Computer Hard Crashes during certain games

Welshy_Rhys · December 16, 2023

Quote

Hi all, I've been having this issue for a few weeks now and thought I'd extend my plea out in hopes someone might know of a resolution.

Within the first 5 minutes of playing my system would shut down, like something tripped the power and start a reboot cycle. I first ran some preliminary tests on my hard drives, RAM, GPU, CPU and the OS in case it was a bad/corrupted file. Later to determine the RAM was faulty and had it replaced. The issue still persisted. I ran the game as an Admin, in compatibility mode, windowed, different DX level in the boot options, consulted their discord to get a "Our game is uniquely resource intensive". I've stress tested this PC with memtest64, prime95 and furmark constantly for hours, updated the BIOS to version 2801. But the game consistently powered off my machine. Without resolution I refunded and moved on. I booted up TF2, same issue. Removed my mastercomfig files and fresh re-install, same issue. A few months pass and I get invited into a Phasmophobia game and I'm getting the same issue again. I've run malwarebytes to find no malware, used a plethora of windows commands to get file integrity, tested my storage for issues. But it runs DOOM Eternal on high at 40fps flawlessly without hiccups, and runs stress tests consistently without any issue.

OS:

Windows 10 Home, 64bit

System Specs:

i7 4790K Quad-Core with a Corsair H105 CFM Liquid Cooler

Asus Z97-AR ATX LGA1150

G Skill Ripjaws Z 32GB (4x8GB) DDR3 1866MHz (Old Ram modules were Corsair Vengeance 16GB (2x8GB) DDR3-1600MHz CL9)

Western Digital WD_Black 2TB 2.5'' 7200RPM Interal HDD

Samsung 850 Evo 250GB 2.5'' SSD (Boot SSD)

EVGA Superclocked ACX 2.0 GTX-980 4GB

Corsair CXM 750W 80+ Bronze Semi-Modular ATX PSU

TP-Link TL-WDN4800 Wi-Fi Adapter

BIOS Version 2801 (Using the updater and the website this is the most up to date BIOS)

Software Culprits (Currently):

Black Mesa

Team Fortress 2

Phasmaphobia

There are no Error Messages or Screenshots to show. Event Viewer logs that it powered off unexpectedly but gave no information as to a root cause.

I have nothing overclocked on the system. I have changed the RAM frequency from "Default" to "DDR3 1866Mhz" and the problem persists on both settings and begun before I replaced the RAM.

I have a cheap PSU tester kit on the way. I have also been recording some of the crashes with OpenHardware Monitor and have a friend looking through the csv files currently.

This issue is now solved by a brand new PSU.

Edited March 10, 2024 by Welshy_Rhys
Solution

Welshy_Rhys · December 16, 2023

17 hours ago, TimedPing said:

I think this is PSU issues, regarding how old the system is if its the OG PSU when you first build the system, I have 2 identical system at the office with the same Corsair RM1200i PSU, and one of them fails with exact problem, and I took it for RMA. second one probably the thermal situation, can you check the CPU and GPU temp? even if its closed loop, old AIO can dry their liquid out.

Hello, thank you for taking the time to respond!

The thermals of the entire system under stress was completely fine. Nothing exceeded 75oC under complete stress. There were a couple sensors on the MOBO reading around 125oC but this fluctuated fairly randomly in spikes of 30-50oC at a time and it's suspected the sensors are bad. I also had the OpenHardwareMonitor open whilst I was gaming and it never exceeded 50oC (From what I could see, I'm hoping the csv files reflect this).

My PSU tester kit comes in a few days and I'm hoping it's just that with the age of the system, something had to give, maybe, fingers crossed.

Edited December 16, 2023 by Welshy_Rhys
Correction

Welshy_Rhys · December 17, 2023

On 12/17/2023 at 1:26 PM, TimedPing said:

can you also update the BIOS firmware, detach the dGPU, and run 24h stress test?

The BIOS firmware was already updated to it's most current version. I'm unsure what detaching the GPU and running a 24h stress test would tell us, as I've had the system run on long stress tests in the past.

EDIT 10/04/2024

The issue has been resolved by plugging in a new PSU. Through some random happenstance a friend had said it was likely due to a bad PSU wire or regulator (From what I remember).