System powers down during demanding gameplay sessions.
System powers down during demanding gameplay sessions.
While mostly playing Valorant and occasionally Apex Legends, my PC often shuts down mid-game. This happens around the halfway point of my second Valorant match. Games such as League of Legends, Minecraft, and Terraria have never caused this issue. The system restarts on its own after shutdown, but I need to manually turn it back on. I’ve updated the BIOS, kept drivers current, and lowered GPU voltage. My system is watercooled with two 360 radiators, eliminating overheating concerns. Event viewer shows no unexpected shutdowns. This problem has persisted for about two years. There’s a hint it might relate to my power supply. My PSU lacks an extra 4-pin ATX port, using only 8 of the 12 CPU power pins. I’m unsure if this is the cause since GPU undervolting didn’t prevent shutdowns. Hardware specs: GPU – Gigabyte RTX 2080 Gaming OC, CPU – Ryzen 9 3900X, Motherboard – MSI MPG X570 Gaming Edge, RAM – 32GB DDR4 2400MHz (Corsair Vengeance LPX), PSU – Corsair RM750x 80+ Gold. Windows – 64-bit. Any suggestions would be appreciated as I’m unsure what’s truly wrong.
Use software to verify the system temperatures instead of relying on the watercooled assumption. Factors like faulty mounts, insufficient thermal paste, a dead pump, air leaks, or debris in the block can still lead to performance problems even with premium water cooling. I suggest using HWinfo64 for detailed temperature readings. Your PSU should comfortably support both the 3900X and 2080 models. The RM750x includes two 8pin EPS12V cables (three available if you have the latest RMx), but a single 8pin cable should suffice—no extra connector needed. Test Cinebench and Furmark simultaneously to observe if the system shuts down, which would stress the CPU and GPU more than typical games. Check whether pressing the case button or flipping the power supply switch triggers shutdown.
I should have been more precise earlier; I monitored temperatures just before shutdown and both CPU and GPU remained cool. I rebuilt the system several times—cleaning the blocks, replacing tubing and fittings—but the issue persists. I’ll review cooling options soon to check for overlooked steps. I was confident in my PSU being sufficient, though it felt like a guess since nothing else seemed to fix it. I plan to stress test it to confirm shutdown behavior. Again, I should have been clearer—I only need to press the case button.
I tested the system, and this was the first time my PC stopped after about 3 seconds. On the second try it passed Cinebench. The Furmark process was still active, so I ran another Cinebench round and it shut down a minute later. I logged the temperature data using my phone—CPU at 87.1°C, GPU at 63.2°C.
I didn't encounter any issues. I tested Cinebench multiple times in a row and everything worked fine. I'll continue troubleshooting tomorrow because it's getting late. Thanks for your support so far.