New System Crash Under Moderate-Heavy Load

[Unimaginative Name]

I recently built a new rig:

8700k

32gb G.Skill DDR4 3200 MHz

ASRock Fatal1ty Z370 Gaming-ITX/ac

ZOTAC GeForce GTX 1080 Ti Mini 11GB

Cooler Master V750 PSU

Cooler Master Dark Rock 4

 

Old Parts:

2TB HDD

256GB WD Black SSD

 

I can browse the internet/stream/watch Youtube videos, do CAD work 'til the cows come home, but anything any more GPU intensive causes either a BSOD, an instant restart, or a freeze then a restart after 10-30 minutes. Temps on the GPU are fine (~71 C under load). Temps on the CPU are also fine. Really just a crap-shoot which one I get. I've cleared the BIOS, updated my BIOS, reseated my GPU, replugged my PSU, changed PSU inputs, I've also run memtest, Windows Memory Diagnostic, and I've tried the System File Checker Tool. All to no avail.

 

My most recent test was to run Furmark. On the couple of times that I've tried this, the system can usually run Furmark fine for 30-40 minutes. I usually then turn the test off because everything seems fine. On this occasion, I then turned on my CPU stress test while Furmark was running. The system crashed after about 10 minutes.  This most accurately represents my real-world use case during crashes, CPU and GPU under load. This makes me believe that this is a PSU power issue.

 

Any ideas?

[Firewrath9]

any OC? the v750 is pretty good.

[Unimaginative Name]

Just now, Firewrath9 said:

any OC? the v750 is pretty good.

I initially had a CPU OC, and the system was stable while doing a CPU stress test. I also had my RAM at the 3200 MHz (DIMM rated speed) and not the motherboard defaults. All seems fine until I tried to game for any longer than 10 or 20 minutes. Currently, everything is at stock speeds, the system still crashes.

[manikyath]

may be worth contacting the store you bought the psu from, and voice your concern over a potential dud power supply.

[ZacoAttaco]

18 minutes ago, Unimaginative Name said:

This makes me believe that this is a PSU power issue.

7 minutes ago, manikyath said:

may be worth contacting the store you bought the psu from, and voice your concern over a potential dud power supply.

So essentially the system crashes when under reasonable to heavy load and it's not exclusive to just the GPU or CPU? Then that would indicate it's most likely a power delivery issue like you suspected.

 

 

[TheGlenlivet]

What does the BSOD tell you?  Any codes?

[Unimaginative Name]

8 minutes ago, TheGlenlivet said:

What does the BSOD tell you?  Any codes?

The BSOD give any number of 2 or 3 different codes. But none of the blue screens ever run to completion or make it past 0% after 10 or 15 minutes. I just tried only a CPU stress test again (1st since CMOS clear) and it crashed after 2 or 3 minutes. I went into my BIOS and found an overheating shutdown feature and disabled it. I'm currently running the tests again and currently with no issue (yet). My thought it that when overclocking is removed from Auto, that feature gets disabled. Unconfirmed though. I'm 20 minutes, I'll add furmark to this test and see what happens. If this is the problem, I'll be relieved, but still pretty irritated. 

[Unimaginative Name]

24 minutes ago, ZacoAttaco said:

So essentially the system crashes when under reasonable to heavy load and it's not exclusive to just the GPU or CPU? Then that would indicate it's most likely a power delivery issue like you suspected.

 

 

That's what I initially thought, but it has now crashed under a CPU only test. Turning off thermal protection in the BIOS didn't do anything either. During troubleshooting, I've run Furmark alone multiple times. It has crashed the system a couple of times. So now it has crashed under a CPU only load and a GPU only load, as well as a combined load. It isn't CPU/GPU/MOBO temp related. I may try the tried and true remove 1 memory stick and retry.

[ZacoAttaco]

1 hour ago, Unimaginative Name said:

That's what I initially thought, but it has now crashed under a CPU only test. Turning off thermal protection in the BIOS didn't do anything either. During troubleshooting, I've run Furmark alone multiple times. It has crashed the system a couple of times. So now it has crashed under a CPU only load and a GPU only load, as well as a combined load. It isn't CPU/GPU/MOBO temp related. I may try the tried and true remove 1 memory stick and retry.

So this would probably still indicate, perhaps a PSU related issue. Try messing with the RAM configuration, wouldn't hurt.

[Unimaginative Name]

1 minute ago, ZacoAttaco said:

So this would probably still indicate, perhaps a PSU related issue. Try messing with the RAM configuration, wouldn't hurt.

So I rebuilt the entire psu, except for the CPU (don't have TIM on hand). Everything except the second HDD is on. My CPU stress test and Furnark have been running simultaneously for 30 minutes now without issue. If this goes well for an hour, I'm gonna put the lid back on the case (Node 304). Induce some heat and see where that gets me. If it crashes then, the only thing I can think of would be GPU overheating causing system shutdowns. 

[ZacoAttaco]

7 hours ago, Unimaginative Name said:

So I rebuilt the entire psu, except for the CPU (don't have TIM on hand). Everything except the second HDD is on. My CPU stress test and Furnark have been running simultaneously for 30 minutes now without issue. If this goes well for an hour, I'm gonna put the lid back on the case (Node 304). Induce some heat and see where that gets me. If it crashes then, the only thing I can think of would be GPU overheating causing system shutdowns. 

I'm guessing you meant rebuilt the entire PC? Good to hear that the CPU no longer seems to be causing the system to shutdown. How did the GPU go?

[Unimaginative Name]

12 hours ago, ZacoAttaco said:

I'm guessing you meant rebuilt the entire PC? Good to hear that the CPU no longer seems to be causing the system to shutdown. How did the GPU go?

Yeah, oops. Not touching that PSU. I ran a simultaneous stress test in open air for 45 minutes. I then put the lid back on and it crashed within 5 or 10 minutes. At this point I know it's a PSU overheating causing crashes. I started the RMA process with cooler master. Pray for me.