unstable driver issue.

Good points but I would like to just point out one thing, never trust software for Voltage reading, they always way off. For a true voltage read its better to use a voltage tester.

Take HWINFO64 for example I use a voltage of 1.131v from Trixx for my GPU overclock, but on HWINFO64 its reads 1.090v

Marine-RX179 · 7 Apr 2013 at 13:32

shankly1985 said:
Good points but I would like to just point out one thing, never trust software for Voltage reading, they always way off. For a true voltage read its better to use a voltage tester.

Take HWINFO64 for example I use a voltage of 1.131v from Trixx for my GPU overclock, but on HWINFO64 its reads 1.090v

I think it is a bit different when it comes to the graphic card voltage. For some reason, the voltage we adjust in these overclocking software don't really seem to match the reading in GPU-Z and HWMonitor etc, so I sometimes wonder if they are give readings in different way (not saying either of them is wrong, but may be their voltage represent slightly different things?).