Possible GPU issues? Driver keeps crashing

Associate
Joined
26 May 2008
Posts
1,210
AS the title says I seem to be having some issues with my GPU.

I've got the latest driver from nvidia installed and I've been getting a lot of system problems since installing. The issue can happen at any time, as I was typing this I opened up the start menu and it happened for instance. Generally I get something like this:

1. Screen goes blank
2. Screen turns back on
3. System either freezes completely with no response from mouse / keyboard at all.

or I get a nice error message saying my graphics driver has stopped and been recovered.

I've done the following:
Ran spybot - No results bar one or two tracking cookies
Ran full malware bytes scan with no results
Ran full Microsoft defender with no results.
Went back to the previous driver - still keeps happening
Tried the latest driver again - still keeps happening

Any suggestions?


Edit:

Probably should include my build:
EVGA GTX 570 - NO O/C
i7 930 - NO O/C
6gb corsair ram
1tb samsgun spinpoint f3
64gb crucial SSD
TRUE cooler
XFX 750 PSU Black Edition.
 
Last edited:
Only thing i can suggest and no idea if it will help at all is a fresh windows install, it normally fixes everything for me

I did a fresh windows install a few months ago, around the start of Decemeber when I was having issues with random shut downs, issues came back it seems
 
Thanks I'll check it out. Hopefully its not that because this monitor is brilliant

CPU temps are all fine, sitting at around 40C atm on all cores.
 
Only thing coming up in it is

"Display Driver nvlddmkm stopped responding and has successfully recovered".
Source: Display
Event Id: 4101

(Its crashing even getting there)
 
Googling that points to a GPU fault, heating issue, any overclocks on the card ? if yes take em off and try it.

Cards 100% stock. I'll give the fan and intake a good clean out when I get home, cleaned it last month though when I installed a new CPU heatsink.

Any other suggestions of what it could be or what to try.
 
Ok so I've downgraded to a previous version of the drivers from a few release ago 310. 90 and its still happeneing.

Just about to run furmark now and see what happens.

Yeah I can't even run furmark. My Screen goes black when trying to run it and nothing
 
Last edited:
Yeah its a real pain, this started happening this seemed to happen all of a sudden. I was playing hitman fine and then it started happening all of a sudden.

I've submitted a ticket with EVGA so lets see what they say
 
Idle temp for GPU is 70C? That's more like what the load temperature should be :p

Yeah the fan is going crazy atm. Going to take it out of the system tomorrow and clean it out again and try it again. Seems to be a common enough issue on the EVGA forums though
 
Shouldnt be anything running in the background using it at all.

So I took the heatsink out, cleared all dust from it (again just to be sure), reinstalled it and the error keeps happening :(

Though its not has hot. Temp is around 50C now
 
I don't know what else to suggest to you, is this card still under warranty (though that might be void now if you removed the heatsink)

Cleaned it as best as I could without removing the heatsink.

Thanks for the advice just waiting for EVGA to get back with RMA details
 
I'm going to check it out this afternoon, I think I was having an issue with windows which was causing the GPU to keep running, it was sitting at 38c this morning when I was using it.
 
Hi,

we do recommend to not use the Furmark Tool on our 500 Series graphics cards!
The driver kick you can see is a security protection of the card, if this detect any issue it kicks the driver to make sure the load is immediately off the card to make sure it do not get damaged. This error can be occured by software or faulty hardware. If you have done a new installation of the OS it do sound like the error is hardware related. This can technically be occured by the card itself, faulty motherboard or faulty power supply. The most simple way to rule out if the fault is with the card is to test it in a different system or another card in your system if you are able to. You can also try to downclock the card with the precision Tool by 20% and have a look if the error still occur.

If you did not get a response from support so far you can send me an email to [email protected] then I can also open a RMA ticket if you wish to.

We do also allow to remove the cooler without void the warranty so you can change the termal paste on the GPU. This can also help to lower the temps.

Cheers

I've just done a fresh install of windows and the issue has completely disappeared.

I'm going to run the EVGA OC scanner as well as, The valley benchmark and the haven benchmark tools.

Is there anything else you would recommend running to test the card?
 
Back
Top Bottom