BSOD - Crashing way too frequently in the past few weeks...

PiCASSiMO · 09 Jul 2016

philc43 said:

Please set the resolution to your default screen resolution. This will be shown on the Advanced Display settings

No problem. 3840x2160 (4K). Should the test run in window or full screen? I'm assuming window so that I can monitor the CPU temperatures via CPUID HWMonitor? Also, for how long should I run this test for?

Thank you again for the quick responses. Much is appreciated.

philc43 · 09 Jul 2016

Yes run windowed and then you can see the HW Monitor. There is a temperature trace that you can turn on in the Furmark test as well. I've suggested 5 min but it depends on the temperature stability really, if temperatures go above 80-85C it is best to stop the test.

PiCASSiMO · 09 Jul 2016

Got around 10-minutes and temperatures hit a max about 41-C (105-F). Computer is still stable after 3.5-hours of use this morning. See attached.

philc43 · 09 Jul 2016

As your temperatures are quite low and stable you can run for a lot longer if you wish. It is strange that you have not had any BSOD considering they happened every few minutes before.

PiCASSiMO · 09 Jul 2016

Funny you should mention the BSOD. After I closed everything, the BSOD showed up again.

I got the recent DUMP file reviewed and it's the same old:

FAILURE_BUCKET_ID: 0x124_GenuineIntel_PROCESSOR_MAE
BUCKET_ID: 0x124_GenuineIntel_PROCESSOR_MAE
PRIMARY_PROBLEM_CLASS: 0x124_GenuineIntel_PROCESSOR_MAE
FAILURE_ID_HASH_STRING: km:0x124_genuineintel_processor_mae

Could this be related to the i7-6700k that can't display in the native 3840x2160 resolution? Should I install the GTX-980 card in my computer and validate that it's not the dedicated GPU issue, rather CPU only?

philc43 · 09 Jul 2016

I'm not sure that fitting the new graphics card will help, I think it is a CPU core fault that gets corrected most of the time according to your event logs but not always - hence the BSOD.

Can you check that all the voltages are running stable for the CPU? I assume you are not overclocking in any way? Make sure that any turbo boost settings are off for the purposes of testing.

axe0 · 09 Jul 2016

Hi PiCASSiMO,

Welcome to the 10forums :)

Please boot in the BIOS and let us know what the 3.3v, 5v and 12v values are.

I have analyzed the dumps, unfortunately they are of no use because the reported error by the CPU is unknown.
Attempting to dump the stack reveals also nothing.

Code:

===============================================================================
Section 2     : x86/x64 MCA
-------------------------------------------------------------------------------
Descriptor    @ ffffe00051b40138
Section       @ ffffe00051b40280
Offset        : 600
Length        : 264
Flags         : 0x00000000
Severity      : Fatal

Error         : Unknown (Proc 2 Bank 0)
  Status      : 0xf200004000010005

Interesting to see, the system up time is almost always a max of 10 minutes, this means that either a service at boot is causing the crashes or we're dealing with a faulty hardware.
The services at boot need a few minutes after Windows has loaded for initializing, but the services are in the POST of the boot process.
We can try a simple diagnostic test to find out whether we're dealing with hardware or a service :)

Code:

System Uptime: 0 days 0:06:34.723
System Uptime: 0 days 0:02:15.750
System Uptime: 0 days 0:02:20.525
System Uptime: 0 days 0:02:29.817
System Uptime: 0 days 0:04:48.045
System Uptime: 0 days 0:02:00.016
System Uptime: 0 days 0:01:36.670
System Uptime: 0 days 0:02:10.295
System Uptime: 0 days 0:10:55.950

Please remove / disable items from the startup in taskmanager
These programs could interfere with the boot process giving you trouble and increase the boot time.

Diagnostics Test

Clean boot

Please try the following:

Open taskmanager
- Go to the tab startup
- disable everything
Search msconfig and open it
- Go to the tab services
- Check 'hide microsoft services'
- Uncheck all other services
- Click apply and click ok
Reboot if asked

See if the problem persist, if the problem does not persist enable in the startup tab in taskmanager 1 item and in the services tab in msconfig also 1 service, reboot and see if the problem persist. Keep doing this until you find the culprit or there is nothing left.
If the problem still persist, try booting in safe mode and see if the problem still persist.

Please note, the GPU is NOT loaded in safe mode, if the CPU itself is not faulty the GPU part could be.

PiCASSiMO · 09 Jul 2016

philc43 said:

I'm not sure that fitting the new graphics card will help, I think it is a CPU core fault that gets corrected most of the time according to your event logs but not always - hence the BSOD.

Can you check that all the voltages are running stable for the CPU? I assume you are not overclocking in any way? Make sure that any turbo boost settings are off for the purposes of testing.

As you wrote this, I had installed the GTX 980. And you were correct, the BSOD came one shortly after installation. Unfortunately for me, I've got a 980 that I'm doing an RMA with. One of the fans is unbalanced and causing contact with the plastic frame, resulting in very annoying and loud noise.

PiCASSiMO · 09 Jul 2016

axe0 said:

Hi PiCASSiMO,

Welcome to the 10forums :)

Please boot in the BIOS and let us know what the 3.3v, 5v and 12v values are.

Attached is the screen that you are probably requesting... in addition, I've taken several more in the attached ZIP file.

+ 3.3v = 3.383 V
+5v = 5.100 V
+ 12V = 12.240 V

Thank you.

philc43 · 09 Jul 2016

Please follow the clean boot instructions as detailed by @axe0

It is not necessarily the CPU that is faulty, the CPU detects and reports the faults but they could and most likely are caused by something else.