F5F Stay Refreshed Hardware Desktop Determine where crashes originate in the latest construction.

Determine where crashes originate in the latest construction.

Determine where crashes originate in the latest construction.

M
msbunny13
Member
62
04-11-2022, 03:08 AM
#1
Initially sorry for the long message. I'm on mobile trying to pinpoint the cause of those hard stutters and crashes. Recently I assembled a new PC – 13600k, MSI Pro Z790A DDR5, Corsair vengeance RGB 2x 16gb, Samsung M2, Thermaltake 750w PSU. A friend lent me his old 1080ti while I waited for a new graphics card. The build was fine physically but things have been tricky since. At first I couldn't install Windows from a bootable USB, it would fail early. I took it to a local shop and they managed to get it running by tweaking BIOS settings – not sure what they did. I was worried components might be the issue, so I ran all available stress tests and benchmarks. The CPU performed well, RAM tests passed after long sessions, SSD is fine. Only the GPU caused problems. Stressing the 1080ti would trigger instantly: GPU usage hit 100%, tests failed with errors, screen would freeze, and the computer would lock up or crash to desktop. I tried games too – same crash under heavy load, no BSODs. To fix it: reinstalled Windows, updated drivers cleanly, reset BIOS, disabled XMP, undervolted GPU, booted in safe mode. Still the same issues. I reinserted the card, used a different PCIe slot, swapped cables from PSU, and even tried an older GTX 1070 which worked perfectly. The 1080ti works fine on my friend’s machine a few years back. It seems during those intense usage spikes it draws so much power that it dims RGB, slows fans, or even crashes the RGB controller. I ran without the RGB hub and connected fans directly to the motherboard – still got crashes. Does this look like the 1080ti is broken? Could my PSU be insufficient? Might RAM and the motherboard be mismatched? Anything I’m missing? I’m glad everything works with the 1070, but I’d like to swap the PSU or RAM while I can get a refund if the mind thinks they’re the problem. Thanks for your help.
M
msbunny13
04-11-2022, 03:08 AM #1

Initially sorry for the long message. I'm on mobile trying to pinpoint the cause of those hard stutters and crashes. Recently I assembled a new PC – 13600k, MSI Pro Z790A DDR5, Corsair vengeance RGB 2x 16gb, Samsung M2, Thermaltake 750w PSU. A friend lent me his old 1080ti while I waited for a new graphics card. The build was fine physically but things have been tricky since. At first I couldn't install Windows from a bootable USB, it would fail early. I took it to a local shop and they managed to get it running by tweaking BIOS settings – not sure what they did. I was worried components might be the issue, so I ran all available stress tests and benchmarks. The CPU performed well, RAM tests passed after long sessions, SSD is fine. Only the GPU caused problems. Stressing the 1080ti would trigger instantly: GPU usage hit 100%, tests failed with errors, screen would freeze, and the computer would lock up or crash to desktop. I tried games too – same crash under heavy load, no BSODs. To fix it: reinstalled Windows, updated drivers cleanly, reset BIOS, disabled XMP, undervolted GPU, booted in safe mode. Still the same issues. I reinserted the card, used a different PCIe slot, swapped cables from PSU, and even tried an older GTX 1070 which worked perfectly. The 1080ti works fine on my friend’s machine a few years back. It seems during those intense usage spikes it draws so much power that it dims RGB, slows fans, or even crashes the RGB controller. I ran without the RGB hub and connected fans directly to the motherboard – still got crashes. Does this look like the 1080ti is broken? Could my PSU be insufficient? Might RAM and the motherboard be mismatched? Anything I’m missing? I’m glad everything works with the 1070, but I’d like to swap the PSU or RAM while I can get a refund if the mind thinks they’re the problem. Thanks for your help.