Question [Hang/Freeze/Crash] - Event ID 14 nvlddmkm, AMD+NVIDIA

Page 6 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.

Aravind92

Honorable
Apr 1, 2014
620
5
11,015
21
I'm also having this problem, it's been very frustrating. I actually RMA'ed my GPUs thinking they were the culprit, not a week after the replacements arrived the problem began again. It's infuriating as I will be in the middle of work and the PC will suddenly lock up.

I've noticed the problem for me tends to be worse if I work a full day in UE4, Substance Painter or any other GPU intensive program and then leave my PC on overnight. That seems to lessen the occurrences a little bit. A little bit being key there. I usually experience these crashes every other day, sometimes daily, but after a lot of tweaking the crashes happen much less often now, but they still do happen.
Another thing I've found that helped a little bit was to up the voltage a little bit to the SoC along with frequently restarting the GPU driver (Win+Ctrl+Shift+B). I'm not sure which made the most difference, but if I was a betting man I'd say it's the frequent restarting of the driver. But if someone else wants to try both of these out and see if they experience less crashing maybe can find out.

I've run through all the basic troubleshooting steps, like testing the cards in another PC, testing this system with an older AMD GPU, and all the hardware seems to be OK. Just found this thread, so at least I now know I'm not alone in this.

Now, if I'm not mistaken, some motherboards with multiple PCIe slots will have some of those slots using chipset lanes instead of CPU lanes, has anyone tested using a slot that passes through the chipset vs directly into the CPU?
I've been thinking of trying a single card in every PCIe slot and see if that makes a difference.

Specs:
CPU: AMD Threadripper 3970X
Motherboard: Asus Zenith II Extreme
RAM: 256 GB Trident Z Neo
GPU(s): 2x Nvidia Titan RTX
Motherboard bios: Latest
Drivers: up to date

Device manager event:

Event ID 14/nvlddmkm

\Device\00000192
0d02(31c8) 00000000 00000000
Hi,

Could you please watch the below video and tell me if that is the same thing you are experiencing as well.

View: https://www.youtube.com/watch?v=zszRpGy-dVI


If yes, please raise this issue with both nvidia and AMD.

Also, does setting the power management in nvcp to high perfromance help, that seems to be the only work around at this time and seems to be working for everyone who's tried it, I personally haven't done it due to the extra heat and power consumption, not ot say it is justa work around and things should work stock. Looking at your work case, you maybe ok with setting it to high performance.

Just letting you know.
 
Reactions: DunkelBlauer
May 14, 2020
2
1
15
0
I've been monitoring since my last post, and haven't had any cause to comment until now...

I can confirm my system is now long term stable again, and its with the following;

Windows: Win10 - 2004 (latest updates)
Driver: 446.14 (installed after DDU safemode uninstall of 436 driver) (no geforce experience)
Bios - No Change still running F12e from 2‎020/03/06
Motherboard: Gigabyte Aorus Pro
CPU: Ryzen 9 3900X
All Settings are default.

I will comment I have since switched from Firefox back to Chrome, due to other issues with Firefox that could not be resolved. But I had this prior to switching to Firefox, as such I never considered it a confirmed cause, but has not recurred at all since making the last changes to my system which were the 100% clean driver install of the 446.14 driver.

Beyond that, I'm sitting at 14 days of up time and my system just laughs and carries on, I'm sticking with the exact config I have now, and won't be changing at all...

No upcoming games are of interest to me so no real driver updates needed and my current games run above 180 FPS... and seen as high as 600 FPS, I would say its a flex, but its Minecraft... so yeah... not really... :ROFLMAO:

Anyway, I hope this helps anyone who may want to try replicating the software setup on their machines.
 
May 13, 2020
9
0
10
0
Hi Guys,

I've contacted both Nvidia and Amd's support on this matter. AMD responded saying it is sent to their engineering team for them to research.

With Nvidia, I got the case escalated to 2nd level, the 1st level agent hardly understood the issue and honestly wasn't very helpful, just giving me the generic troubleshooting steps. But the 2nd level agent has been working and researching on the matter.

