BSOD from nvlddmkm (running verifier) when trying to open GeForce Exp

Hi all,

Sorry in advance for the long and rambling post, but I've tried a few different sets of settings with driver verifier in an effort to pinpoint the problem, but I'm not at all experienced in the nitty gritty aspects of debugging BSODs. I'll try to explain the origin of all of my earlier dump files below in italics, with my most recent dump files pertaining to the text after the italics:

I recently built a new desktop for gaming, but it seemed to BSOD randomly while playing Battlefield 4 (0x50 usually). There was also one BSOD while playing Spec Ops: The Line, but I haven't been able to replicate it.

First, I reinstalled Windows 7, but that didn't fix the problem. Looking at the ensuing BSODs, the 0x50 bugchecks originated from dxgmms1.sys and a single 0xA originated from something called nt, during bf4.exe. I ran Memtest86+ overnight and woke up to see 7 passes and 0 fails, so it didn't seem to be memory. I then ran driver verifier, setting all settings except Low Resolution Simulation to On, verifying only non-Microsoft drivers. This caused eventual BSODs that pointed to nvlddmkm.sys, though I don't remember what I was doing at the time of these BSODs. After this, I did a clean reinstall of NVIDIA 332.21 Driver through GeForce Experience. I also reseated the video card in my motherboard, making sure all the power connections were in all the way. At this point, it appeared that the driver verifier was messing with my ability to browse the internet (pages weren't fully loading, so I couldn't play BF4), so I turned driver verifier off.

At this point, I tried the same settings with driver verifier (everything on except Low Res Simulation) and checked the dxgmms1.sys driver, as well as a driver named ntoskrnl.sys. This, I think, gave me useless BSODs whenever I tried to log in to windows, telling me that my Creative audio driver was trying to do something illegal. I think this is extraneous because the only thing I changed was adding Microsoft drivers to the list, and I read somewhere that having the Force Pending I/O Requests setting on may result in some spurious BSODs if a driver doesn't deal with receiving a pending signal for I/O.

Finally, yesterday, I turned on driver verifier to standard settings to verify all drivers. I ended up playing Spec Ops last night for several hours without issue. However, my most recent dumps, from today, seem to point to nvlddmkm.sys as the culprit, with nvspcaps64.exe being the offending program at the time of dump. I found that every time I tried to open GeForce Experience (which seemed to be running in my taskbar), I got a BSOD pointing to these two things. I believe nvspcaps64.exe has something to do with Shadow Play, but after turning driver verifier off, I saw that Shadow Play was not on during these BSODs. I'm not sure if this means that GeForce Experience is bad, or if it's my hardware.

Lastly, I believe I can return my video card (or maybe it could the the power supply?) to Microcenter up until Friday, if it is actually corrupted hardware. This is obviously not my first choice, but if you guys recommend that, I'm pretty sure it's doable.

Again, sorry for the long post, and attached are my dumps and system information, grabbed through SF Diagnostic Tool. Thanks in advance for helping me out! :) Please let me know if you need any other information and I'll gladly provide it.

Hi Arc,

Thanks for the reply! I'll follow these steps when I get home later this afternoon. A couple quick questions:

How do I know if I'm using NVIDIA chipset drivers? Is it one of those things that I would know if I were using them?
Should I also uninstall GeForce experience?

I'll let you know as soon as I get through all of your outlined steps.

Hi Arc,

Got through all the steps and ran FurMark on my computer. The temperature seemed to top out at about 80 to 82 degrees C in a few minutes, so I stopped it. On my first run, I didn't notice any artifacts, but when running it a second time a couple minutes later, I saw some artifacts pop in and out in the lower left quadrant of my screen. Does this mean the graphics card has some issues, or did I run it too soon after the previous one? I can imagine that repeated heating of the GPU might not be good for the card.

On the upside, I didn't experience a BSOD from running it, though I only ran it for a few minutes. Should I run the test for longer to see if I get one?

I've just exchanged the GPU and it seems fine so far. Stress tested it, and I still see some strange white artifact that pops in and out for a while on the bottom half of the screen during Furmark, though it becomes less frequent as the test goes on. This makes me think that what I'm seeing has nothing to do with the GPU, but I'm not sure.

Still no BSODs after the clean install of drivers, but I'll keep playing games to test it over the next few days. If nothing crops up, I'll mark the thread solved.

Thanks for your help! :)

It's remarkably hard to get a screenshot of what I'm talking about, so I pointed my phone's camera at it and took a video (kind of horrible quality). Here's the video on YouTube: https://www.youtube.com/watch?v=ukYRI70fJmk

As you can hopefully see, there's a white line near the bottom that kind of pops in and out. Over time, it seems like this behavior becomes more infrequent, as the temperature of the GPU gets to its maximum, but I can't imagine why that would be. I also tried this on my TV through HDMI, and it looks the same. Any suggestions? Is this even an artifact?