Rtx 4090 crash - driver or hardware? .The description for Event ID 0 from source nvlddmkm

[swrt45]

Hi. I have 12900K stock

32 GB 2x16 GB DDR4 3600mhz

Palit Gamerock Pro OC rtx 4090

1300W Seasonic Platinum Prime

Aorus Elite DDR4 Z690

Windows 11 2H22. Updated new nvidia drivers.

I swapped from 3090 Ti Gigabyte Gaming to Rtx 4090 Palit Gamerock. Never had single crash in games,apps on 3090 ti. All was perfectly stable.
Today i plugged Rtx 4090 Palit Gamerock. It passed 3dmark tests. But after i downloaded Watch Dogs Legion it freezed on loading screen and crashed.

screen1:

https://i.postimg.cc/6Q3F13nx/1.jpg

eventlog screen:

https://i.postimg.cc/8chYrrQM/2.jpg

2 errors.

1.display driver nvlddmkm stopped responding and has successfully recovered


2.The description for Event ID 0 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
Error occurred on GPUID: 100



Happened once i testing now Watch Dogs Legion no crashing again since. But why it crashed? On 3090 Ti i dont had a single crash,nothing.
Rma card now or just keep it ?


I checked 3dmark speed way,port royal,quake 2 rtx no crashing.Drivers are newest. Also my memory is 100% stable because with 3090 ti was fine all the time.

 

[mkay87]

Drivers.

MASSIVE RTX 4090 Problems. driver or hardware?

www.overclock.net

RTX 4090 PC startet neu/friert ein

Hallo zusammen, seit dem Umstieg auf die 4090 habe ich das Problem, das der PC neu startet. Ganz selten friert der Bildschirm ein oder die Bildschirmausgabe ist weg und ich muss ihn dann manuell neu starten. Das Problem tritt nur sporadisch auf und nur beim surfen/Videos schauen/Dateien...

www.hardwareluxx.de

 

[swrt45]

Someone said:"I have to say a few words about this as I also had this issue (black screen and TDR error at idle) with an Asus Strix OC:

I am 100% sure, that this issue is at least partially related to the card hardware or the GPU itself for the following reasons:

I have returned the Asus Strix that I had this issue with and since then also had a Gigabyte Gaming OC as well as the NVidia FE in the very same PC and have not had this issue with the latter two cards at all, the setup being exactly the same otherwise. I had even tried the 12VHPWR adapter, that came with the Gigabyte card, with the Asus card to rule anything but the card out, but of course that didn't help for the Asus card.

I had tried dozens of things like you guys tried here in the thread, I tried with three different PSUs, I tried the card in two completely different PCs (AMD and Intel), with ReBAR on or off, with DOCP on or off. This is in my opinion all a waste of time. If you have this issue in idle, then you have a card that has an hardware issue and you probably better return or RMA it.

The only thing that "worked" for the Asus card was setting the power management mode to maximum performance, like many of you here have also experienced, which leads to the card consuming 60-70W in idle and constantly running at the boost clock rate instead of 210 MHz when idle.
I didn't need to set the Gigabyte card to maximum performance in order for it not to crash, nor do I have to do that for the 4090 FE.

Maybe this issue can be cured by a BIOS update or a driver update, but for sure the root cause is related to the card itself."


So it can be my card is faulty , if root cause is related to the card itself?

 

[kayawish24]

Someone said:"I have to say a few words about this as I also had this issue (black screen and TDR error at idle) with an Asus Strix OC:

I am 100% sure, that this issue is at least partially related to the card hardware or the GPU itself for the following reasons:

I have returned the Asus Strix that I had this issue with and since then also had a Gigabyte Gaming OC as well as the NVidia FE in the very same PC and have not had this issue with the latter two cards at all, the setup being exactly the same otherwise. I had even tried the 12VHPWR adapter, that came with the Gigabyte card, with the Asus card to rule anything but the card out, but of course that didn't help for the Asus card.

I had tried dozens of things like you guys tried here in the thread, I tried with three different PSUs, I tried the card in two completely different PCs (AMD and Intel), with ReBAR on or off, with DOCP on or off. This is in my opinion all a waste of time. If you have this issue in idle, then you have a card that has an hardware issue and you probably better return or RMA it.

The only thing that "worked" for the Asus card was setting the power management mode to maximum performance, like many of you here have also experienced, which leads to the card consuming 60-70W in idle and constantly running at the boost clock rate instead of 210 MHz when idle.
I didn't need to set the Gigabyte card to maximum performance in order for it not to crash, nor do I have to do that for the 4090 FE.