Could you guys please contact them so that both the companies are aware that it is a widespread issue and swapping hardware components doesn't help. With nvidia please have it escalated.
I've reached out to nVidia about this and to ASUS to see if they can actually release a firmware with AGESA 1.0.0.5 or 1.0.0.6 to see if that fixes the issue. I guess it's time to reach out to AMD as well though I think they'll tell me to get a newer AGESA version.....
 

Aravind92

Honorable
Apr 1, 2014
620
5
11,015
21
I've reached out to nVidia about this and to ASUS to see if they can actually release a firmware with AGESA 1.0.0.5 or 1.0.0.6 to see if that fixes the issue. I guess it's time to reach out to AMD as well though I think they'll tell me to get a newer AGESA version.....
Oh yeah, that's unfortunate, but still reach out to them, refer this thread and perhaps my video and mention that there are people who've updated their AGESA versions and still face the issue.

Couldn't hurt to try and also, reach out to Nvidia and have the matter escalated, keep responding to them and work with them to get to the bottom of this, I think if they can get our help in narrowing it down, we will get somewhere.

The new agent I am talking to their has said he is researching further and is looking to past cases which he thinks this might be tied to, he has given multiple suggestions, like turning off fast start up, keeping of switching the BIOS mode to CSM.
 
Jun 14, 2020
7
1
10
0
I just got the error on my new RMA'd GPU, sad times

But, I just saw there was a BIOS update for my board (TUF-B450-PLUS-GAMING)

Installed it just now, fingers crossed!
 

fluidz

Commendable
Mar 8, 2018
8
0
1,510
0
Question for you all, what browsers are you using?

I don't know if its an unrelated issue but since turning the power settings back to optimal I have am having performance issues in Firefox but not chrome. Googling "Firefox nvidia" seems to show people having driver crashes only when running Firefox

EDIT: I think I have some actual reproducible problems now

I use Blue Iris NVR software, and using the WebUI sometimes the video will jerk down to 5fps and then back up to 35fps (Well above double the FPS of the feed!) every second causing horrible playback

When I look into X1 to see the clocks, its going from 300Mhz all the way to 1650MHz every second which is directly inline with the jerking

If I load up the Blue Iris WebUI in chrome, it sits at around 400-500MHz constantly and has no issues with playback

With the power settings turned to max in the NVIDIA control panel, the issue goes away entirely
When I scroll firefox it makes my 2080ti/gt710 boost to low level 3d clocks, which is expected, but it seems to boost really high, even on the about:config page, and Firefox stutters each time the gpu boosts. This isn't an issue with my onboard graphics card. Seems to be an Nvidia thing. Please take a look at this thread, and see if you can reproduce the stuttering in firefox. https://www.nvidia.com/en-us/geforce/forums/geforce-graphics-cards/5/389515/nvidia-power-states-stuttering-in-firefox/
 
Jul 20, 2020
1
0
10
0
Hi all,

ID 14 keeps happening while processing data with Agisoft Metascan, a software used to generate 3D models and orthophotos from camera images. Some processing stages (e.g. "dense cloud generation") are very GPU intensive as there are several hundred million up to two-figure billion points to be processed.

Agisoft Metashape itself throws a CUDA-Error,
Code:
cudaMemGetInfo(&free_mem_size, &total_mem_size): the launch timed out and was terminated (6) at line 211
Some people have similar problems, which the developer's forum links to driver issues see post here.

The mysterious thing is: after one project triggers the error it will consistently throw errors for the same project, even after the forced workstation restart. However if one switches to another project and works there for a brief moment it is always possible to then follow up with a successful re-run of the troublesome project.

We will be switching to GNU/Linux soon (mostly for Win 10 secretely not supporting more than 32 cores), but some software only runs under win 10, needs GPU support and virtualizes badly. If there is any way I can contribute/complain somewhere - just let me know!

