PC randomly reboots while gaming – tried everything, still no fix

[Ambientcore]

If you haven’t read any of my previous threads, I suggest you do.

https://linustechtips.com/topic/1608244-pc-is-possessed-and-reboots-automatically-with-no-way-to-figure-out-the-problem/
https://linustechtips.com/topic/1605904-pc-reboots-while-gaming-–-at-my-wit’s-end-after-countless-fixes/
https://linustechtips.com/topic/1605781-pc-keeps-restarting-while-gaming-–-need-help/
 

I’m really at my wits’ end. I’m so tired—exhausted. I haven’t had my computer go a single week without shutting down automatically during gaming. It doesn’t matter what games I play—Cyberpunk 2077, The Last of Us Part I, Horizon Forbidden West—it always violently reboots and with very harsh glitch effects, with no display right before the screen goes black. Then I hear the Windows welcome tune. My motherboard gets stuck on the white LED. I keep getting a Kernel-41 error. Fucking Hell!

I assume that has something to do with VGA.
 

Here’s what I’ve done so far (though I’m sure I’ve missed a few things):

• Replaced the motherboard twice

• Upgraded to a higher-wattage PSU

• Replaced the GPU twice

• Replaced the RAM twice (QVL listed)

• Got a new CPU from the AMD Singapore warehouse

• Got a new case

• Installed a new CPU cooler

• Got a new SSD for Windows 11

I’ve also tried all of this:

• Set refresh rate to 60Hz

• No errors were shown on my Sandisk disk management software

• Lowered in-game settings

• Enabled high-performance power plan

• Moved the PC to a different location in the house

• Rolled back drivers

• Installed only GPU drivers (no Adrenalin software)

• Skipped Realtek audio drivers (thought they might be interfering)

• Switched to an old HDMI cable (which performed poorly)

• Undervolted the GPU (made things worse)

• Enabled and disabled DOCP

• Enabled V-Sync

• Limited in-game FPS

• Disassembled and reassembled the PC twice

• Swapped out nearly every part one by one—except for the HDMI cable and the monitor

• Tested on a different monitor at the XFX store (still shut down)

• Disabled Game Mode in Windows

• Disabled automatic restart

• Disabled C-State

• Yes, the PSU and to GPU have separate power cables

• BIOS is updated

• Tried to set ErP Ready to Enable S5

• Tried to set PCIE_1 Mode to Gen 3 

• Turned off PBO

• Reduced my voltage by following this video (didn't help)

• Changed my power supply idle control to typical.

• Ran FurMark + OCCT - VRAM 100% - 20 min (no shut downs)

• Disable turn on fast startup in the power options settings

• Sleep options are set to never

• Downgraded to Windows 10

• Disabled deep sleep within my monitor settings

• ReSize bar is off

• Stopped putting my mouse USB in my case USB hub, and plugged it into my motherboard

• Created GPU hardware acceleration in the registry and disabled it by setting the value to "1"

• Ran Linpack Xtreme (10 tests), all was fine - No RAM/CPU errors shown

• Disabled USB selective suspend

• Put a new mouse

• Disable HDCP in the Adrenalin app

• Uninstalled old mouse driver using USBDeview

• Disable ASPM in the BIOS

• USB power delivery is disabled

• AURA (RGB) is off

• Got a quality brand HDMI cable

• PBO Fmax Enhancer is disabled

• All core curve optimizer sign positive

• All core curve optimizer magnitude is set to 5

• Enabled 4G decoding

NOTHING. WORKED.
 

Current Setup:

• CPU: AMD Ryzen 5 5600X

• PSU: EVGA P6 1000W

• GPU: XFX MERC 6900 XT

• Motherboard: ASUS STRIX B550 Wi-Fi II

• Cooler: DeepCool

• RAM: Patriot Viper Steel 32 GB (3200 MHz)

• Monitor: ASUS VA24E (HDMI)

This has been going on since February, and I have broken down twice. I am on the verge of giving up, and I am exhausted. I am tired. 

@KidKid tagging you since you've been here since the beginning.

