PNY Geforce GTX 660 crashing at 64 degrees.

Toggle sidebar Toggle sidebar
R

Reaperofskil

Reputable
Jul 7, 2014
6
0
4,510
I had first purchased and built my computer ~ 8months ago and it ran great. I had purposefully downloaded stressful games that gave my previous system trouble such as Planetside 2, Skyrim, and witcher 2 just to see what it could do and they all ran great (though not always at max settings). but then i mainly played league of legends afterwards which is not an intensive game at all. About a month ago i installed a game called Black Gold and then trouble arose. The game would consistently crash during either character creation or the massive 200v200 pvp fights giving the "nvidia kernel mode driver crash" and closing the game every time. At first i thought it was just bad coding on their part as it was an alpha game. But then i tried playing Planetside 2 again and got the same crash as well, and then Witcher 2, Skyrim, and Divinity: Originial sin all did it as well. At first i thought it was just a driver issue so i had clean installed the newest driver, then the original driver i got with the gpu disk and they both still caused the crash. Then i got gpu-z and cpu-z to see if something was noticably wrong. The only thing that looked even close to an issue was that before the crashes the gpu would be close to 100% usage and so i thought that might be the cause. So i tried lowering the settings dramatically and the gpu usage went down to ~90% and the games lasted longer but would inevitably crash all the same. So i thought maybe the stock settings on the card got altered or were unstable so i installed msi afterburner to make adjustments. None of the voltage, clock speed, or power limit changes prevented the crash but i did notice that every crash happened when the gpu hit 64 degrees. So i reset to stock settings, tried, and it still crashed. Then i forced the gpu fan speed to constant 74% and then the gpu temp never got above 60 degrees and the crashes stopped though my fan is now really loud.

Computer specs:
Case: Coolermaster Has 912
Motherboard: Asus P8z77
Psu: Evga 600W bronze 80+
Cpu: intel i5- 3570k
Gpu: PNY Geforce Gtx 660 (2gb)
Harddrive: Kingston SSDnow V300 Series 120gb SSD

Does anyone know what might be causing the gpu to crash at such a low temp or how to fix it other than keeping the fan speed at a rediculous speed and speeding up the degradation of my card?
 
Solution
A
Did you attempt to overclock the videocard or adjust the BIOS on the videocard?

It sounds like the card itself is just dead. Quickest troubleshooting step is just to toss it in another system and run a game, see if it crashes.

Easiest solution is just to RMA the card. They do die often enough, it's not uncommon.
Sort by date Sort by votes
well one thing you don't know is the first time the game stressed the card did it do damage to it ?? you may have pushed the card well over a limit and caused something to fail like the thermal tape/paste may have melted away and that part may be now be way overheating but the temp sensor may not be picking it up [for example]
 
Upvote 0 Downvote



Maximum GPU Tempurature (in C)
97 C
http://www.geforce.com/hardware/desktop-gpus/geforce-gtx-660-oem/specifications

Really? This isn't the issue, it didn't melt unless he over volted the card.
 
Upvote 0 Downvote


i guess i failed to mention this. when messing with gpu-z, cpu-z and msi afterburner i had used both heaven benchmark and furmark during the tests so that it was the same thing running when the crashes occured. Still crashing.
 
Upvote 0 Downvote


What error are you getting on the crash? BSOD? Application failure, driver failure?
 
Upvote 0 Downvote
my evga card did this is why I brought it up temp sensor did not pick it up the 2 cards I had in sli the other one still had the thermal tape well in place the other was melted away the card failed just like he is saying.. but it may not be the case here but it sure sounds dang close as to what I seen with mine..
 
Upvote 0 Downvote


Do you notice any artifacting while playing?

I would try completely removing the driver and reinstalling.

The PSU has a good review; http://www.techpowerup.com/reviews/EVGA/600B/11.html
I don't think it's a PSU issue.
However; I would make sure you are using a proper PCIe connector (not a converter of any kind) and make sure the videocard is in the proper x16 slot on the motherboard. Some motherboards have multiple slots for SLI, but only one is for a dedicated single card.


Did you check your CPU temperatures as well?
 
Upvote 0 Downvote


While playing there were no artifacts until seconds before the crash and they tended to vary. Sometimes there would tears, sometimes miscolored pixels, sometimes sections going black. I have already tried clean installing both the latest driver and the driver on the cd that came with the gpu. Cpu temperatures were measured as well and though remained below 60 degrees the exact value at crash varied. I have also tried both of my x16 slots on my motherboard and neither prevented the crash.
 
Upvote 0 Downvote


i dont know what vrm refers to, please explain?
 
Upvote 0 Downvote
Did you attempt to overclock the videocard or adjust the BIOS on the videocard?

It sounds like the card itself is just dead. Quickest troubleshooting step is just to toss it in another system and run a game, see if it crashes.

Easiest solution is just to RMA the card. They do die often enough, it's not uncommon.
 
Upvote 0 Downvote
Solution


overclock the video card. also im going to RMA the card tomorrow but was hoping for a solution that wouldn't take a couple weeks to resolve.
 
Upvote 0 Downvote


I actually mean VRM as in "Voltage regulator module". They provide the power for the GPU and tend to get really hot. some gpus dont even have a heat sink on them and temperature can reach 100c fast and cause a game to crash even though it appears the card is running cool. My old evga ssc 660 didn't even have a temp sensor for the vrm but it was well cooled as it was one big block of metal cooling everything, so I had no problems. Maybe the thermal tape between the sink and the vrms aren't making good contact. As the gpu reaches 100% load, the strain on the vrms increase. as you up the voltage, the temps on the vrms drastically increase. Do you hear a coil wine coming from the GPU when playing games? It could be an indication of a hot running vrm. even at stock, this is possible.

Here is an attached photo from google images. It should look a lot like this. IF you dont have a heatsink on them it could very well be the cause of the issue. If you do, the same could apply. It could be starved of air because of lack of good contact. I mean, they even manage to botch thermal paste applications on the die in the factory.
22104_nvidia-gtx-760-vrm.jpg


and this is an exampe of a gpu with the sink applied.

asus-gtx770-pwm.jpg



In the end, it might just be better to use the warranty. Overheating vrm is as good a reason as any. trying to fix it yourself will most likely void the warranty and I'd want piece of mind knowing my card will last a while.

Can you try something and enable vsync in games that usually cause the crash? if that fixes it then still send it in haha.
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom