Random BSODs: ntoskrnl, dxgkrnl, dxgmms1, nvlddmkm, BCCode 116


  1. Posts : 6
    Windows 7 Home Premium 64bit
       #1

    Random BSODs: ntoskrnl, dxgkrnl, dxgmms1, nvlddmkm, BCCode 116


    Hi Guys. I'd appreciate some help here.

    I've been getting BSODs for quite a while now and would like some help in finding a solution.

    Attached is my SF zip.

    Looking at the dmp files, it seems it's either ntoskrnl or (dxgkrnl, dxgmms1, nvlddmkm). I have no idea what this means.



    Some other info that may be relevant (maybe?):

    Just yesterday, using SlimDrivers, I updated all drivers. This did not help.

    I recently did a memtest and found a faulty stick, which is why I'm down to 3 sticks (I previously had 4). After taking out that stick, I ran memtest several more times and all came back with zero errors. That seemed to reduce (but not eliminate completely) some BSODs.

    Before the onslaught of BSODs occurred, I did have some video card crashes where my screen would black out for a second or two before coming back with a notice saying my video card (or driver?) failed and then recovered (no system shut down or reset, just video). This was completely fixed after I opened up my case and dusted everything.

    Most (if not all BSODs) are followed by BCCode 116 which google tells me is a video card/driver failure? Also, with most BSODs, the monitor turns black (no signal) and the audio loops for a second or so before the system shuts down or restarts. I'm assuming the screen is black and not blue because the video not working.

    I was previously using the newest NVIDIA drivers for my card, but saw some suggestions (through googling) that I could try previous versions to see if they were more stable. Unfortunately, it didn't help. I did the NVIDIA "Clean Install" option every time I installed new drivers. (I did not, however, do the safe-mode remove since I read that "clean install" was essentially the same thing).

    Most BSODs occur when I'm playing games. The games I play do not really require heavy graphics. BSODs happen quite often when playing Civ5, both DX9 and DX11. Sometimes, it crashes when I'm just browsing the web. A more recent BSOD occurred just a few minutes after starting up my computer.

    I'm not sure if this is relevant, but I was able to run 2 3DMark tests without any BSOD, so I'm not sure if the crash really has to do with GPU load.

    I did monitor the CPU and GPU temperatures before some of the BSODs and it did not look like anything was overheating.

    Also, I recently set my BIOS to XMP overclocking (not sure if that's how you say it). Occasionally, I'll get a message on boot that says "overclocking failed". I suspect you guys will tell me to turn it off and see if it still BSOD. BSODs happened with and without overclocking.




    If I need to provide more info, please let me know. Thanks in advance for the help!
      My Computer


  2. Arc
    Posts : 35,373
    Microsoft Windows 10 Pro Insider Preview 64-bit
       #2

    I huess you are overclocking.
    Code:
    CPUID:        "Intel(R) Core(TM) i5 CPU         760  @ 2.80GHz"
    MaxSpeed:     2800
    CurrentSpeed: 2809
    As you are getting BSODs, you should stop overclocking and run all the hardware components like CPU, GPU and RAM to their default settings. Also set the BIOS to default, too.
    How To Clear CMOS (Reset BIOS)

    Check if the Power Supply Unit (PSU) supplying adequate power to the computer or not.
    eXtreme Power Supply Calculator
    At least 30% free is good. Also let us know about the wattage of your PSU.

    Is the computer hot? Report us the heat of the computer after a couple of hours of your normal usage. Upload a screenshot of the summery tab of Speccy.

    Stress test the Graphics Card using Furmark.
    Video Card - Stress Test with Furmark

    Display driver installed is seeming to be latest version ... so not touching it right now. Let us know the results. Depending on those we will go for further troubleshooting.
    _________________________________________________________________________
    Code:
    *******************************************************************************
    *                                                                             *
    *                        Bugcheck Analysis                                    *
    *                                                                             *
    *******************************************************************************
    
    Use !analyze -v to get detailed debugging information.
    
