Windows 10: video tdr failure: nvlddmkm.sys

Dear Windows Ten Forum guys,

I get these random blue screen of death errors and i have tried certain "fixes" i could find when i tried to google this error.
So far nothing worked and im beginning to wonder if i need to replace my pc. I only use this system for my studies at the moment so i consider replacing it as a waste of money. It was working fine up until four months ago.
I have added the log file as instructed. I have also added some system speccs in my profile. Please tell me if i need to add some more speccs.
Can someone please have a look at the log file and/or tell me whats the best thing i can do?

Thanks in advance,
Joseph

:)

 

Video TDR Failure

Hello guys,

While using PC every now and then I get a blue screen error "Video TDR Failure" and what failed "nvlddmkm.sys

Im assuming a video card issue.

Have clean installed windows and hasnt solved issue.

Any advice/tips?

Thanks

 

Video TDR Failure - nvlddmkm.sys

Boot to the Windows 10 desktop

Press Windows key + X

Click Device Manager

Expand Display adapters

Right click the current display adapter

Click Properties

Click Driver tab

Click Rollback driver if the option is available

If not

Right click the current display adapter

Click Uninstall

Exit Device Manager

Restart

Also, depending on the make and model of your computer, you might need to update the driver.

 

Hi j00p,

Welcome to the 10forums *Smile

What kind of fixes are you talking about exactly?

Please update the BIOS from ASUS
P5N-D | Motherboards | ASUS Global


Please uninstall everything of Nvidia using Display Driver Uninstaller and install new drivers from Nvidia. Be sure the clean install box is checked and only install the Graphics driver and the PhysX driver, you can use this tutorial to do so *Arrow NVIDIA Drivers - Avoid Problems - Windows 7 Help Forums
Attachment 82804
Code: 0: kd> lmvm nvlddmkm Browse full module list start end module name fffff801`96800000 fffff801`9749d000 nvlddmkm T (no symbols) Loaded symbol image file: nvlddmkm.sys Image path: \SystemRoot\system32\DRIVERS\nvlddmkm.sys Image name: nvlddmkm.sys Browse all global symbols functions data Timestamp: Fri Jan 29 10:02:12 2016 (56AB2A94) CheckSum: 00C54B7A ImageSize: 00C9D000 Translations: 0000.04b0 0000.04e4 0409.04b0 0409.04e4[/quote]

 

hey there,

woah thanks a lot *Biggrin this looks like something i can work with.
Ill try it tomorrow afternoon as soon as i can. Ill keep you posted.

Greetings Joseph

 

Dear Windows Ten Forum guys,

Thanks again for the help with my driver. I have just followed the above instruction and i am currently using my pc without any problems. So far so good.

Thanks and greetings,

Joseph

 

Dear TenForums guys,

The BSOD problem is back, although they appear to happen less frequently. I have added another log file. Do you think the problem is that the driver updates automatically into one that my system cant hanlde?

greetings Joseph

 

Drivers won't be updated unless you say so.

Have you updated your drivers, they do not look updated.
Code: 2: kd> lmvm nvlddmkm Browse full module list start end module name fffff800`55560000 fffff800`561fd000 nvlddmkm T (no symbols) Loaded symbol image file: nvlddmkm.sys Image path: \SystemRoot\system32\DRIVERS\nvlddmkm.sys Image name: nvlddmkm.sys Browse all global symbols functions data Timestamp: Fri Jan 29 10:02:12 2016 (56AB2A94) CheckSum: 00C50BCB ImageSize: 00C9D000 Translations: 0000.04b0 0000.04e4 0409.04b0 0409.04e4[/quote] I suggest that you try this one, uninstall and install using above instructions.

 

Hi j00p, as I have identical mobo and similar config, I had the same issues - TDR errors and/or frozen system. Because there is absolutely no support for nForce mobos from neither Asus & Nvidia nor Microsoft, I had to make some changes you can try.

1. replacing nForce SATA/RAID controller driver with generic Standard Dual-channel IDE Controller driver in Device Manager
2. disabling all power saving options in Device Manager and in Power Options
3. (the most important change) replacing old Zotac Geforce 550Ti graphics card with new one based on AMD Radeon GPU.

Now is my system definitely rock-stable as it was under Win7, but at the cost of buying new graphics card. I didn't manage to make the combination consisiting of nForce mobo + GeForce + Win10 fully stable (and I tried almost everything), and I doubt it is possible at all. So my suggestion is: try any Radeon (not necessarilly new one) instead of GeForce card.

 

Hi valihrach,

Some of your suggestions are not for TDR problems, TDR problems are video related and things as sata/raid controller driver replacement have nothing to do with video related things.

The third thing is a personal opinion and may not be a solution for the OP, many will say that AMD sucks where many others will say that Nvidia sucks.

 

Some clarification to previous posts:

The only meaning of TDR error is "graphics hardware is not responding and system cannot reset & reconnect to it, so restarts itself". It may have many causes, for example: faulty hardware, bad (not only vga) driver, OS ACPI function call, PCI Express frequency/timing change, PnP reconfig etc. Not necessarily graphics-related cause, only manifestation of deeper system/OS problem (graphics card - mobo - OS interaction). Geforce cards are more prone to these errors than Radeons.

Spent too much time analyzing, googling and trying to solve the same problem on the same hardware, I don't want to cover all details. But some findings and observations:

- nForce-based mobos in conjunction with GeForce cards (systems fully stable under Win7) show TDR errors only under Win10 even if used with the same drivers; these errors manifests itself while the system is lazy, never under heavy load
- stability of my P5N-D system (used for 6 years under Win7 without problems) was improved by disabling all power & sleep options (which modify acpi/pci-e behavior a bit)
- similar effect using generic sata drivers (which impacts pci-e communication too) but never got 100% stable system
- trying very old Radeon HD3650 card - no problems; so I gave up using GeForce card and bought new Radeon - I have rock-solid system again.

2 axe0: Not so personal opinion but sadly personal experience. And borrowing spare old Radeon for testing is relatively easy I think. Sure you may retain your own opinion if you want but in this case imo you are a little bit wrong.

 

I did not focus on your 2nd option, a small typo in my post.

I did not meant the part of replacing an old GPU with a new one, on that part I have nothing to say at all, it was focused on this part

 

hey guys,

The problem persists, even though it appears to occur less frequently. Is switching back to windows 7 an option then? For me it is because i dont really care about windows 10 too much.

 

@valihrach it means the system was unable to recover the drivers, not the hardware, the drivers are the one communicating with the hardware. It has many causes, maybe not always graphics related but all I have seen are 100% graphics related and I have seen a lot
@j00p, that is a decision up to you

 

axe0 says: "it means the system was unable to recover the drivers, not the hardware"
MSDN says: "The GPU scheduler calls DxgkDdiResetFromTimeout when it detects that a hardware time-out occurred."

waste of time