Very Strange Recurring BSOD on Windows 8.1

[Apples555]

I've been having a very strange recurring BSOD on my Windows 8.1 installation. I don't even know what to think. The dump file is attached.

It occurs almost completely randomly. If I am listening to something, the sound output loops a short millisecond of sound and everything completely freezes until the blue screen comes up and the computer restarts. What I have done:

-disabled dynamicticks
-reinstalled, uninstalled, and installed all the compatible versions of the Nvidia video and audio drivers.
-disabled fast startup
-took out one stick of RAM, then upon BSOD encounter switched the other one in with the same result
etc.

Yet the BSOD keeps happening. Where do I look for the cause?

View attachment 66201

EDIT: It happened again, just after posting this thread.

 

[axe0]

Hi Apples555,

Please check Windows Update for updates, install every update available, it may take a couple of times to get them all.

For a clean re-install of the Nvidia GPU drivers, please uninstall everything of Nvidia using Display Driver Uninstaller and install new drivers from your manufacturer.

*** WARNING: Unable to verify timestamp for nvlddmkm.sys
*** ERROR: Module load completed but symbols could not be loaded for nvlddmkm.sys

 

[Apples555]

All updates are installed.

I have tried clean installing many versions of the Nvidia drivers, from the 2010 265 driver all the way to the 327 driver from 2013. Drivers after that result in corrupt graphics. The GPU is a 320m. All of them cause BSODs. Sometimes the sound loops, the screen goes black, but then everything recovers and Windows says "nvlddmkm.sys has crashed but recovered.'

I also disabled the Apple HFS driver, which is a known causer of BSODs.

Bizarre.

 

[axe0]

Please test your GPU, so we can rule out the hardware part
Please note the temperatures of your GPU when idle and under load and post them.

Diagnostic test

GPU TEST

Run Furmark to test your GPU. Furmark tutorial

   Note

Run Furmark for around 30 minutes

   Warning

Your GPU temperatures will rise quickly while Furmark is running. Keep a keen eye on them and abort the test if temperatures rise too high

 

[Apples555]

Just in case, I used DDU to remove all Nvidia drivers and installed the 327.23 drivers from 2013 again.

I ran Furmark for around 5-6 minutes. GPU temperature sharply rose from 71 to 92 degrees C, and stabilized at 91 degrees C. No crashes.

I should mention the crashes never happened under stress. They were always at some sort of idle, usually with a Youtube video playing.

I should add MemTest86, Apple Hardware Test, and Microsoft RAM Check all checked out fine.

 

[Apples555]

It just happened again! Dump attached. Word was open as well as Firefox, which was playing a YouTube video.View attachment 66215

 

[axe0]

Run the DM Log Collector, only a dump is not enough data.

 

[Apples555]

The logs are attached.

View attachment 66241

 

[axe0]

If you have MSI Afterburner installed for measuring the temperatures of your GPU, that is fine, but please remove MSI Afterburner as it is known for causing BSOD's.

Please check your manufacturer for BIOS updates.

Please check Windows Update for updates, install every update available, it may take a couple of times to get them all.
As the crash is caused by the Nvidia Video driver, I suspect

• a bug in Windows which could be solved with the updates from Windows Update
• Nvidia software causing issues
• a driver bug

 

[Apples555]

All Windows and BIOS updates are installed.

I have tried all the Nvidia drivers from when the notebook was made (2010) to now. All of them crash in the same manner.

Afterburner was added only to downclock the GPU, which made it somewhat more stable (it was able to recover instead of BSOD). Regardless, it is now removed. Where else should I look?

 

[axe0]

I ran Furmark for around 5-6 minutes. GPU temperature sharply rose from 71 to 92 degrees C, and stabilized at 91 degrees C. No crashes.

Just noticed this, for the safety of your hardware it is recommended to abort the test when the temperatures are reaching 80-85C.

For what purpose is your macbook mainly used?

 

[Apples555]

It reaches 80 degrees in about 2 minutes.

MacBook is used lightly, mostly for research, light art work, and rarely gaming.

 

[axe0]

That could mean cooling issues, what would explain why the graphics drivers keep showing up.

 

[Apples555]

The BSODs occur exclusively under little load. In fact, it has never happened when the computer is under load and the temperatures are high.

I got new RAM today just in case, which didn't solve the problem. How bizarre!

I'm starting to think it might be the new SSD, because the latest BSOD faulted fs_rec.sys (even though the first such BSOD occurred a year ago with the hard drive I had then).

I'm considering putting in a temporary HD, installing Windows clean, and seeing what happens.

 

[axe0]

Doesn't rule out that the GPU could be damaged.

Why would you think the memory would solve it? I haven't stated anything related to the memory.
The fs_rec.sys is the file system recognizer driver, a driver from Windows and drivers from Windows are rarely the real cause.
If you think your SSD is causing issues, you could try updating the firmware.

 

[Apples555]

It's possible, the VRAM is shared with the regular RAM. I also wanted more.

If the GPU is damaged, problems would appear under OS X as well. The problems only occur under Windows 8.1.

All firmware is updated, confirmed by Samsung's software.

 

[axe0]

Something I forgot, you have a macbook, what rules immediately out the hardware.

Is there any setting enabled that could push the GPU/CPU?
Something like multi-rendering.

 

[Apples555]

The more expensive MacBooks had those, but my 2010 13" only has the 320m.

While cruising around the settings, I turned off Nvidia's power management. It's been 2 days and the system has been stable.

I'll report back if it crashes or stutters again.

On a side note, is my system overheating? At 80 C, the BIOS doesn't even make the fans speed up to their maximum.

 

[axe0]

80C is overheating if your cooling system is not functioning properly yes.
In fact, if the cooling system is not functioning properly, lower degrees is also overheating.

 

[Apples555]

Dang it, it just happened again!

111815-6562-01.dmp 11/18/2015 8:54:05 PM 0x00000133 00000000`00000000 00000000`00000501 00000000`00000500 00000000`00000000 hal.dll hal.dll+539a x64 ntoskrnl.exe+14e8a0 C:\Windows\Minidump\111815-6562-01.dmp 2 15 9600 296,536 11/18/2015 8:54:57 PM

I don't know what to think. Perhaps it is the SSD?