Cheers
Dave

specs:
Minecraft 1080p/low/40 fps PC (/s)
CPUAMD EPYC 7702P - 2.0 GHz - 64 Core, 256MB Cache, 200W, So. SP3
RAM512 GB LRDIMM Samsung DDR4-2666, CL19, reg. ECC
MBSupermicro H11SSL-i, AMD So. SP3 ATX Rev. 2
PSU1600W Corsair AX1600i Professional Series, 80Plus Titanium
GPU2 x NVIDIA Titan RTX, 24GB - none SLI
Nvidia Driver441.66 DCH
OSWin 10 Pro - 1909
 
Jul 20, 2020
1
0
10
0
Hey. I have the same problem with error 14 from once a week to once every 2 months. Configuration:
Ryzen 5 3600 processor
MB Asrock b450m Pro4
RAM Crucial 32GB 2666Mhz (2x16)
VGA Gigabite 1660Super gaming

A week ago I installed BIOS with AGESA 1.0.0.6, it did not help, yesterday this error was repeated.
Set maximum performance again in the graphics card drivers.
 
Last edited:
Jul 20, 2020
1
0
10
0
So this problem has just started for me, the first time was 2 weeks ago, happen once then suddenly it has started happening around twice every few days. but my screens just blinks --> monitor goes black and then back again. and I have to refresh my youtube/ streaming tabs in firefox. The error I get in event viewer

\Device\Video3
0d02(31c4) 00000000 00000000

This pc build is a month old
my specs are
CPU: AMD Ryzen 7 3700X
GPU: ASUS GeForce RTX 2080 SUPER ROG Strix OC
RAM: Corsair Vengeance RGB PRO DDR4 3000MHz 16GB
 
Jul 20, 2020
1
1
15
0
Same issue as everyone is mentioning. New build. Black screen for 2 seconds and then it will either continue as normal, or begin artifacting when I drag windows around or try to access the start menu/task bar. Returns to normal after clicking "Restart" on Windows Explorer in Processes. Until the next time it happens. I've tried the "Prefer Maximum Performance" method as well as turning PCIe off in power settings. Sad face...

Ryzen 9 3900x
Asrock B550 Phantom Gaming ITX
Nvidia RTX 2080 FE
g.Skill Trident Z Neo 32GB
Windows v2004


The description for Event ID 14 from source nvlddmkm cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.

If the event originated on another computer, the display information had to be saved with the event.

The following information was included with the event:

\Device\Video3
0d02(31c8) 00000000 00000000

The message resource is present but the message was not found in the message table
 
Reactions: fluidz
Jul 20, 2020
1
1
15
0
Same here with my build ...
There are no problems while playing.
As soon as Windows is in IDLE mode the Event 14 error messages described here will appear.
Just like the user "Aravind92" showed it in the video above...

I listened to the suggestions here in the forum and set my Nvidia Power Settings to "Performance", since then (and that's almost 2.5 weeks) I haven't had a single freeze!

Ryzen 7 3700x
X570 AORUS ELITE - AGESA ComboAM4 1.0.0.2
MSI - NVIDIA GeForce RTX 2070 SUPER
g.Skill Trident Z Neo 32GB
Windows v2004
 
Reactions: fluidz
Jul 20, 2020
1
0
10
0
Just wanted to say thanks for the advice on this thread!

I was getting this error every time I enabled HDR, the PC would mess up anywhere from 10 seconds to an hour after doing so, sometimes just flat out rebooting itself. It also occasionally did it while not in HDR mode, usually when watching videos in the browser.
Setting the GPU power management to max performance in the Nvidia control panel seems to have fixed this entirely, this makes me very happy!

Thanks once again!
 
Jul 16, 2020
3
0
10
0
Hi,

Could you please watch the below video and tell me if that is the same thing you are experiencing as well.

View: https://www.youtube.com/watch?v=zszRpGy-dVI


If yes, please raise this issue with both nvidia and AMD.

Also, does setting the power management in nvcp to high perfromance help, that seems to be the only work around at this time and seems to be working for everyone who's tried it, I personally haven't done it due to the extra heat and power consumption, not ot say it is justa work around and things should work stock. Looking at your work case, you maybe ok with setting it to high performance.

Just letting you know.
Yep, its the exact same error. I've already raised this issue with Nvidia and AMD many months back when I first got it. At the time both, Nvidia told me it was an AMD problem (after RMAing my cards a few times), and AMD told me it was an Nvidia problem.
Changing power management doesn't fix it for me. What does help significantly for me is frequently restarting the Nvidia GPU driver using Win+Ctrl+Shift+B. Of course this is extremely inconvenient, as every hour or two I have to shut down Max or Unreal Engine and restart the damn driver, but it has seemed to curb the crashes -- for now.
 
