nvlddmkm service stopped, any other ideas?

[Tea-Sir]

Hi there

 

Pesky little bug here. Recently my computer started having a problem with the nvlddmkm driver stopping.

What happens is, my system crashes, just by freezing everything on the screen (it did black screen once), and loop about half a second of audio. I have tried CTRL + ALT + DEL, but that does nothing, leaving it for a few minutes to try and sort it out, doesn't work either. Hitting the reset button does restart the computer.

 

Since no error is showed on the screen, I check the Event Viewer, and find the Event 14 - nvlddmkm, (The only info I get is this:
\Device\Video30cec(3098) 00000000 00000000

 

Then the next event is: Event 4101 - Display, Display driver nvlddmkm stopped responding, and has been restored.

The first time it happened was 2 days after I called Microsoft Support, about the search field that stopped working, I was playing a game of Rainbow Six Siege when it happened.

Again about 24 hours later it happened again, while using PLEX to stream to my Apple TV, though I didn't notice until the next morning, since it had somehow managed to get itself to sleep.

 

During the call to Microsoft, the command to reload my modern apps was run, the one in PowerShell. (And no I don't have the exact command, I am not really PowerShell savvy)

Anyway, I suspected nVidias control panel or driver had been damaged in the process, and downloaded DDU, restarted in safe mode, cleaned my nVIdia driver, restartet the PC like normal, but with my network disconnected, installed the latest driver from nVidias website, and for the past 7 days or so, no problem, until an hour ago, when it happened again, while I was playing Need For Speed Underground 2, of all things.

 

Since DDU didn't work the first time, I don't expect I need to try that again.

But what I am asking is, is there anything I can try, before I go nuclear and just reinstall my system?

My Windows installation is from November 2019, when I upgraded to Ryzen, none of the listed parts is more than a year old:

 

Asus Prime B450-Plus Motherboard

Ryzen 5 3600 CPU, Arctic Freezer 34 eSport Duo cooler

16 GB Ram (2x8GB), 3200 MHz (LPX Vengence)

MSI nVidia GeForce RTX 2070 Armor OC 8GB

256 GB NVMe SSD Black from WD

1 TB Sata SSD Blue from WD

700 watt power supply from Chieftec iArena Series (GPC-700S)

 

Windows is updated to the latest version, 1909.

[Jeff C]

i just started having issues with this myself, did you ever get it figured out?

[Tea-Sir]

Yeah I actually did. Sort of.

Turned out my MSI graphics card and ASUS motherboard didn't want to work together.

 

I installed HWinfo, and noticed that the graphics card never downclocked itself, it was at 1400 MHz constantly. 60c idle temp, 62c to 65c during load. And it used 50 to 55 watts at idle.

 

The really funny thing is, the graphics card does downclock to around 300 MHz in my brothers computer, with the same motherboard and CPU combo. The only difference is, he is running older drivers and older BIOS version.

It also dropped to 25 watt power usage and 45c idle temp.

 

I never got to test if it would work to downgrade my BIOS and drivers, since I already purchased a new graphics card (Gigabyte nVidia RTX 2070 Super Windforce X3 OC), and that works correctly. Idles at 47c, and uses 35 watts at idle.

 

Running the card in Debug mode, did remove the crashes for me. But it has to be enabled every time the PC has restarted.

 

But try to check your card with HWinfo.

[Jeff C]

Thanks for the follow-up, My card and board were playing nice together for about a year (I actually have the same 2070 super) and it just doesnt want to work at all.
I tried hardware info but the "crash" is so frequent now that I cant even run it under load

 

i7-9700k

MSI MPG Pro Carbon

Gigabyte RTX 2070 Super Windforce OC

32 Gig of Gskill Trident-Z RGB, 3200MGhz RAM

Seasonic 850w PSU (80+Gold)

2 512G NVME drives

2 SATA SSD Drives (1 1TB, 1 512G)
GPU and CPU watercooled with EKWB Blocks

2 monitor:  1 Asus 1440p 75Hz TN 27" (used as secondary) and an ACER 1440p 144HZ, 27" Curved Freesync monitor as my primary/gaming display

3 Cable extensions from the PSU to CPU 8pin, the 24 pin, and the GPU
 Current temps under load are in the mid 60s on both CPU and GPU (its 90+ in the house and I havent put in the AC yet)

Nothing is currently over clocked with the exception of the RAM xmp profile enabled.

GPU use to have a 2 overclock settings (the OC curve setting with a +500 MGHZ on the mem, and a manual +40 on the Core and 500 on the mem, currently set to base for troubleshooting)

[Tea-Sir]

Can you get debug mode enabled before the crash?

 

https://www.evga.com/support/faq/FAQdetails.aspx?faqid=59594
 

You could try DDU from safe mode and re install the graphics driver.

That did give some extra time.

[Jeff C]

I'll give that a try, I just bought a 2080s just in case it is hardware, i'll have something to swap with.  Ill keep you updated, Thanks for following up with me btw.
I also attached my most recent error logs

 

[Tea-Sir]

The 3rd screenshot error, I did not have. Could be a nasty driver corruption.

 

My first try would be this:

1. Download DDU from here:Display Driver Uninstaller (DDU) | Wagnardsoft
2. Download the latest nVidia driver from nvidia.com
3. Disconnect internet (pull the ethernet cable or turn off wifi)
4. Restart in safe mode
5. Run DDU, and restart after nVidia uninstall
6. Before connecting to the internet, install the nVidia driver you downloaded earlier
7. When that is installed, connect to the internet, and give your PC a restart, just for good measure.

Let's see what that does.

 

Try HWinfo, to see if your graphics card is running out of spec, and if the crashes continue, try Debug mode. It that doesn't work, the card is most likely dead. Or a recent automatic (or manual, if you can remember doing that), driver update that threw sand in the gears. Until that card has been put into another PC, we just won't know for sure.

 

The second and forth screenshot are of Distributed COM errors. Windows will throw those whenever it is bored, they should be ignored.

 

And you are welcome

[Jeff C]

I'm glad we are on the same page. I just did most of those things. Ran ddu installed an Nvidia driver. I didn't do the latest though just in case that was the one causing me issues. I went from 3 versions back .I think the February/March driver.  I installed just the gpu driver. No physX or audio, no GeForce experience. And set the Nvidia control panel to default. My monitor is currently running at default 60hz I'll slowly up that every couple hours just in case. 

I currently have hwinfo running next to a loop of UNIGINE heaven 

[Jeff C]

This what my gpu looks like under 6hour bench of heaven.  I think everything looks good, and I think the voltage max is supposed to be 1.06

[Tea-Sir]

The fan in the GPU didn't spin up?

Other than that it looks pretty good.

[Jeff C]

Oh my bad. There aren't any fans, I have it on an ek waterblock.  So far it's working but my trust is still shaken

[Tea-Sir]

Ahh, that makes sense then.

It sounds like a corrupt driver to me. But yeah, that kind of instability makes you look at everything every 5 minutes, just make sure it's not going to start again.

[Jeff C]

And that's exactly where I'm at now. I'm still weary of launching a game..but I probably should to make sure.

 

Again thanks for the assist my dude.