Game plays unpredictably, shutting down unexpectedly during Genshin Impact sessions.
Game plays unpredictably, shutting down unexpectedly during Genshin Impact sessions.
My machine occasionally shuts down unexpectedly while playing Genshin Impact. It doesn’t just restart Windows normally; the whole screen goes dark and then restarts completely (showing the MSI logo, booting into Windows). The fans keep spinning without stopping, so I don’t think it’s a simple hard reboot. System details: Microsoft Windows 11 Professional (x64), Build 22000.2538 (21H2), MSI X470 GAMING PLUS MAX (MS-7B79), AMD Ryzen 7 3700X, Gigabyte GeForce RTX 2060 OC 6G, 32GB Corsair VENGEANCE LPX at 3200MHz DDR4, Thermaltake Litepower 550W. Initially, I thought the GPU was the issue because it was getting quite hot (~75°C). Adding an extra fan and improving airflow didn’t significantly help. Using Afterburner to boost GPU fan speed to around 80% seems to reduce crashes. It’s worth noting that resets tend to occur more often after a recent reset, suggesting a thermal problem. I’ve replaced the GPU and RAM, but those changes didn’t resolve the issue. I’m also considering a new power supply (Corsair CV550), possibly because the current one is nearing end of warranty and I’m not confident in its reliability. Could there be another factor at play? Or should I focus on the GPU? I’m holding off on purchasing another GPU due to budget concerns. Thanks for your advice. P.S. I’ve logged data via HWiNFO over the network during a crash, showing a sharp drop to zero when it fails to save.**
I believe that temperature is within an acceptable range for GPU hotspots... However, it wasn't even near full load, drawing only about 100W to 120W. This might be higher than usual or typical. It doesn’t seem like a shutdown caused by overheating. Could be issues with RAM or CPU stability—have you performed any memory tests or stress tests? Also, please share some useful graphs. You went a bit too far in your setup; some measurements aren't reliable and shouldn't rely on software readings like 5V/12V rails. I have no idea about the power supply unit you're using, so it's unclear if it's functioning properly or not. For the Genshin energy usage, it would need to be extremely high to trigger these problems.
I also ran MemTest for some time. Experiment with various RAM timings to check stability across different configurations.
Thanks for the update. I've just completed additional tests. Ran Furmark and it crashes after just two minutes—this suggests the issue might be related to the GPU rather than the CPU. I'm planning to run memtest86 soon, so I'll share the findings then. The CPU cooler is a standard AMD model, and my GPU matches the product image from Amazon. The GPU fan reached up to 2996 RPM during data recording.
It seems your GPU might be experiencing instability. Running the fans at full speed while using Furmark could help. The issue may stem from outdated thermal paste or pads. A 2060 model should have sufficient power, so thermal management is likely the main concern. Good luck!
It seems you're likely correct—Furmark keeps crashing even when fans are at full speed. I'll consider swapping out the thermal paste soon. For now, I'm doing a memory test just to verify my RAM is functioning properly. Appreciate the helpful tips!
It appears the problem might be related to the GPU, especially given the high speed and power usage. The temperature and power settings in Furmark before the crash were likely extreme.
After careful consideration... the issue might still stem from PSU problems. I haven't completely eliminated GPU as a possibility yet. If RAM were the culprit, it would need to be extremely unreliable to cause such rapid crashes across the entire system... which would likely show up during regular tasks like web browsing. Instability usually leads to glitches or driver failures rather than complete restarts. It's rare for NVIDIA cards to trigger shutdowns due to instability, but I can't rule it out entirely. If the GPU is the cause, overheating and protective shutdowns seem plausible, or there could be a power delivery issue. The temperature might drop before any monitoring system registers it, making it hard to pinpoint. Have you ever disassembled, cleaned, or refurbished the card? Or is the PSU simply faulty, causing restarts during load spikes?
Review your earlier message and note that your PSU falls into the low-quality tier, so I’d recommend considering the alternative D-tier model—it’s substantially superior and should perform well as long as it functions properly. For systems with a dedicated GPU, aim for at least C-tier or higher; even Corsair units can work if they’re operational.