[Redchill73]

Setup: Ryzen 5 5600x, 6800xt, windows 10, I built it myself 1.5 years ago
The issue started showing up about a month ago, I updated and factory installed the driver but not solved.

The computer freezes randomly and semi-regularly (every 3h~ avg) when I play games (or sometimes just browsing). I initially just pressed the restart button, but it got more regular and has ruined a few games/disrupted a few meetings. When the computer freezes, the RGB breathing effect in my case also freezes, which suggests that it's not the graphics card but the CPU. Considering replacing the CPU, but I want to see if anyone here knows a solution to this problem.

Updates:
1. I tried clean boot, did not help with this issue
2. Upgraded to windows 11, which seemed to reduce the frequency of this happening, but it is still there. So the problem is at least partially software
3. Upgraded to the latest graphics card driver
4. Upgraded CPU to Ryzen 5700G.


Additional observation:
1. In AMD Software: Adrenalin Edition Version
-> Performance -> Tuning: CPU tuning disappeared after this problem occurring
-> Metrics: All CPU metrics are not showing except utilization
Reinstalling the driver shows the CPU correctly again

HWMonitor right after crash and restart

 

[spunk.funk]

Try a Clean Boot. If that works you can leave it this way, or if you want to find out what is causing it, Add one Service or Startup Item at a time until the stuttering starts happening, then you will know the culprit. If that still doesn't work then proceed to diagnosing the hardware.

 

[Redchill73]

Try a Clean Boot. If that works you can leave it this way, or if you want to find out what is causing it, Add one Service or Startup Item at a time until the stuttering starts happening, then you will know the culprit. If that still doesn't work then proceed to diagnosing the hardware.

Thank you for the suggestion, I'll definitely look into the software side as you suggested.

 

[Redchill73]

Try a Clean Boot. If that works you can leave it this way, or if you want to find out what is causing it, Add one Service or Startup Item at a time until the stuttering starts happening, then you will know the culprit. If that still doesn't work then proceed to diagnosing the hardware.

Tried it, didn't help. I have updated my original post, do you have any thoughts?

 

[MPR]

What power supply did you use in your build? Periodic lockups can be caused by a failing power supply. Use HWMonitor to see what your voltages are reading and if they remain within specs when using your computer. The application can also tell you if cooling might be an issue.

 

[Redchill73]

What power supply did you use in your build? Periodic lockups can be caused by a failing power supply. Use HWMonitor to see what your voltages are reading and if they remain within specs when using your computer. The application can also tell you if cooling might be an issue.

Thank you for the suggestion. I will use it and report back. My power supply is Seasonic Focus GX-750 80+Gold, which has a pretty good reputation.

 

[Rich-M]

Yeah I have never seen a fairly new Seasonic fail, you can'
t do much better.

 

[MPR]

Yeah I have never seen a fairly new Seasonic fail, you can't do much better.

I can hardly count the number PSUs I've replaced in Dell and Compaq computers just after the warranty expired, either when they failed or when someone upgraded a video card only to have it pull more power than the supplied PSU would supply.

Seasonic PSUs are another story, and a 750 W unit should provide more than enough power for all but the highest-end gaming builds. However, it still never hurts to check voltages and temperatures to rule out two potential problem areas.

If the voltages and temps are OK then the next thing I'd check would be the power connectors. If you see the voltages change when you wiggle a connector then you may have found your culprit. Might double-check that all necessary power connectors are plugged in too.

Note about the RX 6800 XT -- AMD was having lots of problems recently with its video driver software. Most users were having to downgrade to an older driver to get their cards to work properly. A new driver was just released last week, however.

https://www.amd.com/en/support/graphics/amd-radeon-6000-series/amd-radeon-6800-series/amd-radeon-rx-6800-xt

 

[Redchill73]

I can hardly count the number PSUs I've replaced in Dell and Compaq computers just after the warranty expired, either when they failed or when someone upgraded a video card only to have it pull more power than the supplied PSU would supply.