[Ambientcore]

I am trying other options, too. Nothing seems to work.

 

Currently changed ErP to Enabled
And set the Power Supply Idle Control to Typical

 

Let us see.

[Brennan Price]

I am starting to believe there is some form of driver incompatibility between the B550 chipset (especially on ASUS B550 motherboards) and the RX 6XXX XT series cards currently whereby your graphics driver is not able to recover properly.

 

I have seen this kind of thing happen before on an X79 chipset with GTX 10 series cards and it never got fixed, the issue was apparent under both Windows and Linux regardless of if I was in game or not. I replaced my graphics card to another generation higher and my issue was resolved. Though I struggle to believe that with a common configuration like yours you're the only one seeing this issue, and others would have raised the same. 

 

I suggested on your original post that it could be the chipset drivers causing the issue and I maintain that might still be the case since the tested components are all pretty much the same. Failing that the issue will be driver related elsewhere but most likely AMD or ASUS driver related. 

 

Rather than looking for Kernel error ID 41, look through event viewer for what happened just before that as it will likely tell you your cause/issue. 

[Ambientcore]

14 minutes ago, Brennan Price said:

I am starting to believe there is some form of driver incompatibility between the B550 chipset (especially on ASUS B550 motherboards) and the RX 6XXX XT series cards currently whereby your graphics driver is not able to recover properly.

 

I have seen this kind of thing happen before on an X79 chipset with GTX 10 series cards and it never got fixed, the issue was apparent under both Windows and Linux regardless of if I was in game or not. I replaced my graphics card to another generation higher and my issue was resolved. Though I struggle to believe that with a common configuration like yours you're the only one seeing this issue, and others would have raised the same. 

 

I suggested on your original post that it could be the chipset drivers causing the issue and I maintain that might still be the case since the tested components are all pretty much the same. Failing that the issue will be driver related elsewhere but most likely AMD or ASUS driver related. 

 

Rather than looking for Kernel error ID 41, look through event viewer for what happened just before that as it will likely tell you your cause/issue. 

Is this my monitor? Is it my AMD CPU? Is it my GPU? I am so confused. I don't think it is my monitor. So, something regarding CPU or GPU? So I should've gone for Intel or NVIDIA? 

Here is the info:
 

4:00:57 - File System Filter 'npsvctrig' (10.0, ‎1978‎-‎05‎-‎14T13:13:30.000000000Z) has successfully loaded and registered with Filter Manager.
4:00:57 - [CRITICAL] The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.
 

Then

4:02:07 - DHCPv4 is waiting on DHCPv6 service to stop
4:02:07 - [ERROR] Device Association Service detected an endpoint discovery failure.
 

Then

4:05:24 - [WARNING] The application-specific permission settings do not grant Local Activation permission for the COM Server application with CLSID 
{2593F8B9-4EAF-457C-B68A-50F6B8EA6B54}
 and APPID 
{15C20B67-12E7-4BB6-92BB-7AFF07997402}
 to the user [computer name]\narut SID (S-1-5-21-832927817-1512349839-1773223518-1001) from address LocalHost (Using LRPC) running in the application container Unavailable SID (Unavailable). This security permission can be modified using the Component Services administrative tool.

4:05:44 - [ERROR] The server Windows.Gaming.GameBar.PresenceServer.Internal.PresenceWriter did not register with DCOM within the required timeout.

[Mumintroll]

Have you tried plugging in the pc from another power outlet, from a different place in your home?

Or even in another house? Thinking it could be something glitchy with the electricity in your home...

[Ambientcore]

7 minutes ago, Mumintroll said:

Have you tried plugging in the pc from another power outlet, from a different place in your home?

Or even in another house? Thinking it could be something glitchy with the electricity in your home...

Did that, still happened. Went to the XFX store, too. It happened there many at times.

