System freezes when under heavy stress, all troubleshooting attempts failed
System freezes when under heavy stress, all troubleshooting attempts failed
So, I've been having this issue for months now, and I'm at my wits end trying to figure out what is going on. My PC shuts down completely unexpectedly during intense gaming loads (BO6, Elden Ring, Cyberpunk, Control, etc), and I've tried just about everything to trace the issue. Here is my build: Part Model CPU AMD Ryzen 7 5700X Motherboard Gigabyte B550M Aorus Pro-P RAM 32GB (4x8GB) Silicon Power DDR4-3200 CL16 GPU MSI RX 6700XT Dual OC PSU Thermaltake Smart BM2 750W 80+ Bronze M.2 Storage Silicon Power NVMe M.2 PCIe Gen3x4 x2 (1x 1TB +1x 512GB [boot drive]) SATA Storage 3x Teamgroup T-Force Vulcan 1TB; 1x Adata SP550 120gb) I have tried just about everything under the sun to diagnose this issue. In order, I have done the following: Checked CPU/GPU temps and replaced thermal paste/pads for both. The system stays around 70C even under load now. Upgraded CPU (was previously a Ryzen 5 5600. Not necessarily done as a troubleshooting step, but still relevant. The issue was present with both CPUs) Swapped PSU and cables (the most frequent answer to this issue seems to be PSU issues, so I bought a Corsair RM750X. This did not solve the issue, so I swapped back to the Thermaltake.) Changed surge protectors Resolved various Critical errors in Event Viewer. There are now no errors in Event Viewer prior to the shutdowns Ran OCCT (no errors or shutdowns) Ran Unigine Superposition (no shutdowns) Reinstalled AMD Adrenalin drivers Re-seated RAM and GPU Ran Memtest86 (no errors) My thoughts on potential next steps are: Try an 850W PSU (my system shouldn't be using that much power, but it's possible it has high spikes?) I'm leaning toward this not being the solution, because I turned on "AC back after power loss" in the BIOS, and it does NOT turn back on when it shuts down, indicating to me that this is not a power issue) Try different RAM (2x16GB maybe, I know 4 sticks can cause instability sometimes) Try a different motherboard (could be a power delivery problem with the board?) I've got no other ideas as to what to do here. If anyone has had a similar issue, let me know what you did to resolve it. I found this thread, where the user fixed his similar issue by tearing his whole build down, buying a new Windows license, etc, but I don't have much faith that it could be that simple.
It's often a final attempt, but if you have another motherboard, why not look for good offers right now? The RAM performed well in testing with memtest. I don't think it's worth spending extra to match the same capacity. You've already tried a different power supply and the problem persists. Before upgrading the PSU, I'd investigate other factors first. What errors were you encountering and what steps did you take to resolve them? Where did you purchase these SSDs? I bought many on Amazon, but they all failed due to overheating controllers. If your computer freezes under stress, these could be the culprits. I swapped all SP SSDs for Western Digital SN550, and they're working perfectly now. Consider buying one SSD and using it as a boot drive. Maybe I'm mistaken, but it shouldn't cost much to test. For monitoring, install tools like CPU-z or HWinfo. Enabling logging will show if temperatures or voltages are abnormal during shutdowns. Let me know if this helps.
Thanks for the response. You're correct that the motherboard seems like a final resort, but I've noticed some hints about the chipset possibly being affected. Swapping it out will be my last resort. Regarding the RAM, I was planning to test two sticks to check for stability when all slots are full before spending more. It makes sense that you'd want to confirm this before buying new RAM. I agree it's unlikely the power supply is the issue. The "AC power back" setting helped solidify my belief it wasn't a power loss problem. In Event Viewer, I saw several errors, all happening just before shutdowns. They included Hypervisor launch failure, missing or disabled SVM in BIOS, VSM not initialized, and Secure Boot issues. Fixing virtualization in BIOS resolved the first two. The Secure Boot update failed due to an error about Secure Boot not being enabled. After resetting Secure Boot, it worked properly. The AMDRyzenMasterDriverV27 service couldn't start because a file already existed, so I had to reinstall the drivers. The GameInput Service stopped unexpectedly six times. A Reddit discussion mentioned a duplicate GameInput on some systems, which could be causing problems. These issues shouldn’t usually lead to crashes, but it’s worth investigating. You might be onto something with the SSDs—those I bought about two years ago. Did your failing drives show up in CrystalDiskMark? I’ll look into current Amazon deals for SSDs.
It seems these items match the time of purchase, possibly indicating a poor batch or fake products. No symptoms or alerts have been reported. They may work one day and vanish the next, either when cooled or never again. This is the most affordable controller I've ever owned. Always verify with other retailers. MemoryExpress and CanadaComputers are more trustworthy than Amazon for computer components. Supporting nearby shops can also make a difference. Good luck!