: Why does the computer keep crashing during VR gameplay?
: Why does the computer keep crashing during VR gameplay?
This occurs frequently when I play Half Life: Alyx, causing a loud clicking/popping noise from my headphones and turning off all my displays. The computer stays on but needs to be powered off and restarted manually. A red light also appears on the motherboard during these incidents. Initially, my RAM was 64GB at 6400MHz, which would cause the game to crash immediately. Lowering it to 4800MHz extended playtime slightly, but the crashes continued. Reducing it further to 3400MHz seemed to stabilize everything.
Later, a crash happened while playing a VR game. After restarting, I entered SteamVR and set up the room, but every time it crashed. Adding another 64GB of RAM resolved the issue temporarily.
Recently, when starting Half Life: Alyx, the computer crashed during the main menu. Removing all RAM and installing older 4x16GB sticks resolved it. Playing for a few minutes then closing the game and turning off the base stations helped. SteamVR remained active while I opened Google Chrome, which then crashed again during startup.
The problem seems to appear when changing or adjusting RAM speed, though it only resolves temporarily. This issue started after acquiring a VR headset. Anyone know what’s going on? Here are my other details.
GPU: RTX 4090
CPU: Ryzen 9 7950X
OS: Windows 11
My first concern is power issues. The PSU isn’t managing power demands well, especially during sudden spikes. I’m not familiar with the AGT1000 PSU, so I’ll have to rely on those who are... What happens during unexpected shutdowns? Files can get corrupted. Windows follows a set routine to prepare for the next start, but if it’s interrupted—like losing power—the problems escalate. More errors are likely, and they’ll vary. Check the Event Viewer—it’s not as user-friendly as Reliability History/Monitor and takes more time.
Get the complete details: motherboard, RAM, disk drives, power supply unit, and connected accessories.
A useful first step is to check Reliabililty Histor/Monitor for error messages, warnings, and events recorded right before or during the failures.
More errors are appearing, and different types of issues are emerging, which raises concerns about the power supply unit.
But additional information is still required.
I own three solid state drives, two of which are Sata and the main C drive is an NVMe. There are also two additional external drives connected through USB. On the setup, I have headphones, a mouse and keyboard, three monitors (24 inch, 42 inch, and 32 inch), and a VR headset. The motherboard is a B650 Aorus Elite AX with 64gb of DDR5 RAM (though I upgraded it to 128gb, which initially helped but later caused issues). The power supply is an AGT1000 rated at 1000 watts.
I checked Reliability Monitor, but it didn’t provide much useful info—it only mentioned that Windows wasn’t shut down properly and noted an unexpected system shutdown. This seems to be caused by me shutting it down manually because all my monitors went black and I couldn’t get them to show anything. I also have to turn off the power strip to restart it, as pressing the power button wasn’t effective at that stage.
I believe the analysis is correct. This appears to be a potential power supply issue. The 4090 model is quite demanding, while the AGT1000 seems to be a budget, average PSU. This situation likely points to a heat or power problem, as mentioned. If you're overclocking, pause and return to the default BIOS settings. It's best to aim for a stable baseline before making any adjustments. Another step would be to remove the extra monitors, as more displays increase pixel manipulation and heat generation. The goal remains reaching a stable baseline before proceeding further.
I'm evaluating options now, but the only other PSU available is a 10-year-old 450W unit. For a 4090, would you prefer a more powerful one or one recognized for better stability?
Other options should work fine. The quality matters more. What is your budget?
I also overlooked noting that my desktop is connected to a UPS. The UPS has a capacity of 1000 Watts.
Tried this alternative approach too, no need to rely on UPS. One of our moderators faced a comparable problem before. Constant troubleshooting with no success.
By chance, he took out the UPS and it worked!
UPS wasn’t providing enough power when the system was under full load.
Switched to a more capable unit.
I removed the UPS and power strip, disconnected the other monitors, reduced the resolution of my main monitor, and it still crashed. I reset the BIOS and re-enabled CSM support while disabling secure boot because login failed without it. I also remembered that my CPU was undervolted due to overheating, and I lowered my RAM speed from 4800MHz to 3400MHz. After resetting the BIOS, my RAM is back at 4800MHz and the CPU clock and voltage are normal. I plugged in my controller, turned on the base stations, and activated the controllers. SteamVR started up shortly after, but it crashed again soon after. It was worse since I wasn’t playing a game at the time. My motherboard had a red LED that hadn’t lit before, which I photographed (ignoring the camera flash glare). Do you know what this red LED indicates?