[KidKid]

4 hours ago, Ambientcore said:

If you haven’t read any of my previous threads, I suggest you do.

https://linustechtips.com/topic/1608244-pc-is-possessed-and-reboots-automatically-with-no-way-to-figure-out-the-problem/
https://linustechtips.com/topic/1605904-pc-reboots-while-gaming-–-at-my-wit’s-end-after-countless-fixes/
https://linustechtips.com/topic/1605781-pc-keeps-restarting-while-gaming-–-need-help/
 

I’m really at my wits’ end. I’m so tired—exhausted. I haven’t had my computer go a single week without shutting down automatically during gaming. It doesn’t matter what games I play—Cyberpunk 2077, The Last of Us Part I, Horizon Forbidden West—it always violently reboots and with very harsh glitch effects, with no display right before the screen goes black. Then I hear the Windows welcome tune. My motherboard gets stuck on the white LED. I keep getting a Kernel-41 error. Fucking Hell!

I assume that has something to do with VGA.
 

Here’s what I’ve done so far (though I’m sure I’ve missed a few things):

• Replaced the motherboard twice

• Upgraded to a higher-wattage PSU

• Replaced the GPU twice

• Replaced the RAM twice (QVL listed)

• Got a new CPU from the AMD Singapore warehouse

• Got a new case

• Installed a new CPU cooler

• Got a new SSD for Windows 11

I’ve also tried all of this:

• Set refresh rate to 60Hz

• Lowered in-game settings

• Enabled high-performance power plan

• Moved the PC to a different location in the house

• Rolled back drivers

• Installed only GPU drivers (no Adrenalin software)

• Skipped Realtek audio drivers (thought they might be interfering)

• Switched to an old HDMI cable (which performed poorly)

• Undervolted the GPU (made things worse)

• Enabled and disabled DOCP

• Enabled V-Sync

• Limited in-game FPS

• Disassembled and reassembled the PC twice

• Swapped out nearly every part one by one—except for the HDMI cable and the monitor

• Tested on a different monitor at the XFX store (still shut down)

• Disabled Game Mode in Windows

• Disabled automatic restart

• Disabled C-State

• Yes, the PSU and to GPU have separate power cables

• BIOS is updated

NOTHING. WORKED.
 

Current Setup:

• CPU: AMD Ryzen 5 5600X

• PSU: EVGA P6 1000W

• GPU: XFX MERC 6900 XT

• Motherboard: ASUS STRIX B550 Wi-Fi II

• Cooler: DeepCool

• RAM: Patriot Viper Steel 32 GB (3200 MHz)

• Monitor: ASUS VA24E (HDMI)

This has been going on since February, and I have broken down twice. I am on the verge of giving up, and I am exhausted. I am tired. 

@KidKid tagging you since you've been here since the beginning.

I'm sorry to hear this. I'll try to help you the best I can!

First, a question. Does the system crash only while playing video games? Try running a CPU and GPU stress test (one by one), and see if it crashes. If it does, the problem is likely power, as mentioned below. If it doesn't crash while running stress tests, reply to me without following the instructions below. I hope I can help!

One thing I noticed you haven't tried yet is a UPS power supply. Too often, in LTT videos, their PCs are unstable and crash due to dirty wall power, and using a UPS power supply has often fixed the crashing. Linus mentioned that a higher quality power supply will help, but even the best power supplies aren't meant to run on dirty wall power. You could also try unplugging any other unnecessary device plugged into the same outlet that your computer is plugged into. Here is what you can try in order-

1. Plug it into a friend or family's wall outlet, in a different house. If it works in another house/building, try #2 below.
2. Try using a UPS power supply. They can be found for under $100.

I hope I can help! Good luck!

[Ambientcore]

33 minutes ago, KidKid said:

I'm sorry to hear this. I'll try to help you the best I can!

First, a question. Does the system crash only while playing video games? Try running a CPU and GPU stress test (one by one), and see if it crashes. If it does, the problem is likely power, as mentioned below. If it doesn't crash while running stress tests, reply to me without following the instructions below. I hope I can help!

