Fix: Unexpected System Restarts During Gameplay
Fix: Unexpected System Restarts During Gameplay
Hi All,
I built a desktop this summer which was working perfectly well until a few weeks ago. Out of the blue, it started restarted (seemingly at random) while playing games. This can happen as soon as the game's open, or it can happen after a few minutes, or it can happen after many hours of uninterrupted play; I really can't discern a pattern. The game - any game, as undemanding as Dota or as taxing as The Witcher 3 - will run seemingly without issue for some length of time, then suddenly freeze briefly, followed by a black screen and an immediate restart of the PC (no BSOD or any other error indicator). These restarts never occur while the system is idle or while I'm running non-game apps like Chrome, Office suite, etc.
EDIT: Since the Speccy link went down, here are my system specs again:
Windows 10 Pro 64-bit
Intel Core i5 4460 @ 3.20GHz
8.00GB Dual-Channel DDR3 @ 799MHz
MSI H81M-E34 (MS-7817) (SOCKET 0)
4095MB NVIDIA GeForce GTX 970 (MSI)
931GB Seagate ST1000DM003-1ER162 (SATA)
NVIDIA Virtual Audio Device (Wave Extensible) (WDM)
XFX TS Series P1550SXXB9 550W ATX12V 2.2 & ESP12V 2.91
I can't for the life of me figure out what's going on. All the components are brand new. The PC isn't getting especially hot - nowhere near maximum rec'd CPU or GPU temps, according to Speccy. This can happen even when the fans aren't especially loud and the game isn't doing anything much more taxing than running the main menu. I've felt the power supply (XFX TS Series P1550SXXB9 550W ATX12V 2.2 & ESP12V 2.91) immediately after crashes, and it's cool (or barely warm) to the touch. I plugged it directly into the wall to avoid any potential power strip issues, so no dice there. Still happens on any outlet around the house.
Despite the lack of indication that there's any overheating going on, I even thoroughly cleaned and replaced the thermal compound on my CPU. Still happens. I'm not overclocking.
All my drivers are thoroughly up to date. I ran Memtest from a bootable USB and found no RAM errors. I've searched for hours and read through hundreds of support threads around the web without finding any helpful advice. Weirdly, I ran all of 3D Mark's stress tests for hours and wasn't able to reproduce the crash under the most taxing GPU conditions - again, these restarts seem to happen at random while gaming regardless of how graphically intense or system-taxing a game is.
Windows Event viewer is showing these crashes as Event ID 41 crashes, source: Kernel-Power and reading "The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly."
So what in the world is going on? Any ideas would be greatly appreciated.
For anyone else searching for this information, I wanted to provide an update: the issue resolved itself after I reapplied thermal paste to the CPU and reinstalled the CPU, its heatsink, and the graphics card on the motherboard. It’s unclear which part was faulty, but I suspect one of them had simply come loose. It was a remarkably basic solution.
Frequent crashes that are difficult to reproduce often indicate problems with the power supply unit (PSU) or graphics card. I experienced comparable difficulties following the installation of a newly purchased GTX 980 Ti. Initially, I examined the PSU, which was operating correctly. However, after several weeks, the GPU ceased functioning entirely, necessitating a return for repair and replacement (RMA). If you can assess your PSU using a multimeter or swap out your 970 into another computer system, you might be able to determine the root cause.
Regarding Arossetti, as I lack a multimeter or alternative diagnostic tools to evaluate the 970’s functionality, are there any methods you suggest for testing the graphics card's performance beyond typical stress tests and benchmarking software?
Following the failure of the Speccy link, my system specifications are listed below:
Windows 10 Professional (64-bit)
Intel Core i5-4460 Processor – 3.20 GHz
8 GB DDR3 Dual-Channel Memory at 799 MHz
MSI H81M-E34 Motherboard (Model MS-7817) (Socket 0)
NVIDIA GeForce GTX 970 Graphics Card (MSI) - 4095 MB
Seagate ST1000DM003-1ER162 Hard Drive (SATA) – 931 GB
NVIDIA Virtual Audio Device (WaveExtensible Device Model – WDM)
XFX TS Series P1550SXXB9 Power Supply – 550W ATX12V 2.2 & ESP12V 2.91
Regarding catchthetide’s inquiry:
Arossetti responded that while he wasn’t aware of any straightforward method to evaluate the GPU independently without specialized equipment, it was a difficult task. A multimeter would cost approximately $50 USD, but even then, diagnosing the issue would remain challenging if the power supply unit (PSU) wasn't the root cause.
It seems necessary to examine each component individually. Based on your account – “...followed by a black screen and an immediate restart of the PC …” – the GPU appears to be the likely culprit.
If the PSU were delivering unstable or corrupted power, the system would typically simply shut down rather than freeze. I recommend contacting MSI to explore potential return authorization (RMA) procedures.
For anyone else searching for a solution to this old post, the problem resolved itself after I reapplied thermal paste to the CPU and reinstalled the CPU, its cooler, and the graphics card on the motherboard. I couldn’t determine the exact cause, but I suspect one of those parts was slightly disconnected. It was an unexpectedly basic issue.