The PC restarts when gaming but not during a stress test?
The PC restarts when gaming but not during a stress test?
Hi.
I’ve been dealing with this issue for about eight months now. In games, I notice a sudden black screen that then restarts the system into Windows. The main hints I’ve gathered are a critical error code 41 in the Event Viewer and a serious hardware warning 193 in the Reliability History.
Here’s what I have:
- Motherboard: MSI PRO B650-s wifi
- CPU: AMD Ryzen 5 7600X
- GPU: NVIDIA GeForce RTX 4070
- RAM: Corsair 32GB KIT DDR5 5600MT/s CL40 Vengeance RGB Grey EXPO
- Cooler: Endorfy Fera 5
- Storage: Apacer AS2280Q4X 2TB
- Power Supply: GIGABYTE P750GM
Additional info:
The PC has gone through two repair shops, but in both cases it came back without any reboots. I’ve already swapped out the PSU. My CPU temperature peaked at 96°C, though before the restart it was usually around 85°C, which doesn’t look like overheating.
Any advice or suggestions would be greatly appreciated.
These are extremely high temperatures. Upgrade the cooling for your PC. What case are you using? Does it feature a mesh front or a closed front? Gaming-grade systems need strong cooling, and with a closed front, overheating is likely. Please share a photo of your system with the side panel removed (upload to imgur.com and post the link). Also, note that your PSU is one of those prone to exploding. I wouldn't recommend using it with an RTX 4070. View: https://www.youtube.com/watch?v=aACtT_rzToI
Despite seeming logical, I doubt it's the main reason. I recorded a temperature of 96°C during a CPU stress test, and the PC performed well. During gaming (with reboots), temperatures hit around 85°C. Also, 7600x seems set to operate at 95°C.
Regardless of the outcome, swap that power supply. It's a well-known ticking time bomb. It could potentially resolve your problem.
These are not typical temperature readings. You must prevent the CPU from reaching those levels.
The CPU temperature at idle is 88°, which seems high even when gaming, especially for the 7600x. Reviews suggest it should be closer to 70-75° during games. Zen 4 CPUs are built to handle heat, but you must consider your overall room temperature. Poor airflow and gradual warming can cause instability over time. Check GPU, drive, chipset, and RAM temperatures via HWiNFO to ensure they remain within safe limits. For instance, DDR5 RAM may become unstable if it reaches 70°. You might try running the system open for a while to observe any crashes. Instability is possible, leading to occasional crashes after gaming sessions. The issue could stem from RAM, CPU, motherboard, PSU, or GPU. Running memory tests like memtest86 or AIDA64 stability checks can help identify problems.
Thanks for your response. The idle CPU is around 45°C in the package. I recently removed my glass panel, installed a fan inside and another large fan outside the case, both providing airflow. The system still restarts within the same time as before. I performed several stress tests, including the ones you mentioned, and all passed without issues.
If the issue appears only during gaming and not under CPU/RAM stress tests (and RAM passes all checks), it suggests the crashes occur solely when the GPU is active—indicating a potential GPU or PSU problem. As John mentioned, upgrading to a better PSU, such as the Corsair RM750X, might resolve the issue. If the problem persists after that change, it’s likely not related to the PSU.
I also performed GPU stress tests without any failures. Since I've tested various PSUs and the problem continued, it's highly improbable the power supply is responsible.
The maximum temperature for this model is 95°C. I anticipate it will slow down before reaching that point. At 96°C it would be around 204.8°F. Higher temperatures also increase power consumption, which can raise temperatures further. It’s unclear if the motherboard’s power protection circuits reset the CPU in such cases. (For situations where throttling was turned off in BIOS) The GPU can handle up to 90°C and draws about 200 watts. I recommend inspecting the 16-pin power connector for any damage or discoloration, ensuring a solid connection. If defective, the card might attempt to extract excess power from the PCI/ bus, triggering protection mechanisms that reset the CPU. (No memory dump or bug check performed)