One thing I noticed you haven't tried yet is a UPS power supply. Too often, in LTT videos, their PCs are unstable and crash due to dirty wall power, and using a UPS power supply has often fixed the crashing. Linus mentioned that a higher quality power supply will help, but even the best power supplies aren't meant to run on dirty wall power. You could also try unplugging any other unnecessary device plugged into the same outlet that your computer is plugged into. Here is what you can try in order-

1. Plug it into a friend or family's wall outlet, in a different house. If it works in another house/building, try #2 below.
2. Try using a UPS power supply. They can be found for under $100.

I hope I can help! Good luck!

Only gaming. I can keep the PC on for HOURS. No reboots or random shutdowns. I can use it as a music production and video editing beast, and I'd be good to go (at least that is what I think).

I will consider a UPS. It is difficult to get the money for UPS right now. I don't know whom to ask which one is good or not. I have seen the prices soar to 1,000 AED or over 200 dollars. 

[Killjoy_NS]

Interesting.
Try to disable "Windows Update" service. It can be pain in the a$$....that thing restarted my PC few times while gaming few years ago just because Windows decided that while I play game, it is a regular time for a restart.

Any "beeps" from BIOS???

Any info on temperatures?

[Ambientcore]

5 minutes ago, Killjoy_NS said:

Interesting.
Try to disable "Windows Update" service. It can be pain in the a$$....that thing restarted my PC few times while gaming few years ago just because Windows decided that while I play game, it is a regular time for a restart.

Any "beeps" from BIOS???

Any info on temperatures?

Every time I play games, the CPU is between 70-80 °C. Same for the GPU.

Beeps? No beeps from the BIOS. Nothing out of the ordinary. 

[Killjoy_NS]

2 minutes ago, Ambientcore said:

Every time I play games, the CPU is between 70-80 °C. Same for the GPU.

Beeps? No beeps from the BIOS. Nothing out of the ordinary. 

 

Try "WhoCrashed". There is a little app/program which can give you some insight into crashing.
Any errors from Windows after restart?

 

[Ambientcore]

29 minutes ago, Killjoy_NS said:

 

Try "WhoCrashed". There is a little app/program which can give you some insight into crashing.
Any errors from Windows after restart?

 

No, I have to hard reboot because my computer is stuck on the white LED (motherboard).

[Killjoy_NS]

4 minutes ago, Ambientcore said:

No, I have to hard reboot because my computer is stuck on the white LED (motherboard).

Can you take a picture of the inside of your case?

It "smells" to me that you didn't do something right.....maybe.

I've seen that you set your ram at 3000......have you tried to put it on a default, 2400?
What is the voltage of your ram when it's on 3600?

Or simply try to "load defaults" in BIOS.

Can you check health of your SSD? DO you know HOW to check health?

[Ambientcore]

2 minutes ago, Killjoy_NS said:

Can you take a picture of the inside of your case?

It "smells" to me that you didn't do something right.....maybe.

I've seen that you set your ram at 3000......have you tried to put it on a default, 2400?
What is the voltage of your ram when it's on 3600?

Or simply try to "load defaults" in BIOS.

Can you check health of your SSD? DO you know HOW to check health?

I have checked the health of my SSD through Sandisk. 

I have run the RAM without DOCP (2400 MHz) and with DOCP (3200 MHz). 

Still, restarts occur. 

I built the computer with a professional, I was there. I made sure everything was put inside perfectly. 

 

Not sure about RAM voltage. 

[Killjoy_NS]

2 minutes ago, Ambientcore said:

I have checked the health of my SSD through Sandisk. 

I have run the RAM without DOCP (2400 MHz) and with DOCP (3200 MHz). 

Still, restarts occur. 

I built the computer with a professional, I was there. I made sure everything was put inside perfectly. 

 

Not sure about RAM voltage. 

 

A lot of ppl are considering themselves professionals...so...
Ram at 3600 should be at 1.35V.

Honestly, I would like to see inside of your PC.
Don't worry, even professionals can forget something.

[Ambientcore]

Currently trying CPU Core Ratio to 40 and set my CPU voltage override to 1.0250. Didn't help.

