| Page 1 of 1 |
|
|
 Posted: Tue, 25th Feb 2014 12:50 Post subject: Video driver has stopped responding and has recovered |
|
 |
So this issue has haunted me for years and through 2 new PCs.
The most recent box is brand new, high end stuff. So it can't be the hardware in it.
Other thing that is similar between both PCs is 2 monitor setup, power+video cables, wall power sockets, extender/splitter that they're all connected to.
Now I've changed 1 of the monitors, with a brand new. Changed the power+video cables.
Now I'm searching for a lenghty power cable, to plug the PC on the other wall socket.
The issue itself is also kind of strange. When I play a new game, using lots of DX11 goodies, after 1-4 hours of play, the video driver starts crashing. Reboot doesn't help. Only leaving the PC and comming back at a later time fixes the problem or playing another DX9 game in the mean time.
It's like something is overheating, but my rig is very well built/cooled and with powerful stuff, so that while playing DX11 games at max settings, all temperatures are under 50C with only the GPU going maybe 60C. And when the driver crashes and I stop playing, they all go to 30-40C within minutes, but if I start playing DX11 right away it crashes within 5min of play.
Also stress tests do jack shit. They can get all my rig to 50C and the GPU to 70C and it's stable for hours, even if I start it right after a video driver crash.
My guess is that the thing that is overheating is something without a diode, a cable outside the case or inside, power installation/cables inside the walls, or a PCI-E socket, or a specific chip/capacitor or something like the NB or SB of the MB, but I can't believe this happens on 2 separate cases with 2 separate brand new MBs and GPUs.
I'm thinking of getting an infrared camera or a laser thermometer. If something is getting so hot, it should be bright red when all of the PC is under 50C.
Of course it could be software related, but then why would id fix itself with time and not restart.
| Lutzifer wrote: | | and yes, mine is only average |
|
|
| Back to top |
|
|
zmed
Posts: 9234
Location: Orbanistan
|
| Posted: Tue, 25th Feb 2014 12:57 Post subject: |
|
|
I think it's actually some windows safety BS called 'Timeout Detection and Recovery'.
I had this problem for ages as well. There is a way to disable it, and only had one or two right after I did it, not a single one ever since (maybe it needed a restart), even in games that were extremely prone to have this happen (like War Thunder).
I can't remember how I did it, stumbled upon it somewhere, involved digging into the registry IIRC.
There were no discernible reasons for the crashes on my end either. No overheats, connections just fine, latest drivers, etc. For me it was an over-zealous safety feature in W7.
|
|
| Back to top |
|
|
|
|
| Posted: Tue, 25th Feb 2014 13:13 Post subject: |
|
|
It's not that. The video actually crashes.
I've set the TdrDelay to 10sec, just when the driver crashes my monitor stays the same frame or black for 10sec before it recovers.
I've even set TdrLevel to 0 (disabled) and when it crashes it just hangs indefinitely without recovering until reboot.
And yes, after setting the registry keys DWORD not QWORD, you need to reboot.
The sound continues with no problem while he video is hung*
| Lutzifer wrote: | | and yes, mine is only average |
|
|
| Back to top |
|
|
zmed
Posts: 9234
Location: Orbanistan
|
| Posted: Tue, 25th Feb 2014 13:34 Post subject: |
|
|
Well, it seems there is more to it then, the edit seemingly stopped it on my rig. Although I also tried other stuffs before that, so it might be a combination of lots of things that finally did it.
|
|
| Back to top |
|
|
|
|
| Posted: Tue, 25th Feb 2014 13:35 Post subject: |
|
|
I still think it's a hardware issue, because it "fixes" itself with time.
| Lutzifer wrote: | | and yes, mine is only average |
|
|
| Back to top |
|
|
|
|
| Posted: Tue, 25th Feb 2014 16:47 Post subject: |
|
|
So I updated the BIOS of my GA-Z87X-D3H from F5 to F8, because it says that it supports Haswell from F7.. and I'm running a 4670K.. 
Couple of things got detected, 1 more case fan also.
