0

PC Specs:

  • Motherboard: Huananzhi X79
  • CPU: Intel Xeon E5 2670
  • GPU 1: Nvidia GTX 750 Ti
  • GPU 2: AMD Radeon RX 580
  • RAM: 32GB (four modules)
  • Storage 1: 512GB NVME SSD
  • Storage 2: 750GB HDD
  • Power supply: Thermaltake Smart 600 80 Plus White

For a while now, my PC shuts down and starts immediately without blue screen with some stressful tasks. When I add up the TDP of each component the sum is at least 200W below the PSU capacity. When gaming, the PC shuts down when rendering new scenarios or when many events happen at the same time. Also when running some Python programs that only use CPU but with a lot of calculations, it shuts down suddenly. I have checked the temperatures of all components, and they do not reach their limit temperature before shutting down, but the performance of the PC drops significantly.

It should be noted that the AMD RX 580 is a new purchase and these events had happened before I had it. But now that I can play more or games or the same games with higher graphics, it happens more often. For example, I could play GTA 5 with my GTX 750 Ti with the lowest graphics settings without problems, but now with the AMD RX 580 with almost the same graphics settings, my PC shuts down when loading the terrain. I don't know if the PSU is not enough or the power delivered to the CPU either or there is something I don't know.

PD: Also, the Python programs' events were before acquiring the RX 580.

EDIT: I already monitored the temperatures before PC shuts down, they are around 40-55 on CPU, 50-60 on RX 580, 32-40 GTX 750ti (celsius degrees).

EDIT: I used the UserBenchmark to look what's going on and PC turns off right after beggining the benchmark on RX580.

EDIT2: The NVIDIA GTX 750Ti is not displaying anything and I only use it for some AI test on python, while I am not using it, it consumes 0.6W. idk if take it out from the pc while make any difference to the suddenly shut downs.

5
  • 2
    Signs point to a CPU Thermal shutdown - monitor your CPU temperature to determine if this is the case Apr 7, 2023 at 3:24
  • 1
    It really sounds like either your CPU is overheating or your PSU is to blame.
    – Ramhound
    Apr 7, 2023 at 3:25
  • 2
    My vote would be for the PSU. Its possible that you have not exceeded 600 watts but still exceeded the power on one of the lines - ie the 3.3v + 5v line have a max of 105W, the -12 volt is 6 watt and the 5Vsb is 12.5 watt. It could also be that the caps on the PSU are going and it can't perform to its rated capacity.
    – davidgo
    Apr 7, 2023 at 3:47
  • 1
    The fact you can’t finish a benchmark indicates a PSU issue.
    – Ramhound
    Apr 7, 2023 at 11:54
  • I also would send the blame to the PSU. A significant power peak over the specs just a few cycles to much is enough to fail but the temperature will not exceed during that short time. — But the PSU also might fail to serve the official specs: — What's the age of the PSU? Older capacitors can fail (you can see the lid vaulting or some liquid leaking to the board). — This can happen to the PSU but also to the capacitors surrounding the CPU.
    – dodrg
    Apr 7, 2023 at 21:13

0

You must log in to answer this question.

Browse other questions tagged .