Maybe this issue can be cured by a BIOS update or a driver update, but for sure the root cause is related to the card itself."


So it can be my card is faulty , if root cause is related to the card itself?

i also have asus strix 4090 OC edition and i was also getting black screen and rebooting prroblems until i enabled prefer performance in nvidia control panal which is temp fix as you said.SO, i think you did a mistake by returning the asus card back because its not Asus card problem it is Asus card Vbios and Nvidia drivers both.Asus already released Vbios V2 for the card and Nvidia said that this week new drivers will comeout which will fix the problem.The problem with asus card is when in idle the mV volt go very low which in result gives nvlddmkm crashes.If Asus cards had hardware problem then it should also crash and reboots while playing game but it does not.it only crrashes in idle mode right ? so card is not a problem here as what i can say.Asus strrix 4090 OC is such a prremijm card and its almost impossible to find it on Germany online shops and i was very lucky to get it.

 

[Azrael_ct]

ASUS released a few beta bioses for Z690/Z790 Boards specifically targeting the problems with the RTX 4090

ROG STRIX Z790-F GAMING WIFI BIOS 0703
"1. Improve system performance
2. Fix the driver compatibility issue with NVIDIA RTX40 series cards
3. We suggest updating the ME firmware to the latest version 16.1.25.2020

With my Z690 Hero i had the problems too, but the new 2201 beta bios fixed it. I couldn't even get to the windows desktop before.

RaptorLake Resources

i will use this as a collection of test bioses/tools/info targeted towards Raptorlake + z69/790 NOTE for Z690: You must update ME on your Z690 to properly support raptorlake (if you are on dual bioses then you need to do this to both bioses) 1) d/l and install ME driver...

rog.asus.com

Don't forget the Intel ME update before.

 

[owned139]

ASUS released a few beta bioses for Z690/Z790 Boards specifically targeting the problems with the RTX 4090


With my Z690 Hero i had the problems too, but the new 2201 beta bios fixed it. I couldn't even get to the windows desktop before.

I have the exact same Motherboard and no issues or crashes whatsoever and i dont use the latest BIOS.

 

[Azrael_ct]

I don't know why the problems are happening for some users. Maybe it's a combination of Motherboard BIOS, other Hardware, GPU vBIOS and driver issues. I had the worst kind of problems - after login to windows the crash occured after a few seconds everytime. Not a single exception. After some research the only solution that worked for me was booting the system in safe mode, and setting the driver to high performance mode via registry. The drawback: 200W power usage (PSU) when idle
Yesterday, i did only the BIOS-Update and reverted the change to the registry - everything is fine since then. In my opinion the majority of the black screen issues can be solved by applying vendor updates, so there is no reason for a RMA

 

[swrt45]

I am gettin occassional crashing during loading games. Watch Dogs Legion ( happened 2 times ) and Metro Exodus EE. Never in game, but in loading . Happen from time to time. Is my card fine? I have PalitGamerock 4090 stock.

Here are always that 2 errors during crash:
2 errors.

1.display driver nvlddmkm stopped responding and has successfully recovered


2.The description for Event ID 0 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
Error occurred on GPUID: 100


It happened on 3 versions of drivers i tried.

 

[FloppyDuck1]

I have the same problem with my Zotac RTX 4090 Trinity.

My specs are:
I7 13700KF, Asus rog strix z690 Gaming F, 32GB DDR5 5600Mhz, Kingston M.2 SSD and WD HDD, EVGA Supernova PSU 1000W. Nothing Overcloced or physically damaged, everything should ne fine. Even the 12vhpwr adapter is fine, no burn or what so ever.

The first month i never had a single problem, played games like 8 hours a day, never was a problem. Last week i got the first crash while gaming CoD, the video output freezed but the sound was continuously playing till i saw a bsod (dpc watchdog violation) and my pc restarted. The crashes once occurred while gaming, but the last few days it came even at idle when watching videos on YT and even when writing the password at the windows login.

So i did research, read all of the tips i have found, clean instal windows, old/new drivers with DDU, updating all drivers, CMOS, Bios beta update, checking the pc for bad connection, setting the desktop display performing by GPU, deconnected every other hardware, but nothing helped.

Only the setting to prefer maximum power in the nvidia settings was a workaround (a little one, because here and there I still had a crash when running 3DMark, but YT was fine with maximum performance, i think) AND when i deinstalled the gpu driver. At this point i was able to watch videos without a single crash but yeah, i cant game. It is worth to mention that i had black dot artefacts when the freezing occurred, so it must be the VRAM or a corrupt power controller/VRAM controller at a hardware level on the GPU and not at software, because none of the released drivers worked for me more and when no gpu driver was installed, the pc don't know what to do with the gpu hardware, so the display output runs at software level. At the point when the pc knows what to do with the gpu hardware, it fails.