But it still crashes. 
| Lutzifer wrote: | | and yes, mine is only average |
|
|
| Back to top |
|
|
|
|
| Posted: Tue, 25th Feb 2014 16:56 Post subject: |
|
|
had the same issue on newer nvidia drivers.
only install normal and physx drivers, don't install all the 3d stuff and geforce expierence and what else there is.
like i said i had the same problem but only while watching or playing on my tv(2nd screen).
after a fresh driver install with only drivers and physx everything runs good.
maybe it's because of my shitty gts250 but i don't know... maybe this will help you.
|
|
| Back to top |
|
|
|
|
| Posted: Tue, 25th Feb 2014 19:20 Post subject: |
|
|
The old monitors used HDMI, the new uses Display port.
Tried only the new monitor and the Case on another power socket - still crashes.
All temperatures under 50C and all voltages steady, only +12 went up by 0.5V after the crash.
| Lutzifer wrote: | | and yes, mine is only average |
|
|
| Back to top |
|
|
|
|
| Posted: Wed, 26th Feb 2014 21:48 Post subject: |
|
|
So I removed the XMP profile of my ram, ran it in default 800MHz - no help.
Even tried to disable turbo of the CPU - still no help.
It's funny that BF4 on max doesn't raise any problems, but AC4:BF and Thief do..
Also if THOSE FUCKERS AT MICROSOFT COULD POST SOME FUCKING DUMP/ERROR CODES FOR VIDEO DRIVER CRASHES IT WOULD BE GREAT!!! 
| Lutzifer wrote: | | and yes, mine is only average |
|
|
| Back to top |
|
|
Invasor
Moderator
Posts: 7638
Location: On the road
|
| Posted: Wed, 26th Feb 2014 22:05 Post subject: |
|
|
I believe this is (video) driver related. Don't know which version would fix it though...
|
|
| Back to top |
|
|
doobzilla
Posts: 1099
Location: Team America's Mount Rushmore Base. Stolen from Indians.
|
| Posted: Wed, 26th Feb 2014 23:50 Post subject: |
|
|
What video card do you have? I believe if you check in your event viewer, you will see that the error is caused by nvlddmkm.sys. If this is the case, and your GPU is of the older variety, you will be best served to find the last set of drivers that would've been made right before the next generation of cards came out. This is what I had to do to resolve my TDR errors way back in the Vista and 8800GTX days...
Also, if you are running any sort of "Dreamscene" desktop-as-a-video software, be sure to get rid of it as Windows Dreamscene is a known culprit.
I hope you get it sorted. Cheers!
|
|
| Back to top |
|
|
|
|
| Posted: Thu, 27th Feb 2014 00:58 Post subject: |
|
|
Oh yea, that's one of the things I tried - setting windows theme to Windows Classic - didn't help either.
| Lutzifer wrote: | | and yes, mine is only average |
|
|
| Back to top |
|
|
|
|
| Posted: Fri, 28th Feb 2014 15:27 Post subject: |
|
|
With or without page file - no difference, still crashes.
| Lutzifer wrote: | | and yes, mine is only average |
|
|
| Back to top |
|
|
|
|
| Posted: Fri, 28th Feb 2014 21:50 Post subject: |
|
|
| Quote: | *******************************************************************************
* *
* Bugcheck Analysis *
* *
*******************************************************************************
VIDEO_TDR_TIMEOUT_DETECTED (117)
The display driver failed to respond in timely fashion.
(This code can never be used for real bugcheck).
Arguments:
Arg1: fffffa8014f68010, Optional pointer to internal TDR recovery context (TDR_RECOVERY_CONTEXT).
Arg2: fffff88007755d84, The pointer into responsible device driver module (e.g owner tag).
Arg3: 0000000000000000, The secondary driver specific bucketing key.
Arg4: 0000000000000000, Optional internal context dependent data.
Debugging Details:
------------------
FAULTING_IP:
nvlddmkm+153d84
fffff880`07755d84 ?? ???
