|
Problem description: I built a new computer over Black Friday/Cyber Monday. The only part that isn't new is the video card, which was working fine in it's old home. It seems to be perfectly stable in Windows, but with games it will sometimes hard lock with a black screen or BSOD. In either case, the crash dump says the error is a DPC_WATCHDOG_VIOLATION probably caused by nvlddmkm.sys, but I don't know how to dig any deeper. I didn't have any games to really stress it with until Christmas, so it's taken me longer than I'd like to pick up on the pattern  .Attempted fixes:
Recent changes: Nope. -- Operating system: Windows 8.1 64bit System specs: Apparently I didn't save the build when I built it, but I'm pretty sure this is all right. CPU: Intel Core i5-4690K 3.5GHz Quad-Core Processor CPU Cooler: Noctua NH-U12S 55.0 CFM CPU Cooler Motherboard: ASRock Z97M Pro4 Micro ATX LGA1150 Motherboard Memory: G.Skill Ares Series 8GB (2 x 4GB) DDR3-2400 Memory Storage: Samsung 840 EVO 250GB 2.5" Solid State Drive Video Card: Asus GeForce GTX 660 2GB Video Card Case: Nanoxia NXDS4B MicroATX Mini Tower Case Power Supply: Rosewill Capstone 750W 80+ Gold Certified ATX Power Supply Wireless Network Adapter: Gigabyte GC-WB867D-I 802.11a/b/g/n/ac PCI-Express x1 Wi-Fi Adapter The RAM is 1.65 volt ram, which I was told in the part picking thread was fine for Haswell refresh. I haven't done any overclocking yet because I haven't felt a desperate need for it (and I'd like to solve the BSODs first), so the RAM is running at 1600 speeds, which is 1.5v anyway. Location: USA I have Googled and read the FAQ: Yes Please help, I want to play games again without worrying that they'll bring my whole system down  .Edit: I'm dumb and forgot to link the crash dump. Edit2: Found a guide on how to use Driver Verifier to try to track down the real problem and (after a mild panic when System Restore claimed it failed when it really didn't) it pointed at BazisVirtualCDBus.sys, which belongs to WinCDEmu. I didn't realize Windows 8 could mount isos natively, so I'll try uninstalling WinCDEmu and see if that fixes things. Edit3: Welp, it didn't fix it. This time it started to lock up for ~1 second ever other second or so. If I alt-tab, it would fix itself for a few minutes, then start doing it again. After a few rounds of that, it wouldn't help anymore, screen went black and the computer restarted. It didn't give me a crash dump this time. The Event Viewer shows a bunch of entries that look like this: quote:The description for Event ID 14 from source nvlddmkm cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer. ZombieApostate fucked around with this message at 23:14 on Jan 27, 2015 |
#
?
Jan 27, 2015 15:05
|
|
|
|
| # ? Apr 24, 2024 12:27 |
|
|
Run DDU http://www.wagnardmobile.com/DDU/ and then install the latest Nvidia beta drivers to be sure that's not the issue.ZombieApostate posted:Ran Memtest for 4 hours without errors I'd try running it overnight at least once (but the RAM is probably okay). If neither of those help I'd temporarily take the GPU out and use onboard video and see if that changes anything.
|
|
#
?
Jan 27, 2015 23:56
|
|
|
Cool, thanks, I just gave DDU a try, we'll see if that helps. I used to use Driver Sweeper, but I think the link I found for DDU the last time I looked for it had some adware crap bundled with it. Thankfully your link was clean. And yeah, I know 4 hours with memtest probably wasn't long enough. I'll do a longer run if I'm still having problems.
|
|
#
?
Jan 28, 2015 08:27
|
|
|
It locked up again. Game video froze except for the mouse, a few seconds later the mouse froze as well, music continued playing. Soon after, both monitors go black and act like they went to sleep. No keyboard or mouse input seemed to do anything. I waited a few minutes to see if it would actually restart like it should if it BSODs, but it didn't. No crash dump generated. Event Viewer shows a single instance of the same error at the end of my first post, then 7 of these warnings: quote:Display driver nvlddmkm stopped responding and has successfully recovered. Neither of these latest two crashes have referenced the DPC_WATCHDOG_VIOLATION I was getting before, so it looks like that might have been from WinCDEmu after all. At least that narrows things down a little. I just reseated the video card and switched the 6 pin power plug to a plug closer to the power supply, but I don't have any good ideas left. I guess I'll have to run a longer memtest run, but it's not sounding much like a RAM problem. Any other good ideas? Is it possible this could be caused by the crappy old electricity lines from my house? My old laptop used to crash in similar conditions, complaining about AMD drivers if I remember correctly. I've got a surge protector, but I doubt it'd help much if the waveform, or whatever the term is, was all hosed up. The computer upstairs never had any problems but I know the basement (where the new computer is) got flooded a number of years back and sometimes the lights flicker a little bit.
|
|
#
?
Jan 29, 2015 10:08
|
|
|
ZombieApostate posted:It locked up again. Game video froze except for the mouse, a few seconds later the mouse froze as well, music continued playing. Soon after, both monitors go black and act like they went to sleep. No keyboard or mouse input seemed to do anything. I waited a few minutes to see if it would actually restart like it should if it BSODs, but it didn't. No crash dump generated. Did this happen when using the onboard video? ZombieApostate posted:Is it possible this could be caused by the crappy old electricity lines from my house? If it's only happening during game I'd think probably not. ZombieApostate posted:I've got a surge protector, but I doubt it'd help much if the waveform, or whatever the term is, was all hosed up. The computer upstairs never had any problems but I know the basement (where the new computer is) got flooded a number of years back and sometimes the lights flicker a little bit. If you're concerned about power the easiest test would be plugging one of these into various outlets: http://www.amazon.com/Gardner-Bender-GFI-3501-Outlet-Tester/dp/B00170KUPC Surge protectors/UPSs/power strips can go bad but that's not a common issue. You could use an outlet tester on those as well.
|
|
#
?
Jan 29, 2015 22:44
|
|
|
I haven't tried playing with onboard video yet. It is currently about 8 hours into a memtest run, no errors yet. I tried the SMP version this time and it froze up after 1 minute 34 seconds, but single threaded works fine. It seems like I'm not the only one with that problem though, so it's probably a software issue with memtest itself. I'll give memtest a bit longer and then try playing with onboard for a while. My prettiiiiieeeeesssss~ 
|
|
#
?
Jan 30, 2015 14:38
|
|
|
|
| # ? Apr 24, 2024 12:27 |
|
|
I think I finally figured this out, and figured I should post the solution in case it helps someone else. I knew my RAM was a little unusual since it is DDR3-2400 (it was on sale~), but I I thought the BIOS would pick timings that would at least work. Apparently that was not a correct assumption. It defaulted to something like DDR3-1600 11-11-12-28. I told it to use the XMP profile timing (DDR3-2400 11-13-13-31), which matched what was on Newegg and I haven't had a single problem since.
|
|
#
?
Mar 29, 2015 19:32
|
|