    BugCheck 116, {fffffa800a0474e0, fffff880051b2ed0, ffffffffc000009a, 4}
    
    Unable to load image \SystemRoot\system32\DRIVERS\nvlddmkm.sys, Win32 error 0n2
    *** WARNING: Unable to verify timestamp for nvlddmkm.sys
    *** ERROR: Module load completed but symbols could not be loaded for nvlddmkm.sys
    Probably caused by : nvlddmkm.sys ( nvlddmkm+8e9ed0 )
    
    Followup: MachineOwner
    ---------
    
    3: kd> lmvm nvlddmkm
    start             end                 module name
    fffff880`048c9000 fffff880`053aa000   nvlddmkm T (no symbols)           
        Loaded symbol image file: nvlddmkm.sys
        Image path: \SystemRoot\system32\DRIVERS\nvlddmkm.sys
        Image name: nvlddmkm.sys
        Timestamp:        Mon May 13 00:39:45 2013 (518FE8F9)
        CheckSum:         00AB6DED
        ImageSize:        00AE1000
        Translations:     0000.04b0 0000.04e4 0409.04b0 0409.04e4
      My Computer


  3. Posts : 6
    Windows 7 Home Premium 64bit
    Thread Starter
       #3

    Thanks Arc!

    I'm at work, but I can address some of your points. I'll get back to you on everything else when I get home.


    Arc said:
    Check if the Power Supply Unit (PSU) supplying adequate power to the computer or not.
    eXtreme Power Supply Calculator
    At least 30% free is good. Also let us know about the wattage of your PSU.
    My PSU is: 600 Watts - XtremeGear Power Supply
    I did a quick google and got some negative feedback regarding it.
    I may be doing the calculator wrong, but it's coming close to 600 watts. Maybe this is the problem?







    Arc said:
    Stress test the Graphics Card using Furmark.
    Video Card - Stress Test with Furmark
    Is 3dMark not a good stress test? My system survived two tests.







    Arc said:
    Display driver installed is seeming to be latest version ... so not touching it right now. Let us know the results. Depending on those we will go for further troubleshooting.
    NVIDIA's latest driver is 320.49
    IIRC, I'm using the 320.18
    Perhaps I did not do the clean install correctly?
      My Computer


  4. Arc
    Posts : 35,373
    Microsoft Windows 10 Pro Insider Preview 64-bit
       #4

    Stopping the overclocking is necessary, at first. The next priority is the PSU. If it is approaching to the maximum of the capacity, have a bigger one, say 750 watts or more.

    3D mark is not bad at all, but give furmark another run.

    And, the system heat after some reasonable time of usage is also needed.

    Will look at the driver at last.

    Let us know when done.
      My Computer


  5. Posts : 6
    Windows 7 Home Premium 64bit
    Thread Starter
       #5

    Thank you again, Arc, for your help.

    Update:

    I set the BIOS back to default.

    Turned on the computer and ran FurMark. BSOD (crash time 6:43PM) immediately after I clicked the ok on the warning about heavy loading to gpu (even before the graphics showed up). Immediately following the crash, the gpu fan spun up very fast. Had to hard reset.

    After hard reset, tried FurMark a second time. This time I got to the graphics, but BSOD (crash time 6:47PM) less than a second into the test. No fast gpu fan this time. Hard reset.

    After hard reset, tried FurMark a third time, got to the graphics, and then BSOD (crash time 6:50PM) less than a second into the test. No fast gpu fan spin. Hard reset.


    (The gpu fan is normally at 40% and is functioning correctly during normal use)



    I was going to capture the speccy after the test, but I don't think temperature has anything to do with it, since it's crashing before there's any heavy use.

