PC reboots while gaming – At my wit’s end after countless fixes

[Ambientcore]

I am at my wit’s end. I feel so defeated, and this is almost a vent post because I don’t know where to go or what to do. I love gaming, but I can no longer play on my newly built PC—the one I waited 10 years to get. After studying for so long at university, I finally have some free time, yet I still can’t play games. My PC keeps shutting down while playing video games, whether it’s Cyberpunk 2077 or The Last of Us Part I—sometimes even when I’m just idling in the menu.

 

I have tried everything, and at this point, I’m considering selling my one-month-old computer.

Here’s what I’ve done so far:

• I RMA’d my GPU due to green screens and artifacts.
• I replaced my PSU, upgrading from 850W to 1000W.
• I ran MemTest86 twice—once at 3000 MT/s and once at 3603 MT/s. No errors were shown.
• I conducted extensive benchmark tests using CPU burner (FurMark) and OCCT.
• I ran extensive GPU stress tests with FurMark and OCCT.
• I removed the back panel, thinking my smaller case was cramping my wiring.
• I plugged my PC directly into a wall socket instead of a power strip or adapter.
• I installed a CPU cooler to regulate temperatures.

All of my components have been tested, yet some people suggest I should replace my RAM or motherboard. At this point, I have no idea what the real issue is. My budget was exhausted long ago, and I just feel lost.

 

I’ve been in touch with @KidKid, but regardless of the game, my PC reboots automatically with a black screen—though the fans keep running and the RGB lights remain on. I have to perform a hard reboot just to get my display back.

 

I have no idea what’s causing this anymore. It feels like a curse like I don’t even deserve to own a PC.

 

I tried lowering my RAM speed to 3000 MT/s, and while it allowed me to game for longer, the PC eventually shut down again. At 3603 MT/s, the crashes happen much sooner.

PC Specs:

• Storage: Samsung SATA 256GB, Western Digital WD_BLACK SN770 1TB
• OS: Windows 11
• Motherboard: ASUS Prime B550-PLUS
• GPU: XFX 6900 XT 16GB
• RAM: Kingston Fury Beast 32GB (2x16GB) 3600 MT/s DDR4
• CPU: AMD Ryzen 5 5600X
• PSU: EVGA 1000W P6

Any assistance or guidance would be greatly appreciated. Thank you.

[TetraSky]

Rebooting while gaming typically points to either overheating or to hardware issue.

The most likely suspect is indeed the RAM or possibly the motherboard.

Though it's strange that you're not even seeing a blue screen and it just goes dark.

 

Have you updated the motherboard to have the newest bios ?

How are your temps while gaming ? How long does it take on average before it fails ?

You mention running memtest "twice"... But is that just through a single pass ? If so, you unfortunately did not fully test the RAM. Usually, to properly test the ram you let it go for hours and hours. Especially if the issue is happens randomly. With 32GB, you would potentially need to let it run for something like 24 hours. A single pass ain't indicative of a successful test.

You could also run Prime95 on Blend with all the available RAM selected.

 

Another solution if you just don't know what the issue is.... Is to just return everything and buy anew. It's what I did years ago when I had an issue that simply didn't fix no matter what I did or change.

[Ambientcore]

10 minutes ago, TetraSky said:

Rebooting while gaming typically points to either overheating or to hardware issue.

The most likely suspect is indeed the RAM or possibly the motherboard.

Though it's strange that you're not even seeing a blue screen and it just goes dark.

 

Have you updated the motherboard to have the newest bios ?

How are your temps while gaming ? How long does it take on average before it fails ?

You mention running memtest "twice"... But is that just through a single pass ? If so, you unfortunately did not fully test the RAM. Usually, to properly test the ram you let it go for hours and hours. Especially if the issue is happens randomly. With 32GB, you would potentially need to let it run for something like 24 hours. A single pass ain't indicative of a successful test.

You could also run Prime95 on Blend with all the available RAM selected.

 

Another solution if you just don't know what the issue is.... Is to just return everything and buy anew. It's what I did years ago when I had an issue that simply didn't fix no matter what I did or change.