Jul 28, 2020
2
0
10
0
I have the same problem on my new build. I get the crash about once per week.
It only happens when a game is running, it never happened while idle or browsing the web.
The mouse starts to lag, the screen starts freezing, the game will often crash then the screen goes black.
It can recover automatically but most of the time I just restart the computer because it takes a while.

The Event Viewer reports the event 14 from nvlddmkm, then a bunch of related errors.



  • Gigabyte X570 AORUS ELITE
  • AMD Ryzen 7 3700X
  • G.Skill RipJaws 5 32 Go (2 x 16 Go) DDR4 3600 MHz CL16 (XMP enabled)
  • ASUS GeForce RTX 2070 SUPER ROG-STRIX-RTX2070S-O8G-GAMING
I tried to change the Power Management mode "Maximum Performance" today, I will report back by editing this message if it worked.
 

Aravind92

Honorable
Apr 1, 2014
620
5
11,015
21
I have the same problem on my new build. I get the crash about once per week.
It only happens when a game is running, it never happened while idle or browsing the web.
The mouse starts to lag, the screen starts freezing, the game will often crash then the screen goes black.
It can recover automatically but most of the time I just restart the computer because it takes a while.

The Event Viewer reports the event 14 from nvlddmkm, then a bunch of related errors.



  • Gigabyte X570 AORUS ELITE
  • AMD Ryzen 7 3700X
  • G.Skill RipJaws 5 32 Go (2 x 16 Go) DDR4 3600 MHz CL16 (XMP enabled)
  • ASUS GeForce RTX 2070 SUPER ROG-STRIX-RTX2070S-O8G-GAMING
I tried to change the Power Management mode "Maximum Performance" today, I will report back by editing this message if it worked.
That is new, can you tell me what kind of games you play and how much of a load that puts on the gpu, It has happened once for me in a game but it was again when the clocks were low cause I had just finished a round going back to menu screen, basically loading screen and that brought the clocks down to idle. And seeing as you are also seeing opengl errors, that must be because you were in a game, cna you check with nvidia to see if this a different issue.
 
Reactions: DunkelBlauer
May 13, 2020
9
0
10
0
Happened again while playing a Youtube video in Firefox. My screen flashed and Event Viewer showed Event 14. Everything seemed to recover and I continued using my PC until just about 30 minutes ago. I had another video playing and opened the new Windows Terminal app to SSH to a server. I noticed that input to the terminal app was extremely laggy and that's when my display completely froze. No keyboard or mouse input would work (numlock key was on and couldn't be turned off; pressing capslock did nothing) but the funny thing is that the audio from the video I was playing was going along just fine. Rebooted the PC and things are back to normal. I opened an RMA ticket with AMD just now to see if a new processor fixes the issue. I'm running AGESA 1.0.0.6 on my ASUS TUF B450M-PLUS GAMING motherboard and an Zotac RTX 2070.
 
Jul 30, 2020
4
0
10
0
I got this error about two weeks ago. Restarted my PC and was fine until a few days ago and it happened again on another game. I changed my power setting to max performance in the Nvidia control panel and haven't had any issues in the last few days.

Only problem for me is that the card never underclocks when it's idle anymore. Is there a fix or workaround for this?

edit: I changed my global settings back to adaptive and then just went into the individual game settings in the nvidia control panel to max performance for every game I have installed and hopefully what this achieves is low idle clocks on desktop (where I've never crashed) and then max performance in games (where I have crashed(
 
Last edited:
Jun 20, 2020
2
0
10
0
I have the same problem on my new build. I get the crash about once per week.
It only happens when a game is running, it never happened while idle or browsing the web.
The mouse starts to lag, the screen starts freezing, the game will often crash then the screen goes black.
It can recover automatically but most of the time I just restart the computer because it takes a while.

The Event Viewer reports the event 14 from nvlddmkm, then a bunch of related errors.



