Temperamental freezes, ongoing issue for 6+months.

KalTorak

Honorable
May 25, 2012
435
0
10,960
Hi All

I have a system that is driving me up the wall and I am out of ideas. Specs are

Mobo - Asus P8Z68-v pro/gen3
CPU - I7-2600k
PSU - EVGA 650w G2 (new)
GFX - Geforce MSI Gaming4G 970GTX
RAM - Corsair Vengance DDR3 2x4Gb 1600Mhz
SSD - Crucial MX100 - 256Gb
OS - Win 10 Pro
Primary Monitor - Asus ROG - PG278Q (Via Displayport)
Secondary - Acer X233H (via DVI-D)


Please note sometimes the system is rock stable with 10 hour gaming sessions and no issues, this has often resulted in a misguided relief that the last attempted fix has resolved the problem. I cant replace the whole damn system and while I have swapped out parts for other known working pieces and not had issues this could be the result of the issues being so randomly occurring.

The problems experienced are the following.


- Crashes when gaming. This can be either while gaming or sometimes loading chrome or a new tab. This can occur if the game is windowed or fullscreen. This happens for multiple titles (from factorio to WH:Total War)
- Screens freeze but chat programs eg discord/skype remain active for 10-15s then die. A single press of the power button sometimes instantly kills the PC, but more often than not it requires a 5s hold.
- Resets (can repeat up to 3 times (ish) including hard resets with 10s poweroff) sometimes result in screens still not displaying, sometimes it freezes on windows loading however in both instances it does load as discord loads back in and I can speak to and hear friends.


Troubleshooting steps taken.


- Temp monitors show both the CPU and GPU remain under all the thresholds and no throttling takes place. Had these running during more than one instance of freezing.
- GFX card re-seated
- GFX card moved to a different PCI-E slot.
- PSU replaced with EVGA G2
- Drivers updated
- Drivers downdated
- Drivers stripped entirely and reinstalled. (Via Guru3d Display Driver Uninstaller
- Win 10 re-installed.
- Memory Checks out OK through all memtests.
- Games installed to seperate SSD. Steam games verified cache.
- Aida stresstest passed. No temp issues
- Furmark stresstest passed. No temp issues


I am honestly open to ALL sane suggestions on what could cause this.
 

Barty1884

Retired Moderator
So, it doesn't appear to be drivers, temperatures or the OS.....
Unfortunately, you may well be dealing with a slight hardware issue, making it really, really tough to detect if you can't replicate the issue on demand.

You've checked your RAM with Memtest, and you've stressed your GPU and replaced the PSU.
That only really leaves the CPU/Motherboard or Storage.
The storage is fairly easy to test, using the manufacturer's software....... check both your SSDs and any HDDs and see how they're holding up.

Beyond that, you're left with the CPU and Motherboard.
CPUs rarely 'die' without being over-volted. They're either DOA, or will work forever (generally).

Inspect the motherboard for any blown (unlikely) or slightly bulging capacitors (possible).
With maybe only one cap having a slight bulge, it would explain why the issue is so inconsistent.

If that's the case, caps can be repaired with the correct knowhow (I believe.... I don't have it! lol). Or you could replace the motherboard .
 

KalTorak

Honorable
May 25, 2012
435
0
10,960
Didnt think about testing the storage, thats a good shout thanks.

The CPU has been overclocked stably since install at 4Ghz with only a small voltage boost, I cant remember the exact voltage increase, but its nothing remotely approaching an overvolt. I remember when I first overclocked that thing, I couldnt believe how easily it went to 4, hell it went much further but I dialed it back.

Ill see if any of the caps are bulging, but I dont think that would cause this kind of infrequent stability, still I am out of ideas so lets see :)

Unfortunately if its the caps, its new mobo time, I dont have the patience for it, and a mistake could fry everything.


Once I have tested the storage and inspected the mobo I will post back with an update, currently at work.