Video driver keeps crashing and recovering until the system crashes

Apr 29, 2018
4
0
10
Hello. For some time now I've been having issues with my video driver crashing and rebooting when certains programs open (worst offenders seem to be DXDiag, Steam Library and the Discord standalone. The browser version works just fine). It had stopped for some time but I opened up League of Legends tonight and at the Select Champion screen it happened again.

I think it's a bigger issue than just the driver, because the system itself crashes after a few of these.


Nero Home also crashes at startup but that's a program I haven't used in years and has been doing that far longer than this problem has appeared.


Programs that open at the start up: Hamachi, Steam, Discord, CC Cleaner, NVidia Control, Daemon Tools Lite.

I have tried reinstalling Discord and Steam; rebooting the system; rolling back the driver and going back to a new one.





My specs are:

8GB Ram, i5 3570K, 1050TI (raw, not overclocked or anything). Driver is 391.01, OS is Windows 7 Ultimate 64-bit.


In a much more intensive game like BATTLETECH, it works just fine, never having crashed once in 8 hours of gameplay.

Other programs that seem to be causing trouble are Chrome, WMPlayer and foobar2000, so maybe it has something to do with an audio codec?

Thank you.

UPDATE: It also happened when watching some MP4s, but not others. It also happens when watching a stream on Twitch.tv and a video on YouTube at the same time (but only sometimes), so maybe it's a RAM thing?
 
Apr 29, 2018
4
0
10


No BSOD, just turning off after some time. Sometimes it reboots, sometimes it turns off. Another thing that happens is that until it truly crashes but after the video driver has crashed, audio continues playing. Here is the screenshot:

fb2YEEC.png


(in a free translation, the error description means: 'Not possible to find Event Identification 14. The component that generates this event isn't installed in the local computer or is damaged' this information is included with it [again, rough translation]: "\Device\Video7
ffffffff(ffffffff) ffffffff ffffffff

the resource message is present, but the message wasn't found on the characters/messages table chain")

The warning is that the driver crashed and the description says something roughly on the lines of: 'the video driver nvlddmkm stopped working and has recovered succesfully'.

There's also been two recent errors:
4KeZk53.png
(Translation: DCOM has run into error '1084' when trying to start the WSearch service with arguments "" to start the server) and
IyxbnvK.png
(Translation: The service Network List Service needs the Identification service from Network Localization/Finding, but did not start due to the following error: It wasn't possible to start the service or dependency group)

This problem has apparently been going on since 06/March, but this has only been happening for a week, two at most.
 

incorrector

Proper
Apr 29, 2018
69
0
160
first of all,i suggest you to totally remove the old display drivers (use DDU) and install the latest ones.
then,if the problem persists, run a memtest until 2+ passes and check for errors
 
Apr 29, 2018
4
0
10


Sorry for the long time before this one. I'll do memtest soon, a bit hard to set it up for now but should be able to until Monday.

Using DDU worked great! It hasn't solved the issue 100% but it has made it much better. The driver still crashes on opening dxdiag and when I tested Slay the Spire, the game crashed (though it took much longer to do so than previously and I'm on a beta build so that might have something to do with it), but it seems to be able to hold much longer before doing so.
 
Apr 29, 2018
4
0
10



Sorry for the long wait, was only able to run memtest today! I think I found the problem. Memtest didn't accuse any errors, but the CPU was stated as 103C, I checked BIOS temp and while system was 35 C, CPU was 96 C
20180528_194318.jpg



Checked with Core Temp too and it reiterated the results.

unknown.png


Is it possible that the system is somehow reading Fahrenheit as Celsius? It doesn't feel that hot when I move my hand over the shell, so it's restricted to the CPU, the fan is working quite well.

edit: Because I have the setting to lower clock rate and voltage automatically in case of a high temperature, so it might be that a faulty temperature sensor indicates it, the CPU lowers its clock to the point where it won't run anything and crashes or something? Is that possible/plausible?