  • Gigabyte X570 AORUS ELITE
  • AMD Ryzen 7 3700X
  • G.Skill RipJaws 5 32 Go (2 x 16 Go) DDR4 3600 MHz CL16 (XMP enabled)
  • ASUS GeForce RTX 2070 SUPER ROG-STRIX-RTX2070S-O8G-GAMING
I tried to change the Power Management mode "Maximum Performance" today, I will report back by editing this message if it worked.
I have almost the same build as you except RAM/GPU.
I had no crash since July 14, before that it was 2-3 times per week.
Running F20 BIOS, AMD chipset drivers 2.07.14.327 and NVIDIA driver 451.67.
 
May 29, 2020
2
2
15
0
Event 14 is a catchall for many kinds of issues - could be a bad GPU, Incompatible RAM, etc. etc.. The particular problem which initially was the "spirit" of this thread was for the particular issue of the PCI Express speed change causing the crash and for my GB X570 Pro it is indeed fixed in the F20 BIOS or you just have to change in Nvidia control panel power management to prefer maximum performance from adaptive. If this is your issue I can guarantee that either of those things fixes it as I have now gone more than long enough without a single crash first with the control panel change and then with that back set to adaptive and the BIOS change.....so again if it is not fixed for you it means you have an another issue like RAM incompatibility, etc. It did take them 1 year to fix this issue which is pathetic.
 
Jul 8, 2020
2
0
10
0
I had this happening to me 2 weeks after I upgraded from a 1060 to the rtx 2060. When it happened, the screen froze and eventually I got a black screen with a mouse cursor (happened twice and both times while in game). Maybe the PC'd recover eventually, but I just rebooted manually because it took too long. I updated my driver from 446.14 to 451.48 and everything seems fine... until another 2 weeks later when I started getting event ID 14 messages in the event viewer. It occurred twice while in desktop, once when the display was turned off during idle, another while just browsing with a youtube video playing in the background. I also get them while in game twice. The difference here was that since I updated to 451.48, I no longer get the stutter/freezing event ID 14, I instead get the black screen flash (like the monitor got disconnected) and got dumped to windows desktop (if I were gaming). I guess that was better than having the PC hardlocked? I I don't know if this all has something to do with the PCIE speed change problem that's described here, but while in game it seems to happen when the clock speed is hovering below 1000mhz.

I notice that with driver 446.14, with the freezing/stuttering, I got this message on the event ID 14:
"\Device\Video3
0cec(3188) 00000000 00000000"
With Driver 451.48 however, where the display driver (the black screen flash) crashes and recovered immediately, I got this message on the event ID 14:
"\Device\Video3
0d02(31c8) 00000000 00000000"

My system spec is B450m Steel Legend running on BIOS P2.10 (very old, AGESA 1.0.0.2, this is what came with the board when I got it; I have an irrational fear of bricking the mobo while updating the BIOS)
Ryzen 5 3600
Colorful RTX 2060
16gb of DDR4 3200mhz Klevv Bolt X
Cooler Master Master watt lite 600w

PCIE link power management is off, and HW acceleration is disabled on my browsers. I haven't tried running with maximum power setting in the nvidia control panel. The card I'm using is the cheapest rtx 2060 I could find, and as such, runs fairly hot on max clockspeed due to the fairly basic cooling solution.

Honestly I kinda want to just RMA the GPU to eliminate it being the problem, but when I inquired the shop I bought it from, they asked me to bring my whole PC for testing. Given how sporadic the issue is and how difficult it is to replicate, I doubt they'll replace it.
 
Last edited:

jinchuriki

Prominent
Sep 9, 2018
116
2
585
0
I've upgraded some parts of my PC around 8 months ago, even since I'm experiencing this issue. I just upgraded my CPU, mobo, RAM, PSU. (2700x, phantom gaming 4 x570, 2x16GB RAM 3200mhz)

Still using my GTX 770 4GB, and it's happening to me as well. Not all that often..But it does and it's super annoying.

I thought of upgrading to another GPU, but I don't know if that will even solve the issue for me.
 
Jul 30, 2020
2
0
10
0
Same issue. When playing stream or video long time with Chrome or Potplayer, the screen is suddenly stuck, I need reset my computer.

