Question PC shuts down instantly and restarts while playing Fortnite (and other games)

Status
Not open for further replies.
Nov 2, 2019
2
0
10
Situation
When playing Fortnite for about 10 minutes or less, my PC will shut off instantly (without a BSOD) and restart itself. No other games are causing this issue. I reviewed the Event Viewer and the System log shows that a Kernel-Power, Event ID: 41, Task Category: (63) event has occurred. I have also disabled "Automatically restart" within Startup and Recovery in the Control Panel but no BSOD shows up.


PC Specifications
I built the PC myself about 2.5 years ago with the following components. I am running Windows 10.

CPU: i5-6500 3.2 GHz Quad Core
MOBO: Gigabyte GA-B150M-D3H-CF
GPU: MSI Radeon RX 480 (8GB)
PSU: EVGA SuperNova G2 550W 80+
RAM: Corsair Vengeance LPX (2x8GB) DDR4-2400
SSD: Samsung 850 EVO-Series (250GB)
Stock coolers only.


Attempted Solutions/Tests (in order)

(1) Failing PSU: Initially I thought the PSU was failing based on other threads from here, Reddit and other websites. I was able to use my EVGA warranty and obtained another PSU (of the exact same model) but the restarts continue.


(2) Resetting my PC: I uninstalled Fortnite and GPU drivers prior to wiping my computer which did not solve my problem. I did a reset of my PC (keep my files) and this actually stopped the crashing and restarts for about two weeks. I had no issues playing Fortnite until yesterday, when the restarts began again.


(3) Updating the BIOS: I went to the Gigabyte website and found the latest BIOS firmware for my MOBO. I updated from F16 to F24. No success with this either.


(4) Stress testing (and HW Monitor):

I downloaded Primer95 and Furmark to stress test my PC. I ran the Prime95 Smallest FFT Torture Test for ~8.5 hours and the CPU temperatures did not exceed 50 Celsius and the PC did not crash or restart overnight.

In the morning, I ran the Smallest FFT together with Furmark for 30 minutes, which induces a greater demand over a long period of time than Fortnite would (before crashing which again is ~10 minutes). The maximum GPU temperature was 75 Celsius and the CPU again remained under 50 Celsius throughout the 30 minute test. Surprisingly my PC did not crash and restart under the very high demands of the tests.

I ran the Windows Memory Diagnostic Tool and it shows no problems with my RAM. I feel that my issue is not likely to be caused by RAM.

HW Monitor is running while Fortnite is running, and it does not show any abnormally high temperatures and shows that all the fans are spinning. It also shows 100% CPU utilization and all 4 cores running at maximum.


(5) Miscellaneous: I cleaned out the dust in my PC using compressed air. I also moved my PC to a more open location and plugged it into a different outlet. No luck with any of these.


Closing
Any and all advice and suggestions are welcome. I am opening to trying any new solutions. I am pretty frustrated with this issue and I am not sure where to go from here. I am open to buying new PC components but would like to narrow down the cause and prefer not to upgrade everything. Thanks.


EDIT: I tried running two other games: Arkham Knight and The Witcher 3 (two older games that I can run on high graphics with steady FPS) and the crash and restart now occur when playing them too.
 
Last edited:

InvalidError

Titan
Moderator
Running everything at constant heavy load is easier on the various VRMs than dealing with the large transients that occur in most normal workloads and games in particular. Your problem here may be an unfortunate combination of transient responses between the CPU VRMs, GPU VRMs and PSU - every component may be technically fine individually but aren't playing nice with each other.

One thing you may be able to try without having to change any hardware is change your power management profile to Performance to raise the minimum power state of everything so the PSU and various VRMs have smaller min-max transients to cope with.
 
Situation
When playing Fortnite for about 10 minutes or less, my PC will shut off instantly (without a BSOD) and restart itself. No other games are causing this issue. I reviewed the Event Viewer and the System log shows that a Kernel-Power, Event ID: 41, Task Category: (63) event has occurred. I have also disabled "Automatically restart" within Startup and Recovery in the Control Panel but no BSOD shows up.


PC Specifications
I built the PC myself about 2.5 years ago with the following components. I am running Windows 10.

CPU: i5-6500 3.2 GHz Quad Core
MOBO: Gigabyte GA-B150M-D3H-CF
GPU: MSI Radeon RX 480 (8GB)
PSU: EVGA SuperNova G2 550W 80+
RAM: Corsair Vengeance LPX (2x8GB) DDR4-2400
SSD: Samsung 850 EVO-Series (250GB)
Stock coolers only.


