Getting a TDR timeout with the nvlddmkm driver whilst gaming

Alza77

Reputable
Jun 18, 2014
13
0
4,520
Pretty much every time I load up certain games and play them for around 30 minutes to an hour my computer will get this error. The screen will turn black but I can still hear some audio and I will see "Display driver nvlddmkm stopped responding and has successfully recovered." This error only appears in GTA V, The Forest, Rust, Dead Island Riptide, Middle-Earth: Shadow of Mordor and Dying Light. The games will run flawlessly on high settings at 60fps and then the error will just randomly happen.
I have had this issue for just under a year now which was when I built this rig myself but it only seems to arise when I am playing games.

Here are my specs:
Motherboard: Gigabyte GA-Z87-OC
CPU: Intel Core I7 4770K 3.5GHz
GPU: Gigabyte Geforce GTX 780
PSU: XFX ProSeries 850W XXX Semi-Modular
RAM: Corsair Vengeance 4x4GB DDR3
OS: Windows 7 Pro 64-bit

Things I have tried to troubleshoot the issue:

- Uninstalled/Installed the latest nVidia drivers countless times using different programs such as DDU, Driver Sweeper, RevoUninstaller and CCleaner. I even spoke directly to a few people on the nVidia live chat but to no avail.

- I have ran Memtest86+ on all of my RAM sticks both together and individually on about 4-5 passes each time and no errors were detected on them.

- I have stress-tested my GPU using Furmark and 3DMark and have seen temperatures max out at about 72C So I assume the temperature is not an issue.

- I have tried my GPU in my brother's rig and have not been able to replicate the error.

- I have flashed my motherboard BIOS from F3 to F8

- I have formatted my SSD and reinstalled the OS but I am still having the error when playing games.

- I have checked my dump files after my last error occurred using WhoCrashed and it has definitely outlined a video TDR timeout.

Before this XFX PSU I had a Corsair GS800W which I RMA'd and it turned out faulty, however the error is still present. I have RMA'd my GPU as well but it was found to be fine. I am thinking about RMAing my motherboard as well to help root out any issues with it. Any further suggestions and help will be appreciated.
 

maxalge

Champion
Ambassador


Install msi afterburner, and set power limit to +25. Disable ULPS.

Set windows power options to high performance.


Run a game. Monitor temps make sure you turn on cpu temp monitoring in OSD.

 

Alza77

Reputable
Jun 18, 2014
13
0
4,520
I've installed MSI Afterburner but am having trouble with this power limit setting you mentioned, I'm also not sure what ULPS is. Would you be able to elaborate a bit? I have set the power options to high performance now. Something I've noticed in MSI Afterburner is the Memory Clock and GPU Clock are fluctuating a lot in idle. Seems to drop consistently too.

GPU Clock: 953MHz
Memory Clock: 3004MHz

Then dropping to

GPU clock: 324MHz
Memory Clock: 324MHz

Voltages are showing 0mV and idle temperature of GPU is 38C
 

maxalge

Champion
Ambassador


Disable ULPS is under "settings" button. Check it.
It is a power saving mode that can cause issues like blue screen/stutter.


Yeah leave voltage alone, Just increase power limit to +25. Then make sure you click save and choose a profile number to save the settings to.
 

Alza77

Reputable
Jun 18, 2014
13
0
4,520


The ULPS under 'AMD Compatibility properties' checkbox is greyed out and I can't change it. My power setting is at 100 at the moment, I cannot increase it by 25 and it won't go any lower than 52.
 

maxalge

Champion
Ambassador





Oh, you are using the default skin. Silly me.

Try increasing it to something like 105.



Give that a go with a game, see if it helps.
 

Alza77

Reputable
Jun 18, 2014
13
0
4,520


It seems to be okay and by that I mean that GTA V has not crashed on me yet. GPU maxes at 72C and CPU maxes at 60C. What could this mean about my computer if it is this that is preventing the crashes?

EDIT: I just got a crash on Dying Light. Apparently the game crashed 2 seconds before the nvlddmkm error appeared in the eventviewer.
 

Alza77

Reputable
Jun 18, 2014
13
0
4,520
Okay it has started doing it again even though I put the power to the 105% limit. I'm at my wits end here as I really want to sort this issue out. Any other ideas as to what could be causing it?

WhoCrashed still states this:

On Mon 18/05/2015 22:18:44 GMT your computer crashed
crash dump file: C:\Windows\Minidump\WD-20150518-2318.dmp
This was probably caused by the following module: nvlddmkm.sys (nvlddmkm+0x1405D4)
Bugcheck code: 0x117 (0xFFFFFA8010480010, 0xFFFFF88004FA35D4, 0x0, 0x0)
Error: VIDEO_TDR_TIMEOUT_DETECTED
file path: C:\Windows\system32\drivers\nvlddmkm.sys
product: NVIDIA Windows Kernel Mode Driver, Version 350.12
company: NVIDIA Corporation
description: NVIDIA Windows Kernel Mode Driver, Version 350.12
Bug check description: This indicates that the display driver failed to respond in a timely fashion.
A third party driver was identified as the probable root cause of this system error. It is suggested you look for an update for the following driver: nvlddmkm.sys (NVIDIA Windows Kernel Mode Driver, Version 350.12 , NVIDIA Corporation).
Google query: NVIDIA Corporation VIDEO_TDR_TIMEOUT_DETECTED
 

maxalge

Champion
Ambassador






Your temps are good, so go ahead and increase power limit to max.

But don't touch voltages.

As increasing to 105 did help a bit right?


Have you blown out the dust on the gpu fans?
 

Alza77

Reputable
Jun 18, 2014
13
0
4,520


The increase in power helped for a while but now GTA V has started crashing again under the nvlddmkm error. I have blown off the dust from the GPU fans indeed. Any idea what could be going on component-wise though? I don't want to have to keep worrying about opening MSI Afterburner and increasing voltages before I want to play any demanding games. I would rather figure out if any components are damaged and get them replaced to fix this issue. Thanks again for your help.
 

Alza77

Reputable
Jun 18, 2014
13
0
4,520
Think I am going to go forward with the motherboard RMA to see if the GPU slots are damaged in some way or something. So far that's the only thing I've got as a lead.