@KidKid @Brennan Price

[spacepickle]

I think someone already mentioned this, but have you tried stress-testing individual components?

[Ambientcore]

3 minutes ago, spacepickle said:

I think someone already mentioned this, but have you tried stress-testing individual components?

Testing now FurMark - OCCT VRAM (100%) together. 10 min - no reboots. 

[spacepickle]

15 minutes ago, Ambientcore said:

Testing now FurMark - OCCT VRAM (100%) together. 10 min - no reboots. 

OCCT might be a good one to try as well if FurMark does fine. This will help you weed out exactly the component that is giving you issues. https://www.ocbase.com/download

[Ambientcore]

8 minutes ago, spacepickle said:

OCCT might be a good one to try as well if FurMark does fine. This will help you weed out exactly the component that is giving you issues. https://www.ocbase.com/download

No shutdowns. Nothing. A-OK. I am so confused.

[KidKid]

14 hours ago, Ambientcore said:

No shutdowns. Nothing. A-OK. I am so confused.

I noticed you said you get Kernel Error 41. That is related to power, thus #3 in the questions is probably what you should try next.
Some questions-

1. How long did you stress test for?
2. How long does it take to crash in games?
3. Did you try a UPS power supply, or plan to? This may help, since Kernel Error 41 is a power related problem. The power supply is fine, but the wall power may still not be clean (like I mentioned earlier), making it hard for any PSU to run properly.
4. Did you take your PC to any repair shops? Did it crash then in the repair shops? If it didn't, you probably need a UPS power supply (this is not a PC power supply, but rather an uninterruptable power source for any device, PC or not, and can be found for under $100)

I'll be here and do my best!

[Ambientcore]

1 hour ago, KidKid said:

I noticed you said you get Kernel Error 41. That is related to power, thus #3 in the questions is probably what you should try next.
Some questions-

1. How long did you stress test for?
2. How long does it take to crash in games?
3. Did you try a UPS power supply, or plan to? This may help, since Kernel Error 41 is a power related problem. The power supply is fine, but the wall power may still not be clean (like I mentioned earlier), making it hard for any PSU to run properly.
4. Did you take your PC to any repair shops? Did it crash then in the repair shops? If it didn't, you probably need a UPS power supply (this is not a PC power supply, but rather an uninterruptable power source for any device, PC or not, and can be found for under $100)

I'll be here and do my best!

I ran the FurMark test with OCCT for 20 minutes. I can barely last 20 minutes in The Last of Us Part I game.
Sometimes within 10 minutes, sometimes within 20 minutes, sometimes not for hours.
Yes, it crashed at the repair store around March when I had my GPU replaced. They replaced it twice. 
I don't have money for a UPS. If I do, I'll get it in the future. 
Yes, when I went to replace my GPU for the first time, it crashed there, too. They replaced it again on the same day and time. Came home, crashed the next day.

[KidKid]

26 minutes ago, Ambientcore said:

I ran the FurMark test with OCCT for 20 minutes. I can barely last 20 minutes in The Last of Us Part I game.
Sometimes within 10 minutes, sometimes within 20 minutes, sometimes not for hours.
Yes, it crashed at the repair store around March when I had my GPU replaced. They replaced it twice. 
I don't have money for a UPS. If I do, I'll get it in the future. 
Yes, when I went to replace my GPU for the first time, it crashed there, too. They replaced it again on the same day and time. Came home, crashed the next day.

If you have extra hardware, maybe you could sell the not needed parts to raise funds. Just an idea.

[Tan3l6]

Maybe test with Linpack Xtreme stress test - it really seems to be more thorough test than OCCT.

It shows RAM and CPU stability quite fast.

 

In my experience - RAM test shows as fail in the command line, while CPU instability just freezes the machine.

[The K-Man89]

On 5/22/2025 at 6:49 AM, Mumintroll said:

Have you tried plugging in the pc from another power outlet, from a different place in your home?

Or even in another house? Thinking it could be something glitchy with the electricity in your home...

I was going to say the same thing, depending where you live there could be problems with the power delivery to your home triggering these safeguards.