BSOD Problems. Having Trouble Pinpointing the Cause.

[Scurra]

Since building my new rig I've been having intermittent BSOD issues, last night however it got to the point where I can't even get the system to boot into Windows 10 without restarting or bluescreening moments after loading into windows. (Interestingly I also can't get it to boot from a USB to do a fresh install/windows repair).

The trouble is the bluescreen errors I've been getting never seem to be the same thing. I've been getting 'Memory Management' errors 'Kernel Security Check Failure' 'IRQL not less or equal' and 'IRQL gt zero at systems service' issues to name a few.

Because I don't seem to be getting a consistent errors I've been having real trouble nailing down the cause of the problem. The PC spec is as follows:

Intel i7 5820k
Gigabyte X99-Gaming 5P
Corsair Vengeance 16GB (4x4GB) DDR4 3000MHz Quad Channel
Samsung 950 Pro 256GB
Crucial MX100 512GB
SuperfFlower Leadex Gold 850W
Corsair H110i
2x Gigabyte GTX 770 Windforce in SLI

I did have a mild overclock on the i7 at 4Ghz and the RAM running its XMP but disabling that doesn't help.

I'm at work currently but could really use some any advice anyone can throw my way in diagnosing what could be causing this before I waste another evening trying to sort it.

 

[Hagar]

Perhaps tweaking the voltages for the CPU may identify or cure the IRQ errors.

Run memtest from boot to check the memory.

 

[Scurra]

Perhaps tweaking the voltages for the CPU may identify or cure the IRQ errors.

Run memtest from boot to check the memory.

I'm not sure the system is stable enough to get through a memtest, half the time I seem to be stuck in a boot loop at the moment. I'll give it a shot though as it would be good to rule out something for sure.

At stock clocks shouldn't the default CPU voltage be fine? My first thought was unstable overclock but like I said in the OP dropping the BIOS back to optimised defaults doesn't seem to make any difference.

 

[V4NT0M]

At stock clocks shouldn't the default CPU voltage be fine? My first thought was unstable overclock but like I said in the OP dropping the BIOS back to optimised defaults doesn't seem to make any difference.

Did you reset the bios to default or just change the clocks/voltage back?

I think it's worth a try putting a little more voltage through it even at stock and double check that nothing is at the wrong setting.

Memory can be annoying to check, try with one stick and then another and then in different ports, if all else fails see if someone can lend you a stick to try.

Definitely sounds like memory or CPU.

 

[Scurra]

Did you reset the bios to default or just change the clocks/voltage back?

I think it's worth a try putting a little more voltage through it even at stock and double check that nothing is at the wrong setting.

Memory can be annoying to check, try with one stick and then another and then in different ports, if all else fails see if someone can lend you a stick to try.

Definitely sounds like memory or CPU.

I did a full BIOS reset so everything should have gone back to factory settings.

Thanks for the suggestions I'll give them a go tonight.

 

[Scurra]

Well memtest is not going well.

Spoiler

Would this many errors indicate something more wrong than a duff module? Is there anything else I can do to check this?

 

[ED209]

Well memtest is not going well.

Spoiler

Would this many errors indicate something more wrong than a duff module? Is there anything else I can do to check this?

its just testing the ram, if you have cleared the cmos back to factory settings its going to be faulty so RMA it

 

[Scurra]

It certainly seems that way.

I've had limited success with upping the voltage. I can now make it back to the windows 10 login screen I put in my PIN and windows loads and then immediately BSDs with memory_management errors again.

I've opened an RMA request with OCUK

No gaming for me for a while.