Greetings,
 

No overheating, I promise. I cap my FPS at 70 to minimize GPU and CPU usage. I also monitor my temperatures using AMD's software, and both my GPU and CPU stay below 70°C.
 

When I kept the RAM at 3000 MT/s, my PC took around five hours (give or take) before rebooting to a completely black screen while the fans were still running.
 

I ran MemTest86 with four passes, each lasting approximately 4.5 hours. I’m not sure what a single pass means. I also don’t think I could sleep at night with my PC running and so many lights switched on.
 

I have never used Prime95 on Blend. I don’t know what it is or how to use it. How long does the test take? 

Returning the whole thing, ya... Well, considering this, too. Feel so broken, feel so defeated. 

[KnoT]

Did you check GPU Hotspot Temps?
 

[Ambientcore]

1 minute ago, KnoT said:

Did you check GPU Hotspot Temps?
 

I have the GPU temps via Adrenalin software. No issues with my temperature. I have run OCCT (VRAM) + FurMark. It ran for 20 minutes with no issue. 

[OhYou_]

idk just push higher SoC voltage, like 1.15v and see if the problem goes away.
look for it in overclocking settings.
I think your integrated memory controller is just mid though it *should* do 3000

[Ambientcore]

2 minutes ago, OhYou_ said:

idk just push higher SoC voltage, like 1.15v and see if the problem goes away.
look for it in overclocking settings.
I think your integrated memory controller is just mid though it *should* do 3000

What is a memory controller? I am very new at these things. Idk where the SOC Voltage is located. If I get a new RAM it will be at 3200 MT/s. It is similar to the connectivity my CPU is compatible with. 

[OhYou_]

Just now, Ambientcore said:

What is a memory controller? I am very new at these things. Idk where the SOC Voltage is located. If I get a new RAM it will be at 3200 MT/s. It is similar to the connectivity my CPU is compatible with. 

idk then
so like you put this together and then did nothing and it started crashing?
go into bios where you are setting the speed to 3000 or 3600 or whatever, and just turn off XMP/DOCP, idk what it looks like, look for memory profile, set to auto or whatever option isnt XMP.
then boot windows and verify if your memory is now running at 2133mhz using something like cpu-z or whatever tool you have been using.
now test if it is stable

[Bwithnewcast]

Tried reinstalling windows compleatly? Tried running ddu and reinstalling graphics drivers? May also be a software problem

 

Also check if you have the latest bios update installed?

[Ambientcore]

2 minutes ago, OhYou_ said:

idk then
so like you put this together and then did nothing and it started crashing?
go into bios where you are setting the speed to 3000 or 3600 or whatever, and just turn off XMP/DOCP, idk what it looks like, look for memory profile, set to auto or whatever option isnt XMP.
then boot windows and verify if your memory is now running at 2133mhz using something like cpu-z or whatever tool you have been using.
now test if it is stable

I gave it to a professional for assembly and spent money on the service. It worked until my GPU started acting up. So, I sent it in for a replacement. Now, my PC randomly shuts down or reboots with a black screen. I thought it was the PSU. I changed that, too. The only thing I can do is disable DOCP, set it to no profile, and let it run at 2400 MT/s. Maybe it’ll work... I haven’t extensively tested my computer at 2400 MT/s.

[Ambientcore]

Just now, Bwithnewcast said:

Tried reinstalling windows compleatly? Tried running ddu and reinstalling graphics drivers? May also be a software problem

Re-installed Windows twice. No budge. 

[OhYou_]

1 minute ago, Ambientcore said:

I gave it to a professional for assembly and spent money on the service. It worked until my GPU started acting up. So, I sent it in for a replacement. Now, my PC randomly shuts down or reboots with a black screen. I thought it was the PSU. I changed that, too. The only thing I can do is disable DOCP, set it to no profile, and let it run at 2400 MT/s. Maybe it’ll work... I haven’t extensively tested my computer at 2400 MT/s.

I would highly recommend trying it at 2400. this will almost completely rule out a memory/cpu instability issue.
 

