Unstable shutdowns following system upgrade.
Unstable shutdowns following system upgrade.
Hello, thank you very much for taking the time to read my post and support me through this challenge. I appreciate your patience as I work on finding a reliable answer. Before starting this topic, I conducted some research, hoping to discuss the matter openly and share practical ideas or recommendations. I believe many experienced members of this community have likely encountered similar issues, making it more likely that a solution exists. Recently, I upgraded my system with a new GPU and case while keeping the old parts intact. However, after the change, my computer began experiencing occasional crashes. During startup into Windows, there’s a brief freeze lasting a second or two. Occasionally, while browsing or performing everyday tasks, it shuts down completely—this has occurred three times in a month so far. Gaming sessions are equally problematic; the system freezes abruptly without warning, resembling a sudden power loss. The Windows event logs show a Kerenel-Power Error ID 41: “The system rebooted without a clean shutdown. This could stem from unresponsiveness, crashes, or unexpected power loss.” While I’ve looked up possible causes, it seems the power supply unit might be insufficient despite my belief that the upgraded GPU shouldn’t be the issue. I own this PSU for nearly two-thirds of its lifespan alongside previous components and have never had failures—until now. Monitoring temperatures confirms they’re within normal ranges (CPU 44°C, GPU 51°C, both stable during idle). During gaming, the GPU reaches 60–75°C, while CPU stays around 50–60°C. My PSU has been functioning well until this point. I noticed the PSU felt unusually warm and emitted a faint odor—neither hot nor burning, but definitely unusual. After attempting to restart and powering back on, it worked temporarily. This suggests either a PSU limitation or an overheating concern. I’m unsure if upgrading it would resolve the problem or if the issue lies elsewhere. Should I consider replacing the PSU with a Corsair RMx 1000W model? I’m not technically inclined, but I took the time to reassemble and clean my system carefully. Everything operates normally except for these sporadic crashes. My complete specifications are as follows: Case – NZXT H9 Elite; Cases – 4x 140mm fans, 4x 120mm; Motherboard – ASUS ROG STRIX X570-F; CPU – AMD Ryzen 9 5900X (12 cores); Cooler – Kraken X53 RGB; Thermal Paste – Arctic MX-4; RAM – G.SKILL Trident Z Neo RGB (32GB, 2x16GB); GPU – NVIDIA GeForce RTX 4080; Cables – CableMod C-Series Pro ModMesh 12VHPWR, 3x PCI-e; Monitors – 1x 1440p, 2x 1080p; Drives – Samsung SSDs (850ECO 250GB, 860ECO 250GB, 970ECO Plus 500GB); PSU – Corsair RMx Series RM850x; BIOS – Windows 11 Home 22H2. I’m still analyzing the situation and seeking a clear path forward.
Have you experimented with reducing your GPU's power settings? Trying around 80% can help assess stability. It’s possible the GPU’s power spikes sometimes, causing the PSU to activate safety limits. This might not be accurate, but it’s worth exploring how lower power usage affects overall system performance.
The anti-cheat tool operates through Windows-level programming which may lead to crashes, PC restarts, shutdowns, BSODs, or loss of monitor signals during multiplayer sessions. Even after closing a game, the anti-cheat can remain active in the background, potentially interfering with your next actions. Remove all recently played multiplayer games, restart the computer to confirm no anti-cheat is running, and only play offline games without anti-cheat features to test stability.
Shutdowns and restarts are very uncommon due to software issues, and I haven't encountered an AC cause. The problem usually lies with faulty OEM drivers. @xpto If temperatures are normal, the PSU is likely the culprit. Otherwise, any component could be responsible, though the GPU being new makes it a possibility. The motherboard might have a short circuit, so you could attempt breadboarding it. It's also worth checking for a power or reset cable issue on the case—try unplugging them and powering it back on by touching the power button pins with a metal object like a screwdriver. If none of these apply, it might be another component, which can be frustrating to troubleshoot.
You kept RDR2 operational and occasionally relied on the in-game benchmark to ease GPU strain. I capped the power usage at 80%, which has been stable so far, but I also experimented with 100% (default) and 105%. The power draw stayed within the allowed range, which is favorable. During gameplay, the TDP percentage remained near the limit, varying slightly based on rendering demands. No crashes have occurred yet, though this event seems unusual. I don’t think the GPU itself is defective—it’s functioning as expected. There’s nothing clearly suspicious about the new hardware upgrade. Here’s a snapshot showing the GPU’s performance during runtime. While the PSU felt warm and the fan ran, it appeared to be handling the load adequately. I’m puzzled by the sporadic crashes; there must be an underlying system issue. Previously, with Windows 11 and the old GPU, I experienced no problems. After a fresh Windows 11 install, I kept using the previous GPU. I used DDU to remove the old drivers and install the new ones for the updated GPU. I considered a full Windows reinstall to test, but I’m hesitant due to uncertainty. I don’t feel the need to update my BIOS unless it’s required for the AMD Ryzen 9 5900x support. I suspect my GPU is stable enough without changes. I lack experience with breadboarding, so I prefer avoiding potential mistakes. The RAMS appear normal, and speeds look correct. By the way, do you know of any tools that can display total system power consumption?