CPU: AMD Ryzen 3700X
Motherboard: ASUS TUF GAMING B450M-PRO S (BIOS version 2202, agesa 1.0.0.6)
GPU: ASUS 1650 Super
Memory: ADATA DDR4 2666Mhz 16G*2

Windows 10 version 2004 (19041.388)
Driver tested with 442.74 and 451.67

Event ID 14
\Device\Video3
0cec(3098) 00000000 00000000
 
Aug 4, 2020
1
0
10
0
I've been experiencing similar black screen crashes shortly after this PC was built about 9 months ago.

As a last resort and before finding this thread, I reinstalled my Windows about a week ago, but after experiencing another black screen crash a few days ago, I decided to register here and share my pain.

Specs
CPU: AMD Ryzen 3700X
Motherboard: X570 UD (Ultra Durable, BIOS F11)
GPU: MSI RTX 2070 Armor 8G
Memory: G. Skill Ripjaws V, 16GB (2x 8GB), DDR4 3200 MHz

Windows 10 Home, version 2004.
NVIDIA Driver Version: 451.67.

Event ID 14
\Device\Video3
0d02(31c4) 00000000 00000000

Description
The crashes seem to occur randomly, at least once a week or once a month. Doesn't matter if I'm in a game or just using Firefox. Sometimes the display driver will recover and I can continue using the computer normally. The types of crashes have varied slightly:

The screen will go black, all sound will cut out and my monitor will lose signal. The computer stays on, the lights on my mouse and keyboard remain on. While waiting for several minutes, my monitor will occasionally regain the signal for a few seconds, but the display remains black and the signal is lost again. This process seemingly repeats forever.

Black screen, intense stuttering. Mouse cursor is still visible, but stutters a lot.

Black screen, fans go to a 100%. Only way to restart is by holding down the power button. (happened only once so far).

To me this definitely seems like a driver issue. Before reinstalling my Windows, I've uninstalled and reinstalled my drivers several times, usually trying out the latest ones to see if the issue would resolve itself, but that has not fixed my issue.

The F20 update for a lot of X570 boards seems to have fixed the issue for some people. I'll probably try that next and report back if it helped.

https://forums.tomshardware.com/threads/hang-freeze-crash-event-id-14-nvlddmkm-amd-nvidia.3594431/post-21751815

I didn't see much discussion around user aeternus-derentis post.
I saw that he had a follow-up post made almost a month after this one where he says his system has been completely stable since reverting his drivers back.

https://forums.tomshardware.com/threads/hang-freeze-crash-event-id-14-nvlddmkm-amd-nvidia.3594431/post-21868912
 
Last edited:
Aug 4, 2020
3
0
10
0
Ok well, sometimes I can go a day or two with nothing happening, and sometimes it will crash two or three times in a day. Always with "The description for Event ID 14 from source nvlddmkm cannot be found." in the event log. I will usually have a few seconds of performance stuttering before the display shuts off. I think the black screen may be because I'm using a non-standard monitor setup (55 inch LG OLED on both computers). The computer is still working fine, just the display goes black, and I can occasionally recover to desktop with the ctrl-win-shift-B to save any work in progress but I'll need to reboot before using any 3d graphics again or the driver will immediately crash a second time.

I have been emailing back and forth with Nvidia tech support over the last week... they have all the logs and of course all of my drivers and bios are fully updated. The last gem I got from them was to reseat my video card lol (I did anyway just for kicks but I'm fairly certain that's not the issue here...) I'll update the thread if we find a solution but I've tried the other tweaks suggested here (power settings etc) and nothing seems to have helped yet.

Sounds really similar to whats happening to mine, did you ever find a fix or whats going wrong in general?

For me the tower is still working, and any audio either from music or streaming still plays on my speakers, the black screen happens mainly when I'm using Firefox or Chrome, mostly when watching videos but sometimes when browsing normally. Has also happened once using Discord, I've turned off hardware acceleration for everything and changed the power settings but still no luck.

I can still use GPU intensive programs such as Premiere, Maya, Substance Painter and Unreal without the error happening so.
 

ASK THE COMMUNITY

TRENDING THREADS