Solved BSOD and random crashes

  • Hi there and welcome to PC Help Forum (PCHF), a more effective way to get the Tech Support you need!
    We have Experts in all areas of Tech, including Malware Removal, Crash Fixing and BSOD's , Microsoft Windows, Computer DIY and PC Hardware, Networking, Gaming, Tablets and iPads, General and Specific Software Support and so much more.

    Why not Click Here To Sign Up and start enjoying great FREE Tech Support.

    This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.
Status
Not open for further replies.
scifo76

scifo76

PCHF Member
Sep 7, 2020
10
1
48
Hey guys,

This is my first post here, reaching out to see if someone can help me with my computer problem.

For about a year, I've had random complete shutdown of my computer. Sometimes it's a BSOD, which lists the problem is with the "nvlddmkm.sys" file (NVIDIA driver), but sometimes there is no BSOD, the screen just go black and the only way to reboot is the click the physical power button on the desktop tower to first switch it off, and then back on again.

I managed to read the latest crash dump file in a software called BlueScreenView, and the top 3 sys-files are highlighted in red (indicating a problem?), and those are dxgkrnl.sys, dxgmms1.sys and nvlddmkm.sys. All 3 pointing towards GUI drivers.

The problem is not consistent - it can be weeks (and months) without any issues at all, and then I can have 2-3 crashes in a day.

I have updated to the latest nvidia drivers, still same issue. After reading up a bit, I changed to an earlier version. Same issue. A year ago, I replaced my old hard drive to a new SSD drive and did a complete new Windows 7 installment. Computer worked well for months, and then the problem started to come back.

Windows 7 Home Premium SP1
Inter Core i7 870 @2.93GHz
8 GB Ram
GeForce GTX 960
SSD 860 EVO 1TB

I have tried to find solutions for countless hours on the internet, but it seems I'm just guessing now, hence my post here.
The computer is a pretty old HP desktop from 2011. The only thing I've upgraded in it is the GUI and the SSD.
I don't play any modern games on, so I know my GUI isn't working much at all. Runs like a horse when it's working :)
I could just buy a new computer, but I would like for You guys to at least get an attempt of helping me narrowing down what the real issue is here. I guess the worst case scenario is the PSU or the Motherboard, cause I can't replace them. But if this is a problem with the GUI, I'd be glad to just buy a new one. Software problem? Even better.

So please, let me know how I can help you help me. Do people still use Sandra to post reports or are there other files that I can upload for you to analyse? Crash dump files?

All help will be much appreciated.

Thanks!
 
First glance at your psu,the minimum psu is 500 watts for that 960 and given the age of your pc,it may be time to buy a new psu .. Also that error points to your graphic card..

Please post that bsod collection app log..
 
Thanks for the reply.

I was worried the PSU might be too weak. ****, I need to figure out if I can change it myself :cautious:

I still can't upload the zip-file, it says it's too large. I attach the latest .dmp-file here, does that help?
 

Attachments

The psu is old but that error like i said points to the gpu.. The guys i tagged will help with that..
 
  • Like
Reactions: scifo76
Update: I ran a FurMark test for 20 minutes with no issues. I tried to find the FurMark log file, but I'm not sure which file it, so I took a screenshot of the results shown when the test was finished.

I also would like to mention, that the ONLY game I'm playing on this computer is Everquest Project 1999, which is an emulation of an old game, which has a game engine from 2001, so I'm pretty sure that neither my GUI or PSU are being pushed to the limits here.

I also did the following things today:
I opened the tower and vacuum cleaned everything. I took out my GUI and air-cleaned it as well. I also changed the CR2032 battery next to the motherboard, that has never been changed since I bought this computer in 2011. I also changed the position of the tower, I lifted it from the floor and moved it the wall a bit, so now there is nothing "blocking" it from the sides/bottom. I think it gets less "clogged" this way.

Let me know if there are other things you want me to test here. And again, thanks for the help so far. The computer has been 24 hours without a crash now, so troubleshooting this problem might be difficult.
 

Attachments

  • Furmark-test.jpg
    Furmark-test.jpg
    97 KB Views: 15
Ok you really should not use a vacuum to get the dust out..

The temps on that old gpu is a little high and given it's age it may be problematic,when new the max temps are 98c..
 
Ok, no more vacuuming, thanks for letting me know (y)

About the temperature: When I'm playing Everquest, the temperature on the GPU does not go over 50c (I've measured with MSI Afterburner during many long gaming sessions). When I ran the FurMark test, I heard the fan started to spin a bit, which is never does when I play EverQuest. The FurMark test pushed the GUI significantly harder than when I play EverQuest, and the GUI still passed that test.
With these facts, doesn't this smell like a software/driver issue?
 
Update: I was asked by Nvidia support to run the 3DMark Firestrike benchmark test, and it passed without any issues.
 
Nvidea drivers were the cause of the latest crash.

I noticed that you have no AV protection installed, not a good idea when running an unsupported version of Windows and especially when you have one of the most widely exploited programs like Java installed.

What I would suggest, upgrade to Windows 10 for free while you still can, for this you will need a couple of USB thumbdrives, a small capacity one to save your present drivers to and one that is 16GB or above for downloading a Windows 10 ISO to, let us know what you think.

NB: The PSU is old and of poor quality but when out of the box new it would have been ok for an ASUS GTX 960 which requires the system to be powered by a minimum of a 400W PSU that has at least 29 amps on the +12V rail and a six pin supplemental power dongle.
 
Hello Veeg!

I do not have the possibility to upgrade to Windows 10 for free at the moment, but I'll keep that as a future option if the moment comes when I give up on my current setup.

I haven't had a crash in about three weeks, since I:
  • removed all the dust inside the the computer and dust layers on the internal fans
  • placed the computer away from the wall and floor, so nothing is blocking its sides/bottom
  • replaced the CR2032 CMOS battery next to the motherboard
  • uninstalled 'Microsoft EDGE' and 'Bonjour'
  • changed Windows Media Player Network to Start-Up "Disable" in services.msc
I cannot say that any of these actions actually have solved my problem, because last winter, I had months without any problems and then it came sneaking back up a few months ago. Speculating out loud here now - since I've had more problems during summer than winter, it could be related to heating. But then again, my GUI temperature meters haven't showed any high numbers around these crashes...

If it's possible, can you keep this thread open, so when I have a crash again (which most likely will happen), I can post about it here?

Thanks again for all the help! Great forum!
 
Good to hear.. if you have to open a new thread,please refer to this thread in your new post.

Marked solved:
 
  • Like
Reactions: scifo76
Status
Not open for further replies.
Top Bottom