My next step is to RMA and so you have to do that, because it seems to me like you have a faulty gpu and that you will get more and more of such crashes. (Für die Deutschen hier, Gewährleistung beantragen!)

 

[swrt45]

card passing all stress testing games and benchs.

 

[FloppyDuck1]

card passing all stress testing games and benchs.

Ye, but i had the same problem in the protocol and sometimes when idle, as you, as many in other forums/threads mentioned. It is 100% linked to the same problem when the problem is mostly caused when it is not at "prefer maximum power", as it is a workaround for my crashes. It can be a micro cut or electrical damage in the gpu hardware and when setting to maximum power, the output gets through there where it should go on the board. In my opinion ether you will get the same problems as I have in the future or u will have a slightly faulty gpu and none of the drivers will help you out. Its not of a big deal to let your gpu check, at least here in germany.

I would suggest you to test your game that caused the problem and setting the card at maximum power for one day straight or more. As i said it was for me a little workaround, the problem still occurred at sime times, but not that often. To figure out if this is a micro damage.

 

[swrt45]

Should i rma card?

Like i said.I downloaded 3 games: Metro Exodus Enhanced,Watch Dogs Legion,Portal Rtx.
And on all of these 3 games i had crash on loading screen in my first launch ( after download ). Not happened again since. Now i can launching game forever and it will never crash. So why it crashed on first time? Thx

 

[waynesmith3615]

I have the same problem with my Zotac RTX 4090 Trinity.

My specs are:
I7 13700KF, Asus rog strix z690 Gaming F, 32GB DDR5 5600Mhz, Kingston M.2 SSD and WD HDD, EVGA Supernova PSU 1000W. Nothing overlocked or physically damaged, everything should be fine. Even the 12vhpwr adapter is fine, no burn or what ever.

The first month i never had a single problem, played games like 8 hours a day, never was a problem. Last week i got the first crash while gaming CoD, the video output was freezed but the sound was continuously playing till i saw a bsod (dpc watchdog violation) and my pc restarted. The crashes once occurred while gaming, but the last few days it came even at idle when watching videos on YT and even when writing the password at the windows login.

So i did research, read all of the tips i have found, clean install windows, old/new drivers with DDU, updating all drivers, CMOS, Bios beta update, checking the pc for bad connection, setting the desktop display performing by GPU, disconnected every other hardware, but nothing helped.

Only the setting to prefer maximum power in the nvidia settings was a workaround (a little one, because here and there I still had a crash when running 3DMark, but YT was fine with maximum performance, I think) AND when I uninstalled the gpu driver . At this point i was able to watch videos without a single crash but yeah, i cant game. It is worth mentioning that I had black dot artifacts when the freezing occurred, so it must be the VRAM or a corrupt power controller/VRAM controller at a hardware level on the GPU and not at software, because none of the drivers released worked for me more and when no gpu driver was installed, the pc don't know what to do with the gpu hardware, so the display output runs at software level. At the point when the pc knows what to do with the gpu hardware, it fails.

My next step is to RMA and so you have to do that, because it seems to me like you have a faulty gpu and that you will get more and more of such crashes. (For the Germans here, apply for a guarantee!)

Hello from the UK.

I have exactly this problem with the same GPU. Did you RMA this, and what was the outcome?

I sent mine in 2 weeks ago and they said No Fault Found and returned the card back to me. Been using and old GTX970 in the interim with no problems, hooked up the 4090 again last night, and the same problems started again.

I don't know what to do now, it's very frustrating.

thanks in advance

 

[IronAge]

@waynesmith3615

tried other than displayport output signal ? tried HDMI ? tried negative offset for GPU Cores and VRAM ?

 

[warmachine79]

Same issue here, what helped for me is limping into MSI Afterburner if it does not crash before and then VRAM -500. Not a crash since then. However, I will RMA the card now as that is not a good solution. Also had the black dots.

 

[OCBBlackEdition]

Gibt anscheinend nur ganz wenige NT die eine RTX4090 richtig ansteuern könn

 

[warmachine79]

Gibt anscheinend nur ganz wenige NT die eine RTX4090 richtig ansteuern könn

Was kann man denn dabei falsch machen und gibt es breiter diskutierte Anhaltspunkte, dass es da Probleme gibt?