Jump to content

Whenever I run a CPU stress test, it runs fine until I stop it, at which point my computer restarts. I have tried both the OCCT CPU test, and Prime95 SmallFFT. I have tested by using singular ram sticks and it still restarted with either of them. Everything is at stock with nothing overclocked.

I originally started testing because Helldivers 2 would restart my computer within a minute of getting into my ship (this was with the GPU installed), now it restarts right after I start the game.

 

I have tried reseating the CPU and it did nothing, I also checked and could not see any bent pins in the socket. 

 

Is this more likely to be a motherboard/cpu issue or a power supply issue?

 

My specs are:

  • Windows 11 Pro 64-bit
  • Asrock Z890 Pro-A Wifi (2.21, latest stable release)
  • intel Core Ultra 7 265k
  • 2x16GB Team Group T-CREATE Expert (currently on stock 4800MT/s)
  • MSI Spatium Series M482 2TB SSD
  • Seasonic G-Series 650w (from an old build, first time being used in ~5 years)
  • ONIX ODYSSEY ARC B580 (currently completely removed from system)
Link to post
Share on other sites

Bios and drivers up to date? PSU could be a factor, though  you'd need a PSU tester or another unit to test in your system to confirm.

Main Desktop: CPU - i9-14900k | Mobo - Gigabyte Z790 Aorus Elite AX | GPU - Sapphire AMD RX 7900XTX Nitro+ 24GB RAM - Corsair Vengeance Pro RGB 64GB 6400mhz | AIO - Arctic Liquid Freezer III 360mm | PSU - Corsair RM1000X | Case - Hyte Y40 - White | Storage - Samsung 980 Pro 1TB Nvme /  Sabrent Rocket 4 Plus 4TB Nvme / Samsung 970 EVO Plus 2TB Nvme / Samsung 870 EVO 4TB SSD / Samsung 870 QVO 2TB SSD/ Samsung 860 EVO 500GB SSD|

 

TV Streaming PC: Intel Nuc CPU - i7 13th Gen | RAM - 16GB DDR4 3200mhz | Storage - Crucial P3 Plus 1TB Nvme |

 

Phone: Samsung Galaxy S24+ - Ceramic White 512GB |

 

Link to post
Share on other sites

Sounds like a voltage issue. Either your CPU is faulty due to the sudden drop in voltage (stock issues, or poor undervolt/overclock), or a failing PSU. My guess is on the later—try purchasing a new PSU and see if it fixes your issues.

Jude 1:23

And on some have compassion, making a distinction; but others save with fear, pulling them out of the fire, hating even the garment defiled by the flesh.

Link to post
Share on other sites

 

33 minutes ago, MidnightStars said:

Sounds like a voltage issue. Either your CPU is faulty due to the sudden drop in voltage (stock issues, or poor undervolt/overclock), or a failing PSU. My guess is on the later—try purchasing a new PSU and see if it fixes your issues.

Thank you! I will try purchasing a new psu for now and see if that fixes it, I tried updating to the beta bios and disabling the igpu/putting the dgpu back in, but those didn't work.

Link to post
Share on other sites

@Greth

 

Use HWiNFO to see what the AC / DC loadline values are set to. Reducing these values can help reduce large changes in voltage going to the CPU when a full load stress test suddenly ends. I believe the default loadline values for both of these is still 1.10. I have an older 10th Gen system where I had to lower both loadline values to 0.25 mOhm. This finally stopped the full load to no load crashes. 

 

I also have an older Seasonic power supply. That might have been the source of the problem. Nice I did not have to spend money replacing it. I have been using it for years without any issues after the loadline adjustments. Watch the VCore voltage reported by HWiNFO for any spikes or drops when testing. 

 

No need to run a lengthy stress test. You probably only need to run Prime95 for about 15 seconds to get it up to full load. 

 

image.png.585f2103d82246282bc73803c1c79d3e.png

Link to post
Share on other sites

On 2/16/2025 at 1:54 AM, unclewebb said:

 

 

Use HWiNFO to see what the AC / DC loadline values are set to. Reducing these values can help reduce large changes in voltage going to the CPU when a full load stress test suddenly ends. I believe the default loadline values for both of these is still 1.10. I have an older 10th Gen system where I had to lower both loadline values to 0.25 mOhm. This finally stopped the full load to no load crashes. 

 

I also have an older Seasonic power supply. That might have been the source of the problem. Nice I did not have to spend money replacing it. I have been using it for years without any issues after the loadline adjustments. Watch the VCore voltage reported by HWiNFO for any spikes or drops when testing. 

 

No need to run a lengthy stress test. You probably only need to run Prime95 for about 15 seconds to get it up to full load. 

 

image.png.585f2103d82246282bc73803c1c79d3e.png

@unclewebb

Thanks! I found the setting, it seems to automatically be set to 0.76 mOhm, however, if I change the setting in the UEFI away from auto to anything in its accepted range, it gets pegged to 1.8 mOhm. I'm assuming there is some other setting I need to change or there is a bug in the beta UEFI, but I have not been able to find anything on it.

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×