    Current idling CPU temperature is 46 C
    Current idling GPU temperature is 53 C (is this high?)
    Current idling HDD temperature is 40 C

    Attached is an updated SF zip.
    Last edited by dapoofyhairdude; 12 Jul 2013 at 21:37.
      My Computer


  6. Posts : 6
    Windows 7 Home Premium 64bit
    Thread Starter
       #6

    And now I sit in front of my computer twiddling my fingers, hoping that Arc is awake at 8AM on a saturday morning.
      My Computer


  7. Arc
    Posts : 35,373
    Microsoft Windows 10 Pro Insider Preview 64-bit
       #7

    So you can understand why I wanted you to give furmark a try.

    If the Graphics crash is caused by stress testing, it is the situation that the GPU is failing to take the stress. Which indicates to a failing GPU.

    The GPU temperature is not too high, but is tending to be high.

    Before locking the suggestion as a conclusion, I would suggest you to use the onboard graphics, disable/remove the graphics card. If the onboard graphics produce the same BSOD, the problem might be caused by any other thing, or any more component is contributing to the BSOD. But if it all right, then it is he GPU only that is failing.

    And, upload a screenshot of Speccy. It will be helpful to draw some necessary inferences.
      My Computer


  8. Posts : 6
    Windows 7 Home Premium 64bit
    Thread Starter
       #8

    Thanks Arc!
    Another very interesting update:



    I had some time before your latest response, so I did the following things to see if any would help.



    I opened up my case and dusted everything again. I paid particular attention to the PSU and tried to get it as clean as possible. I also put the tower on a wooden board since it was sitting on carpet and the PSU's intake is coming from below the case. (I hope I didn't already burn out the PSU).

    After doing so, I booted up and ran FurMark. Again, BSOD (crash time 9:36) less than a second into the test.



    Next I decided to try an even older NVIDIA driver to see if it was more stable. (I previously was using 320.49 and 320.18; both did not help). I "clean installed" 314.22. Restarted the computer, then ran FurMark. Interestingly enough, this time it did not BSOD immediately. It ran for ~5 minutes and then BSOD (crash time 10:07). This piqued my interest since it was a different result than before.


    After a hard reset, I installed an even older NVIDIA driver: 314.07. Again, "clean install" and then a system restart. This time, FurMark ran for an entire hour before I manually hit "ESC" to end the test. No BSOD! I started FurMark again, just to make sure I wasn't just lucky. It ran for 5 minutes and then I stopped it again. I have attached a screenshot of the test running along with speccy on the 2nd monitor at the 1 hour mark of the test. CPU @ 65C and GPU @ 82C.


    I've also attached a new SF zip.



    Also, in response to your suggestion to use onboard, my mobo doesn't have a vga out =/



    Now I'm not sure how to proceed. I guess I can run some other programs that previously caused BSOD and see what happens.
      My Computer


  9. Posts : 6
    Windows 7 Home Premium 64bit
    Thread Starter
       #9

    Dang. Just tried Rift, one of my games that caused BSOD.

    Prior to installing an older driver, it used to crash 1-2 seconds after the game loaded the first menu screen.
    This time, I got through much more: I got through changing to windowed mode and changing resolution. Finished creating a character. Got through two seconds of the intro cutscene and then BSOD. Same type as before.


    It seems the old driver helped, but did not fix the issue. Do you think it's worth going to even older drivers? Or am I just wasting my time?
      My Computer


  10. Arc
    Posts : 35,373
    Microsoft Windows 10 Pro Insider Preview 64-bit
       #10

    If you are behind a stable version of nvidia display driver, this one is one year old, but the most stable one I have seen lately. It never crashed on its own. Try GeForce 306.23 Driver

    But as it is almost evident, it is the GPU that is failing. Still, give the suggested driver a try, for your mental satisfaction.

    I guess the system is not very old, and the GPU is within warranty?

    Need to know the CPU temperature too, after an hour of normal usage.
      My Computer


 

  Related Discussions
Our Sites
Site Links
About Us
Windows 7 Forums is an independent web site and has not been authorized, sponsored, or otherwise approved by Microsoft Corporation. "Windows 7" and related materials are trademarks of Microsoft Corp.

© Designer Media Ltd
All times are GMT -5. The time now is 11:03.
Find Us