So I've done some reading around the forums about the issue of the monitor going black in the middle of gaming, browsing, working, etc. and I have the same issue, but a few questions to address. PC are specs at the bottom of this post.
I built this PC for digital content creation (3D modeling, digital painting, real-time rendering). The monitor going black issue is completely random on my PC at different intervals and time periods and seems to happen mostly when I am working on a 3D scene. It has happened before while gaming as well, or browsing the web.
The strange thing about this error was that even though the monitor goes black, any music that was playing stays on for about 30 seconds or so before the PC restarts it's self. After the PC has restarted a window pops up stating the issue produced a minidump file. Here's some copied text from the minidump:
Attempt to reset the display driver and recover from timeout failed.
Arg1: fffffa801d247170, Optional pointer to internal TDR recovery context (TDR_RECOVERY_CONTEXT).
Arg2: fffff88011324ae4, The pointer into responsible device driver module (e.g. owner tag).
Arg3: ffffffffc000009a, Optional error code (NTSTATUS) of the last failed operation.
Arg4: 0000000000000004, Optional internal context dependent data.
Use '!findthebuild' command to search for the target build information.
If the build information is available, run '!findthebuild -s ; .reload' to set symbol path and load symbols.
FAULTING_MODULE: fffff80003614000 nt
nvlddmkm!_NULL_IMPORT_DESCRIPTOR <PERF> (nvlddmkm+0xc3bae4)+0
fffff880`11324ae4 4883ec28 sub rsp,28h
So I had a though about the PSU. I purchased the PSU for this PC in 2008 when I was wanting to finish this build, but never had the funds. The PSU was unused for a period of time from 2008 until 2011 when I finished the build. One of the strange things I've noticed when I completed this build is when I press the power button to turn it on, it stays on for one second and shuts off. Four seconds later, the PC turns back on and starts up. This happens every time I turn off the surge protector it's plugged into. When I originally purchased this PSU, I knew nothing about power draw or current requirements of PSUs. So when I checked out the label on the side of the PSU, I noticed that its current requirements are 18amps. The surge protector the PC is plugged into is rated at 15 amps and the breaker for the outlet is only 20 amps. I had been wondering if the current draw of the PSU had been the issue.
In addition to checking out the minidump file, I also checked out the Event Viewer and it recorded on the day and time of the minidump an error with a file named nvlddmkm.sys, which is apparently the Nvidia driver. I keep the drivers updated weekly because of this issue, as it makes it nearly impossible to get any work done (thank goodness for autosaves and backups). What I'm trying to figure out is why the most recent driver is causing an issue with my PC. Here's the text from the error reporting webpage:
Address a problem with NVIDIA Graphics Driver
NVIDIA Graphics Driver has stopped working properly.
A driver update, if available, might prevent this problem from recurring.
There are several ways of locating and installing driver updates, but it is best to let Windows do this for you. Try the first step below, which describes the process. If it doesn't produce a driver update that solves the problem, then try the remaining steps in the order given.
> Check for optional third-party updates
> Try updating NVIDIA Graphics Driver manually
> Check the NVIDIA Corporation website for driver updates
> Check your computer manufacturer's website for driver updates
My questions: Could the PSU be to blame with the current draw, causing it to lose power and cause issues? Could it be the drivers?
Any help is appreciated. PC specs below.
Intel Core i7 K875 CPU
EVGA NVidia 460GTX 2Win GPU
16GB DDR3-1333 RAM
GIGABYTE GA-P55A-UD4P Mobo
Western Digital WD5000AAKX 500GB SATA HDD
ABS Tagan ITZ Series 1100W PSU
Soundblaster Audigy 2 SPU
Windows 7 x64 Ultimate
Acer X203W + Acer AL1916W monitors
Ah! Forgot to mention a few other things. I do not believe temperature is an issue either. I've been using the EVGA Precision app to monitor temps to ensure the GPU doesn't get too hot. The temps usually never crest over 60C, and they are currently sitting at 45C / 50C for Core1 / Core2 respectively. When working in DCC software the temps usually climb to about 62C and to a max of 63C. Gaming it gets a bit more intense at about 80C per core.
Based on the description of the problem the two most likely reasons are:
1. Faulty drivers - though if you have been updating frequently it becomes unlikely that every new driver revision would have the same problem. Still, if I were you, I'd download the last WHQL driver (very likely not to be the latest driver) and install it after _fully_ uninstalling the current driver. Since doing this doesn't cost anything, that's the first thing I'd try.
2. If 1. above does not solve the problem, I'd strongly suspect that the video card hardware is malfunctioning. I say this because the information you posted about the minidump clearly indicates that the problem is linked to the video card. If the drivers are not the culprit that leaves the video hardware holding the bag. If you have access to another video card, that would be an easy way of figure out if it is the culprit.
Lastly, I suppose temps of 80C per core are in the acceptable range but, they are certainly outside my "comfort zone". The temp may not be an issue at all but, if there are easy and reasonable steps you could take to lower that maximum, I'd give that a shot too.
Thanks for the reply. I'll give the WHQL drivers a try. I attempted a clean install of the drivers that came with the GPU using Driver Sweeper, and cleared out all Nvidia-GPU-related files. After installing the CD drivers, I still had the same error (this time while randomly posting to another forum). I'll download a WHQL driver and perform another clean install and see if anything changes.
do you got a good air draft in the case at 80 C if the card giving error then you need to refresh her also dust off the fan
Yeah dusting off the fans is something I need to do, and usually often. But the card only hits 80C per core when playing a game like FONV or Far Cry 2. I have EVGA Precision sitting on my desktop so I can monitor the temps, and on average when this issue happens the temperature is usually idle or light load temp at about 49C. But this issue has plagued me since the first week I had the card installed in the rig.