Seasonic PSUs are another story, and a 750 W unit should provide more than enough power for all but the highest-end gaming builds. However, it still never hurts to check voltages and temperatures to rule out two potential problem areas.

If the voltages and temps are OK then the next thing I'd check would be the power connectors. If you see the voltages change when you wiggle a connector then you may have found your culprit. Might double-check that all necessary power connectors are plugged in too.

Note about the RX 6800 XT -- AMD was having lots of problems recently with its video driver software. Most users were having to downgrade to an older driver to get their cards to work properly. A new driver was just released last week, however.

https://www.amd.com/en/support/graphics/amd-radeon-6000-series/amd-radeon-6800-series/amd-radeon-rx-6800-xt

Thx for the additional suggestions. I do not see a problem in HWMonitor, but maybe I will have to wait for the issue to detect the anomaly (I haven't been using the computer much, so I have to wait a bit for the next occurrence). More intense usage seems to trigger the issue. I have attached the HWMonitor log here. If you can see something worth noting, please let me know.

I suspect the AMD driver is the culprit, as it is the only software I know that changes after a crash (disappearing CPU performance and tuning option). Also, updating the AMD driver seems to drop the problem occurrence a lot (or maybe it's the windows 11 update, I did the two back-to-back, so not sure).

 

[MPR]

Thx for the additional suggestions. I do not see a problem in HWMonitor,

I suspect the AMD driver is the culprit, as it is the only software I know that changes after a crash (disappearing CPU performance and tuning option). Also, updating the AMD driver seems to drop the problem occurrence a lot (or maybe it's the windows 11 update, I did the two back-to-back, so not sure).

Your voltages look OK and I can see no temperatures that might cause your processors to clamp down.

While Windows will allow you to keep working for a while when it want's to reboot to finish an update, it usually doesn't "play nicely" if you do. Every time my system seems to be acting up I look and, sure enough, there's as Windows up date wanting to reboot the computer in order to finish installation.

Speaking of updates. To clear everything and make sure that you are starting afresh, it's good to use the restart option (chose upgrade and restart rather than upgrade and shut down). This is because Windows startup defaults to a semi-hibernation "off" state that saves the kernel to the hard drive rather than completely loading it anew each session.

 

[Redchill73]

Your voltages look OK and I can see no temperatures that might cause your processors to clamp down.

While Windows will allow you to keep working for a while when it want's to reboot to finish an update, it usually doesn't "play nicely" if you do. Every time my system seems to be acting up I look and, sure enough, there's as Windows up date wanting to reboot the computer in order to finish installation.

Speaking of updates. To clear everything and make sure that you are starting afresh, it's good to use the restart option (chose upgrade and restart rather than upgrade and shut down). This is because Windows startup defaults to a semi-hibernation "off" state that saves the kernel to the hard drive rather than completely loading it anew each session.

Ye, I have checked the Windows update multiple times. Definitely not the issue here since no update has been pending while the freeze occurs.

 

[SpywareDr]

What I'd do next is take it back down to bare bones out on a table. Nothing but

• motherboard (on a non-conductive surface)
• CPU (w/heatsink & fan)
• RAM
• PSU
• video (including monitor)
• keyboard and mouse

Now see if it will boot into the BIOS/UEFI. If not, try a different PSU and/or RAM. Once you do get into the BIOS/UEFI reset it to factory defaults.

Now try running an Operating System (*nix) from a bootable USB flash drive. Exercise it some (web) and then leave it running overnight.

If it's still where you left it the night before, power down, hook up your drive with Windows on it and try the same, (exercise it, web, games, whatever), and leave it running overnight.

If it's fine the next morning, power down and reassemble everything back in the case. As you continue beyond the bare minimum though, only add one or two pieces of hardware at a time and then test to make sure what was added isn't causing problems. You get the idea ...