Stability testing recommendations.

[BasicCable]

What stability test should I take?

A little bit of background

X99 Msi gaming 7
5820k @4.0GHz with an h100i
970 Gtx
32GB DDR4 2666 (8x4gb)
Corsair 860ax

System rebooted yesterday morning. Event ID 41 I believe. Kernel power loss or something around those lines. No BSOD, no .dmp file. System had an idle vm and a YouTube video playing at the time with only browsers and idle programs opened.

The PSU is brand new. 2 weeks old. Switched after a Evga p2 850w gave me similar issues but much worse in frequency and severity. OC has been stable otherwise under an older PSU and when the Evga didn't go on a failing spree.

I've stressed the system with multiple stability tests today.

Asus realbench for 3hrs. (32 Gb tested)
Furmark GPU and 90% CPU load (10 threads) for 90 mins.
OCCT Cpu test 1hr.
Intel Burn Test for 40 mins.

Temps range from high 50s to mid 60s on the CPU side.
GPU wise mid 70s.

I'm at a complete loss. So any direction and advice wiuld be appreciated.

 

[n0ns3ns3]

this might be related to MB.
like the overload on power delivery that is either overheating or faulty.
Since you are using liquid, there is no direct airflow on MB MOSFET. might be the issue.

 

[BasicCable]

this might be related to MB.
like the overload on power delivery that is either overheating or faulty.
Since you are using liquid, there is no direct airflow on MB MOSFET. might be the issue.

Im assuming the power delivery is under the heatsink directly above the Socket? If so, I just touched it and it was very warm. Almost hot in fact. It is in the middle of stress test right now though. The case is a corsair 760t and the H100i is in push config exhausting on thentop of the case.

Anything I can do to cool the thing further?

 

[n0ns3ns3]

The components in the power circle are usually rated for over 100C - you can not touch that without getting burned
850w quality PSU (and AX is good one) is more than enough for your system.
So I'd suspect that the issue is with MB.
First thing to do, would be resting the BIOS and running CPU at stock speed. If the problem does not occur, start to raise the clocks. Be careful with voltage.
Do not use "adaptive" for the stress testing.
P.S.
I guess you already updated the MB BIOS to the latest available.

 

[BasicCable]

The components in the power circle are usually rated for over 100C - you can not touch that without getting burned
850w quality PSU (and AX is good one) is more than enough for your system.
So I'd suspect that the issue is with MB.
First thing to do, would be resting the BIOS and running CPU at stock speed. If the problem does not occur, start to raise the clocks. Be careful with voltage.
Do not use "adaptive" for the stress testing.
P.S.
I guess you already updated the MB BIOS to the latest available.

100C? That would make sense.
Honestly I havent been able to replicate the issue.This is the first reboot in a month with 3 weeks on a 750w ax, and 2 weeks on a 860w ax. On the faulty Evga 850w p2 (2 of them) it was an easy process but I'm 10 hours into pounding the set up with stress and I have nothing. That's why I was looking into more intensive stability testing.


I'll update/config Bios (I think its just a broadwell-e compatibility update though) after this current 2hr run of testing and also look into a memtest overnight.
Running a 1.165v @ 4.0GHz. God, I just hope its not the PSU again.

 

[n0ns3ns3]

3 faulty top tier PSU in a row sounds not probable.