Attempted Solutions/Tests (in order)

(1) Failing PSU: Initially I thought the PSU was failing based on other threads from here, Reddit and other websites. I was able to use my EVGA warranty and obtained another PSU (of the exact same model) but the restarts continue.


(2) Resetting my PC: I uninstalled Fortnite and GPU drivers prior to wiping my computer which did not solve my problem. I did a reset of my PC (keep my files) and this actually stopped the crashing and restarts for about two weeks. I had no issues playing Fortnite until yesterday, when the restarts began again.


(3) Updating the BIOS: I went to the Gigabyte website and found the latest BIOS firmware for my MOBO. I updated from F16 to F24. No success with this either.


(4) Stress testing (and HW Monitor):

I downloaded Primer95 and Furmark to stress test my PC. I ran the Prime95 Smallest FFT Torture Test for ~8.5 hours and the CPU temperatures did not exceed 50 Celsius and the PC did not crash or restart overnight.

In the morning, I ran the Smallest FFT together with Furmark for 30 minutes, which induces a greater demand over a long period of time than Fortnite would (before crashing which again is ~10 minutes). The maximum GPU temperature was 75 Celsius and the CPU again remained under 50 Celsius throughout the 30 minute test. Surprisingly my PC did not crash and restart under the very high demands of the tests.

I ran the Windows Memory Diagnostic Tool and it shows no problems with my RAM. I feel that my issue is not likely to be caused by RAM.

HW Monitor is running while Fortnite is running, and it does not show any abnormally high temperatures and shows that all the fans are spinning. It also shows 100% CPU utilization and all 4 cores running at maximum.


(5) Miscellaneous: I cleaned out the dust in my PC using compressed air. I also moved my PC to a more open location and plugged it into a different outlet. No luck with any of these.


Closing
Any and all advice and suggestions are welcome. I am opening to trying any new solutions. I am pretty frustrated with this issue and I am not sure where to go from here. I am open to buying new PC components but would like to narrow down the cause and prefer not to upgrade everything. Thanks.

Override fan profile on gpu and run it 100%. 10:1 graphics card going bad. Also try down clocking by 200MHz the gpu. Let us know if it is more stable after that.
 
Nov 2, 2019
2
0
10
Running everything at constant heavy load is easier on the various VRMs than dealing with the large transients that occur in most normal workloads and games in particular. Your problem here may be an unfortunate combination of transient responses between the CPU VRMs, GPU VRMs and PSU - every component may be technically fine individually but aren't playing nice with each other.

One thing you may be able to try without having to change any hardware is change your power management profile to Performance to raise the minimum power state of everything so the PSU and various VRMs have smaller min-max transients to cope with.


I modified the power platform Balanced to High Performance, which unfortunately did not resolve the restart issue.


Override fan profile on gpu and run it 100%. 10:1 graphics card going bad. Also try down clocking by 200MHz the gpu. Let us know if it is more stable after that.


I downloaded MSI Afterburner and set the fans to 100% and also downclocked the GPU, this did not prove successful either
 
Nov 5, 2019
2
0
10
Hi, I've had issues like this before. PSU does sound like the issue but hey if you checked this then you've checked it! If you have multiple sticks of RAM try pulling all out but one and seeing if the issue persists (and then try different sticks one at a time if it persists) give your PC a clean maybe. its 2.5 years old maybe a bit of love wont go amiss (especially if you smoke lol) even if its not the cause. try CPU-Z to monitor the temps/ voltages. As digitalgriffin says above try setting your Vdroop to another setting and see if that helps. Give all the plugs on the mobo and GPU a wiggle (stuff can become loose sometimes). Grasping at straws maybe usb mouse or keyboard could be issue try swapping them out if you got some old kit lying around (especially if you've spilt your beer on them). I had a GPU once that crashed at stock speeds you could try underclocking your core and mem speeds to see if that helps with EVGA precision X or the MSI equivalent, afterburner I think
 
Last edited:
Nov 5, 2019
2
0
10
sorry, rereading your initial post to see if I've missed anything and I see you already cleaned it already. I would also try upping your ram voltage slightly and after that, lowering their speed to 2133 . Also try removing graphics card and giving the pci-e contacts a clean might help (de-static yourself so you don't fry the card). You could try a prog called "whocrashed" and that might give you some ideas. Last idea ...try downloading the latest intel B150 chipset drivers, and try disabling your integrated graphics "Intel® HD Graphics 530 " in your device manager.
 
Last edited:
Status
Not open for further replies.