[Ambientcore]

1 minute ago, OhYou_ said:

I would highly recommend trying it at 2400. this will almost completely rule out a memory/cpu instability issue.
 

Wouldn't that be bad for my CPU? I’ve been told that running my memory at a lower speed isn’t good for my CPU. I don’t care about FPS since I limit it to 70 anyway. I rarely, if ever, play competitive games—I only play single-player games, mainly open-world and RPGs.

[Bwithnewcast]

Is the ram slotted in the a2 and b2 slots? That would be the light grey slots not the dark, had that problem with my asus motherboard when i tried to run the ram at 4600 mhz instead of 2133 mhz

[OhYou_]

1 minute ago, Ambientcore said:

Wouldn't that be bad for my CPU

in no way would it ever be bad to run ram at DDR4’s JEDEC standard speed.
that is what turning off xmp does. XMP is overclocking, if you turn it off, the ram then runs in the actual ddr4 spec.

[Ambientcore]

2 minutes ago, Bwithnewcast said:

Is the ram slotted in the a2 and b2 slots? That would be the light grey slots not the dark, had that problem with my asus motherboard when i tried to run the ram at 4600 mhz instead of 2133 mhz

I have installed it in alternative slots. So, | (|) | (|) - the ones in brackets ( | = RAM slot). 

[Timme]

Open Win+S->Event Viewer
Un administrative events or under Windows logs you should see an error report.

[Ambientcore]

3 minutes ago, OhYou_ said:

in no way would it ever be bad to run ram at DDR4’s JEDEC standard speed.
that is what turning off xmp does. XMP is overclocking, if you turn it off, the ram then runs in the actual ddr4 spec.

But why is my RAM advertised at 3600 MT/s?

[Ambientcore]

4 minutes ago, Timme said:

Open Win+S->Event Viewer
Un administrative events or under Windows logs you should see an error report.

Kernel-Power: the system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

[strange13930]

51 minutes ago, TetraSky said:

Rebooting while gaming typically points to either overheating or to hardware issue.

The most likely suspect is indeed the RAM or possibly the motherboard.

Though it's strange that you're not even seeing a blue screen and it just goes dark.

 

Have you updated the motherboard to have the newest bios ?

How are your temps while gaming ? How long does it take on average before it fails ?

You mention running memtest "twice"... But is that just through a single pass ? If so, you unfortunately did not fully test the RAM. Usually, to properly test the ram you let it go for hours and hours. Especially if the issue is happens randomly. With 32GB, you would potentially need to let it run for something like 24 hours. A single pass ain't indicative of a successful test.

You could also run Prime95 on Blend with all the available RAM selected.

 

Another solution if you just don't know what the issue is.... Is to just return everything and buy anew. It's what I did years ago when I had an issue that simply didn't fix no matter what I did or change.

 

8 minutes ago, Bwithnewcast said:

Is the ram slotted in the a2 and b2 slots? That would be the light grey slots not the dark, had that problem with my asus motherboard when i tried to run the ram at 4600 mhz instead of 2133 mhz

 

4 minutes ago, Ambientcore said:

I have installed it in alternative slots. So, | (|) | (|) - the ones in brackets ( | = RAM slot). 

it been asked twice, is your bios up to date?

[Ambientcore]

Just now, strange13930 said:

 

 

it been asked twice, is your bios up to date?

Yes, I checked this morning. It is.

[wONKEyeYEs]

1 hour ago, Ambientcore said:

My PC keeps shutting down while playing video games

 

20 minutes ago, Bwithnewcast said:

May also be a software problem

Shut down any software other than your game.

I had this problem with a 6900XT, it was caused by AIDA64.

[Timme]

5 minutes ago, Ambientcore said:

Kernel-Power: the system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.

Is it event id 41?

[Ambientcore]

11 minutes ago, wONKEyeYEs said:

 

Shut down any software other than your game.

I had this problem with a 6900XT, it was caused by AIDA64.

Never heard of AIDA64 in my entire life. Haha.

[Ambientcore]

19 minutes ago, Timme said:

Is it event id 41?

Yes.