Showing results for 
Search instead for 
Did you mean: 

Computer shutting off during high GPU load

Level 7
I already have another thread on this here, but I am not able to find a way to change the title to be more relevant.!

Long story short: My PC Hard powers itself off when under high GPU load. The Temperatures are fine with CPU around 66c, and GPU at 71c MAX No logs occur during this process. No DUMP files, No nothing.
I can get the PC to crash via a couple means. One is to run FurMark Burnin test with 8x AntiAliasing, Another way is while playing PlayerUnknown Battlegrounds.
The same PC was stable with a EVGA GTX 760 SC running both of these tests.

Things I have tried:

Swapped GPU: MSI Armor 1080 TI FOR Asus ROG Strix 1080 TI OC -- Both card resulted in crashing.
Swapped PSU: Swapped Corsair RM750X out for Corsair RM1000I PSU.
Swapped Modular Cables on PSU to different slots.
Used different Modular cables from the PSU to each 8 pin connector on GPU.
Used separate PSU to power MOBO and GPU
Swapped: ROG Maximus IX Code for MSI Z270-A PRO (using i7 7700k in each Mobo)
Swapped: i7 7700k for Celeron G3930. (First tested the Celeron in the Maximus Code, Then Pulled the Code, Reinstalled the MSI board and tested again)
Removed: All components from case, in case there was a short or something.
Assembled all components on my desk with only BARE ESSENTIALS
Installed new SSD, Crucial M550 256GB SSD with fresh copy of Windows 10 Pro. Fresh Drivers from Nvidia site after removing the VGA drivers that were automatically installed by Windows via DDU version, Installed Nvidia display driver version 384.76

Everything has resulted in crashing.

TESTED: Swapped out ASUS ROG STRIX 1080 TI OC for MSI ARMOR RX 480 OC Edition. Running Furmark, Will report back in a couple hours. - Failed
TESTED: Installed ASUS ROG STRIX 1080 TI OC in another machine and running FurMark on it.- Failed
TESTED: Swapping Corsair RM1000I with SeaSonic X750 PSU - Failed
TESTED: Swapping GSkill Trizent Z RGB 3200 RAM with Gskill RipjawsX 3000 RAM - Failed
TESTED: Removed Corsair RM1000I, Corsair H100i v2 and Switched DIMM Slots. Ran for 1 hour. (best result yet)
Replaced PSU with Seasonic X-750 and Replaced Corsair H100i v2 with a copper Zalman heatsink.
TESTED: Removed Zallman heatsink and Seasonic X-750 and Reinstalled original components. - Crashes again within 10 minutes.
TESTED: Replaced Corsair RM1000I with Corsair H100i v2 using Liquid Metal (Coolaboratory Liquid Pro) TIM

TESTING: I put the 1080 TI in another pc (MSI z270-A Pro, with a Celeron G3930 CPU and Corsair RM750X PSU, WD Velociraptor drive, W10) over night and ran FurMark on it. It lasted 17 hours with no crashes. Currently Adding pieces of my build to this setup to see when crashes start to occur. This should theoretically narrow down the bad egg. So far I have my ASUS ROG STRIX 1080 TI and GSkill Trident Z RGB 3200MHz RAM installed. I will run a FurMark test after every piece I install.
EDIT: GPU ran FurMark 17+ hours, = no crash Added GSkill Trident Z RGB 3200MHz RAM, Ran for 1+ hour = no crash Added i7-7700k = no crash
Added Corsair RM1000i = Testing NOW

MOBO: MSI Z270-A Pro
CPU: i7-7700K No OC
RAM: 16GB GSkill Trident Z RGB 3200MHz (DIMM A1 B1)
HDD: 1x Western Digital Velociraptor
PSU: Corsair RM1000I 1000W
Cooler: OEM Intel Cooler

So if anybody has any solutions/Suggestions/Advice, It would be greatly appreciated. This Custom build has been giving me issues for almost 5 months now and I would just like to get back in the game.

Level 7 pretty sure it was this, I'll have to double check when I get home

Level 7
OK, I am currently trying the other DIMM slots on my mobo. Figure it's a long shot and doesn't make any sense. But if it works I don't care. I will try these drivers after. Thanks again for your input, I greatly appreciate it!