Issue arises during prolonged GPU usage New model experiences failures after extended GPU workloads
Issue arises during prolonged GPU usage New model experiences failures after extended GPU workloads
Building a new system with the 7800x3d Msi 4090 850w Corsair PSU Asus B650-A worked but struggles under heavy GPU loads. Crash occurs consistently around the 8-15 minute marks during testing. Using heaven benchmark confirmed high temperatures—CPU stays below 80°C, GPU under 70°C, while in benchmark. Swapping in an older GPU (1070) didn’t trigger a crash after 20 minutes. All PSU connections were reset and reconnected. Fast boot was disabled, RAM moved to alternate A2 and B2 slots. Looking for suggestions or solutions to this unusual problem.
I’d consider upgrading the power source. A 4090 can handle around 400W when under load, reaching 450W if pushed harder. There are also spikes that can double the GPU’s rated power for a short time, enough to cause the OCP of an underspecified power supply to activate. How does it actually happen? Is it just a way for the system to shut down from the application running on the GPU, or does the whole PC stop working?
The video display will become unresponsive for approximately one minute before displaying a blue screen error stating: DPC Watchdog Violation.
System began cycling during Windows boot. Reinstalled Windows without problems. When the video driver was detected, cycling occurred with a black screen lasting about a second, followed by another 5 seconds. System remained unresponsive but did not reboot. Disconnected power supply and GPU; tested GPU on another machine (Windows 10). Same issue—black screen and unresponsiveness when graphics drivers were installed.
4090's rank high in power-hungry cards... I wouldn't use it without a 1200W PSU, most think 1000W is enough—but I've seen some problems too. Right now I'm using a 1600W PSU anyway. The fewer the strain on your PSU, the longer it lasts and stays in good shape over time.