DEFAULT_BUCKET_ID: GRAPHICS_DRIVER_TDR_TIMEOUT
TAG_NOT_DEFINED_202b: *** Unknown TAG in analysis list 202b
BUGCHECK_STR: 0x117
PROCESS_NAME: System
CURRENT_IRQL: 0
STACK_TEXT:
fffff880`039db4f0 fffff880`072c0737 : fffffa80`14f68010 fffff880`0730d038 fffffa80`14f68010 fffff880`0728e843 : watchdog!WdDbgReportRecreate+0xa3
fffff880`039dba10 fffff880`072c13fc : fffff8a0`0312a830 fffff8a0`0312a830 00000000`00000080 fffffa80`14f68010 : dxgkrnl!TdrUpdateDbgReport+0xcb
fffff880`039dba60 fffff880`072956b3 : 00000000`00000001 fffffa80`1072a000 00000000`00000000 fffff880`000000a9 : dxgkrnl!TdrCollectDbgInfoStage2+0x220
fffff880`039dba90 fffff880`072c1f4f : fffffa80`10729a58 ffffffff`fffe7960 fffffa80`14f68010 00000000`00000000 : dxgkrnl!DXGADAPTER::Reset+0xef
fffff880`039dbb40 fffff880`0739103d : fffffa80`13307110 00000000`00000080 00000000`00000000 fffffa80`10729410 : dxgkrnl!TdrResetFromTimeout+0x23
fffff880`039dbbc0 fffff800`02f5b2ea : 00000000`fffffc32 fffffa80`103b8610 fffffa80`0ca4b040 fffffa80`103b8610 : dxgmms1!VidSchiWorkerThread+0x101
fffff880`039dbc00 fffff800`02caf8e6 : fffff880`03365180 fffffa80`103b8610 fffff880`0336ffc0 00000000`00000000 : nt!PspSystemThreadStartup+0x5a
fffff880`039dbc40 00000000`00000000 : fffff880`039dc000 fffff880`039d6000 fffff880`0b804d70 00000000`00000000 : nt!KiStartSystemThread+0x16
STACK_COMMAND: .bugcheck ; kb
FOLLOWUP_IP:
nvlddmkm+153d84
fffff880`07755d84 ?? ???
SYMBOL_NAME: nvlddmkm+153d84
FOLLOWUP_NAME: MachineOwner
MODULE_NAME: nvlddmkm
IMAGE_NAME: nvlddmkm.sys
DEBUG_FLR_IMAGE_TIMESTAMP: 52f651cf
FAILURE_BUCKET_ID: X64_0x117_IMAGE_nvlddmkm.sys
BUCKET_ID: X64_0x117_IMAGE_nvlddmkm.sys
Followup: MachineOwner |
so this is the contents of the dump file created when the video driver crashes, but I can't get any useful information for now out of it..
| Lutzifer wrote: | | and yes, mine is only average |
|
|
| Back to top |
|
|
|
|
| Posted: Sat, 1st Mar 2014 00:14 Post subject: |
|
|
tried reverting to 331.65 as they claim it's the last very stable driver - no help
and a pic of an actual video driver crash
| Lutzifer wrote: | | and yes, mine is only average |
|
|
| Back to top |
|
|
LeoNatan
☢ NFOHump Despot ☢
Posts: 73196
Location: Ramat Gan, Israel 🇮🇱
|
| Posted: Sat, 1st Mar 2014 00:34 Post subject: |
|
|
Check if you have a C:\TEMP folder.
Search your hdd for "iehighutil.exe".
|
|
| Back to top |
|
|
|
|
|
| Back to top |
|
|
|
|
|
| Back to top |
|
|
|
|
| Posted: Sat, 1st Mar 2014 13:37 Post subject: |
|
|
My GTX570DCII had similar issues back in the days. I needed to bump vcore.
However, after getting new GPU, I sold it and the buyer has had 0 issues with it. Makes me wonder, what might have been the problem?
New GPU works fine under loads etc. And GPU before 570 worked also fine.
i5 6600k @ 4.3 GHz | MSI z170 Gaming M7 | 32GB Kingston HyperX Fury | 850 Evo 500GB | EVGA 1070 SC | Seasonic X-660 | CM Storm Stryker
|
|
| Back to top |
|
|
|
|
| Posted: Sat, 1st Mar 2014 15:09 Post subject: |
|
|
I preffer underclocking, than upping the voltage.
| Lutzifer wrote: | | and yes, mine is only average |
|
|
| Back to top |
|
|
|
|
|
| Back to top |
|
|
| Page 1 of 1 |
All times are GMT + 1 Hour |
