HELP! Continual full crashes when playing PUBG on my system. I don't know what's wrong, not a PC expert at all...

[technobolt15]

Operating System

Windows 10

Computer Specs (PSU, GPU, CPU, RAM, Motherboard)

Intel Core i5 6500 3.2GHz QuadCoreCPU
MSI ATX B150 Gaming M3DDR4 Motherboard
DDR4 8GB(2x4G) G.Skill 2400Mhz Ripjaws V RAMKit
120GBKingston 2.5" SSDNow UV400 SATA 6Gb/s SSDDrive
1TBWD3.5" 7200rpmSATA 6Gb/s Caviar BlueHDD
ASUS RX480 8GB PCIe VideoCard ( and now GTX 1070, both of which having the issue)
Cougar ATXSolution 2Case Black (No PSU)
700 Watt Cougar CMX700 80 Modular Bronze Power Supply
ASUS24x DRW-24D5MTDVDWriter OEMAsus PCE-N15 Wireless N300 PCIeNetworkAdapter
120mmCorsair "Air Series" AF120 WHITELEDQuiet Case Fan PNCO-9050015-WLED


Speccy Link
http://speccy.piriform.com/results/MB7ilZL2EINtGpeCPJO6C55

Description of problem:
Hey guys, I'm really having long term issues with my PC. It will flat out crash then reboot when playing GPU intensive games. PUBG mainly. I've checked windows logs and I'm getting Kernel Power 41 errors. Underclocking my gpu to -30% frequency and 50% power allowance in AMD Wattman seemingly fixed the issue however it began erroring again at this state. I turned GPU frequency down to -50% and it allowed for me to run PUBG, albeit not at 60FPS (RIP the dream).I replaced the AMD card with a GTX 1070 as I wanted an upgrade anyway, to no avail, and I am still having the same issue, even more so with this graphics card running at a higher core clock and memory speed.Another interesting point is that playing in fullscreen with a decreased resolution scale does not crash instantly, and it is inconsistent (sometimes it will crash, sometimes it won't). Windowed borderless is unplyable, and everything looks terrible at 1366x768.It is flat out unplayable at a high GPU frequency, simply because the PC will reboot.


When this issue began:

July 2017

Recurring issue:
Yes

Date of purchase:
June 2016 I purchased and built my PC, no issues up until now.

Under Warranty:
Yes

Cause/Steps to recreate the issue:

Running PUBG at a normal- non underclocked Core and Memory clock on the graphics card. It is very frustrating.

What I've tried so far to resolve the issue:

Decreasing core and memory clocks speed seemed to help, although this is unachievable with the new 1070. Decreasing screen scale also did, I don't know how resolution plays into this.Temperature is not an issue, have monitored these.Tried a different GPU - no dice. I've also tested RAM with memtest86 and tested my HDD (not my SSD though)

 

[technobolt15]

Also here is a HWInfo test I ran when starting up PUBG.

https://docs.google.com/spreadsheets/d/1iqa47SkEvRRgLAeToTPkFOChsb1NivtyGUQmivfantc/edit?usp=sharing

 

[BrandonYoung]

If you can eliminate the OS as a candidate, I would suspect the PSU just by the sounds of the issue.

You can get (or use) a bootable USB Linux distro to check for OS issues, not sure if you'll have luck reproducing heavy GPU use however...

 

[technobolt15]

Could I somehow fresh install windows OS?

I'm not sure how to do this Linux stuff

 

[BrandonYoung]

Before you start that path, have you attempted increasing CPU voltage slightly?

 

[technobolt15]

How do I do that safely haha?
And how will this help? Do you suspect the CPU has issues?

 

[BrandonYoung]

I would reboot the PC, enter your BIOS, and increase your CPU Vcore voltage the smallest increment possible and note if it helps/hurts.

You have updated your Nvidia GPU drivers and uninstalled your AMD GPU drivers right?

My thinking:
If its a PSU issue, (in theory) if you increase the power draw to the PSU, it will have more issues.
If its a CPU issue, (in theory) increasing the voltage slightly could help stabilize things.
If its an OS/Driver issue, updating drivers/fresh install of Windows could help.

EDIT:
It may not be a bad idea to test your SSD, but if this were the issue, I would expect to see issues outside of games as well.

EDIT2:
You mentioned earlier that temperatures were not an issue, this included CPU temperature?

 

[technobolt15]

I would reboot the PC, enter your BIOS, and increase your CPU Vcore voltage the smallest increment possible and note if it helps/hurts.

You have updated your Nvidia GPU drivers and uninstalled your AMD GPU drivers right?

My thinking:
If its a PSU issue, (in theory) if you increase the power draw to the PSU, it will have more issues.
If its a CPU issue, (in theory) increasing the voltage slightly could help stabilize things.
If its an OS/Driver issue, updating drivers/fresh install of Windows could help.

EDIT:
It may not be a bad idea to test your SSD, but if this were the issue, I would expect to see issues outside of games as well.

EDIT2:
You mentioned earlier that temperatures were not an issue, this included CPU temperature?

Right, yeah I've reinstalled OS and drivers to no help at all. It is just PUBG too, no other games as of yet. I've uninstalled and reinstalled that twice completely also. I'll try the CPU thing, also CPU temps are very good.

 

[neatfeatguy]

You seem to have done everything except troubleshoot your power issue - which is what your system is telling you if you're getting Kernel Power 41 error about the sudden power disruptions when your computer shuts down since you've ruled out temps.

So if things aren't overheating for the GPU or CPU, then it comes down to a possible overheating issue with the PSU or simply a fact you have a bad power issue. Bad power could be a couple of things: it could be the PSU itself or it could be bad power at the location you have the computer plugged in at.

If you don't have another PSU to test with you can at least get your computer plugged into a completely different outlet. If that means moving your computer to another room and testing it out, so be it.

Constant bad power issues can cause other sensitive equipment in your computer to become damaged. Get this sorted out as soon as possible or you can run the risk of causing something else to get damaged in your computer.