Graphics Drivers Crash with Snow On Screen, Computer Freezing

LewisOakwood

Reputable
Jul 20, 2014
7
0
4,510
-------------------The Problem:-------------------
Lately my computer has been black screening and Showing snow like in the image below and then giving me the error " Display Driver has stopped responding and has recovered" or black screening, Snowand freezing all together. Sometimes it only takes and hour sometimes it takes 4 hours, its pretty random and no specific program seems to be causing the issue.

*******EDIT*******
I have been running on board graphics for the past 4 hours with no crashes, so i'm assuming it is an issue with my graphics card. Now my question is it the hardware or the software. When the gpu crashes I see no voltage spikes or anything that looks abnormal so I don't know how to tell besides RMAing the card but I'd rather avoid that if I can. Any ideas on how to know for sure?

-------------------I have tried:-------------------
1. A Clean driver wipe in safe mode with Beta drivers, current drivers and 3 older driver versions.
2. I have tried changing the PCI slot that my 680 is in.
3. Ran a full virus and malware scan. Also checked every process running on my computer.
4. Ran a Microsoft FIX IT relating to "Timeout Detection".
5. Pulled the card apart and blew out dust/cleaned, replaced thermal paste.

-------------------Some Important Things to Note:-------------------
1. I do not believe it is a heating issue because it has crashed at 30C and at 60C. It has crashed when I'm running firefox or chrome or if I'm playing a video game.

2. I did go a clean wipe and format of windows 2 weeks ago and this was not happening before then and this started 2 days after the format and wipe. I assume this is just a coincidence but who knows.

3. When I freezes All sound freezes too. Like if i'm playing a song it will just freeze it into a loop of the last .1 seconds.

4. My CPU is overclocked to 4.1 GHz, Base clock is 3.5GHz. It has been running like this since I got it over a year ago and temp never goes over 60C and voltage never over 12.9V

5. I can leave my computer running over night with the screens off and it will not crash.

-------------------My Specs:-------------------
OS: Windows 7 Ultimate 64Bit
Motherboard: ASUS P8Z77V Deluxe
CPU: 3770K - Liquid cooled
Graphics Card: GTX 680 - Liquid cooled
Main Drive: 120G Intel Solid State
Backup Drive: 1TB Caviar Black
Screens: 1x HDMI, 1xDVI
PSU: 1050W - Corsair HX1050

-------------------Pics-------------------
Graphics Drivers recover error (this is someone elses picture, i have tried many different driver versions):
8c43a94e0a2b8066ca91d811c95c04df.png

Snow on Screen During Most Recent Crash:
95dbc0e32904c53beadcc2ca6f511aa6.png

When I Restarted Windows after the most recent crash. This happens about every other time I have to restart but it is always gone at the windows log in screen.
2990064ad2c680dcccfe5dc080071139.png

GPU Stats At most Recent Crash
e9d552689d602dff6aa45099f083d4e0.png

CPU Stats At most Recent Crash
68d2873a2e3a0f893c23600cb2f0e7bd.png
 

LewisOakwood

Reputable
Jul 20, 2014
7
0
4,510


I will try this but if i am not using my graphics card then it is not running the Nvidia drivers correct? Which means if it is a driver/software conflict it may not crash because i'm not running the drivers right?
 

kenghtt

Honorable
Oct 23, 2013
137
0
10,710
Yes that is true, but as of right now we dont know what is causing the problem. it may be your cpu, psu, ram...etc...so by checking if it works on your motherboard graphics, atleast we will know if it is your gpu or not :)
 

LewisOakwood

Reputable
Jul 20, 2014
7
0
4,510


I have been running on board graphics for the past 4 hours with no crashes, so i'm assuming it is an issue with my graphics card. Now my question is it the hardware or the software. When the gpu crashes I see no voltage spikes or anything that looks abnormal so I don't know how to tell besides RMAing the card but I'd rather avoid that if I can. Any ideas on how to know for sure?
 

john TJ

Distinguished
Jul 12, 2014
840
0
19,360
if there is no change after driver re instalation the it might be the gpu thats causing the issue ,check that gpu in another machine and check wheather it works
seeing that black screen i dont think that is a software issue

 

LewisOakwood

Reputable
Jul 20, 2014
7
0
4,510


I am running a Accelero Hybrid on it in the Option D configuration with 2x 252CFM 120mm in a push pull config on the radiator (loud as hell but i dont mind). However i also have 2 x 150CFM 120mm fans on the side that blow into the case directly on the card. and then a 200mm that exhausts out the top.

There are also mini heat sinks on most of the Vram chips like this:
ab9ac0345aee84c2bba9dd64aceda242.png


Pic for reference (only included gpu cooling):
81996a639c382b92296adb85dbc893e0.png
 

LewisOakwood

Reputable
Jul 20, 2014
7
0
4,510


Sadly I do not have another computer to test it in. I am trying another clean wipe and re install of drivers right now, I figure it can't hurt.
 

LewisOakwood

Reputable
Jul 20, 2014
7
0
4,510


I ran a Render Test to check if its the RPMs are changing, they are changing a little but not a huge amount but then again the card inst getting hugely hot under 100% load. It usually levels off at just under 60C no matter how long I run it. I have ran the test for a 15 min now without crashing.

This is the Fan at Idle well sorta, I had just ran a render test for about 2 min restarted the program and then let it cool for about 3 min:
e4f12b28a680682978dd900b276cad04.png


This is the Fan after a render Test has been running for 5min:
a9c5a053018c1052dad69a2e4409251c.png


********Edit********
For the record I have clean wiped and re installed the latest drivers in safe mode before this post. Also I noticed the card was sagging a bit in the PCI Slot (not bad but enough) so I Zip tied through a screw hole to to raise the end up and make the card sit level in the case.

Also I figured out how to Manually set the Fan to 85%, Rpms seem to be ramping up just fine:
25587d2a1659940eafe9e4da06e32fe7.png

 

LewisOakwood

Reputable
Jul 20, 2014
7
0
4,510
After 25 Min running the GPUZ Render Test:
c505f2187badfbe7298246923ad84361.png


Going to run a EVGA OC Scanner Burn in and see what the results are there. I will post them here.

*********EDIT*********

5 min into EVGA OC Scanner GPU Burn Test:
a98090f48fcc326d6b305fcfb7495930.png


Seems to be holding Steady at 73C/74C, The only thing I see possibly wrong is the big green PWR bar, I assume it means there is an issue with the power because the TDP is at 110% but I don't really know.
edd56c2536eefd55225f682973abec08.png

 

kenghtt

Honorable
Oct 23, 2013
137
0
10,710
so most likely it is your GPU. How long have you bought it? Do you have warranty? Is it possible for you to bring it in and replace for another one?

Do your clean install again and see how it works. And download the correct drivers from the Official Site. If it still freezes, then its most likely a Hardware Malfunction. Which then you either can bring it back to the store or buy a new one :)
 

kenghtt

Honorable
Oct 23, 2013
137
0
10,710
I also wouldn't rule out your PCI slots either. If you have someone who is willing to let you borrow their GPU to test on your pci slots, that would be awesome too. If their GPU doesn't work on your pc but works on their own PC, then it would be your slots that are at fault. :)

(Most likely this won't be the issue, but it's still a possibillity)