Constant Nvidia Driver Crash on Startup

Toggle sidebar Toggle sidebar
S

Sandstone

Honorable
Dec 18, 2013
6
0
10,510
This has been a problem for several months but had just gotten much worse.

99% of the time on startup my system will crash in under 2 minutes if it starts at all. Usually I get a blackscreen before I even see my desktop and the monitors say "No Display Input". If I get to desktop I'll get an immediate "Nvidia Kernel Mode Driver has stopped responding and recovered", and that will mean any program I launch will lead to a complete crash. Event viewer shows nothing about the crashes, only telling me that the system didn't shut down correctly. If i get the lucky 1%, the system is perfectly fine and I simply don't shut it off for a month or so.

I have a GTX 970 with up to date drivers clean installed.

I've reseated the GPU and memory, confirmed that the memory times are correct in BIOS, and changed the PCIe slot to Gen2, all at the recommendation of others with similar issues. Nothing has improved. The issue is largely ignorable but as soon as I need to restart for any reason I'm out 1-2 hours attempting to get it restarted. I'm typing this after spending 3 doing just that.

All help and advice greatly appreciated.
 
Solution
S

This may sound crazy but try both sticks of RAM (Assuming you are in a 2x8GB config) separately in slot 1 and see if the crashing still occurs. Also try underclocking your cards core and memory clock by 100Mhz in Afterburner to see if it runs stable like that. Try those in whichever order you'd like but I'd recommend the underclock first as it's easier than opening the PC.
Sort by date Sort by votes
S

Sandstone

Honorable
Dec 18, 2013
6
0
10,510
GTX 970
i5-4430@3.00
16gb RAM
ASUS Z87-A mobo
Seagate 750w PSU, I'm not sure on the model. Its 7+ years old.

It doesnt shut off on its own, only if I shut it off myself, or need to restart to install or update software. But when that happens I'll not be able to get it going easily again.
 
Upvote 0 Downvote
S

Sergeant_Sneaky

Admirable
Mar 25, 2015
3,193
0
7,960

Check your GPU temps are normal. What OS are you on? If you have a friend with a suitable PSU to try that would be helpful for elimination.
 
Upvote 0 Downvote
S

Sergeant_Sneaky

Admirable
Mar 25, 2015
3,193
0
7,960

This may sound crazy but try both sticks of RAM (Assuming you are in a 2x8GB config) separately in slot 1 and see if the crashing still occurs. Also try underclocking your cards core and memory clock by 100Mhz in Afterburner to see if it runs stable like that. Try those in whichever order you'd like but I'd recommend the underclock first as it's easier than opening the PC.
 
Upvote 0 Downvote
Solution
S

Sandstone

Honorable
Dec 18, 2013
6
0
10,510


Yes! I'm running 4x4 RAM so I couldn't try that, but underclocking my GPU seems to have fixed it entirely. I've just restarted a few times and it worked first try each time. Thanks a million.
What does this mean though? Not enough voltage to the GPU to support default speeds? CPUID's HWMonitor says 0.86V to the GPU, I've never overclocked or known anything about voltages to know if that is right.
 
Upvote 0 Downvote
S

Sergeant_Sneaky

Admirable
Mar 25, 2015
3,193
0
7,960

Okay, that may only be a temporary fix I'm afraid, if the card is degrading but if it is just a matter of unstable stock clocks then it should stay okay. You do have the right to return the card for not working as intended though. You are probably spot on, I'm guessing the voltage isn't high enough at stock, for whatever reason.
 
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