Computer seems to be half-way shutting off during gameplay it used to handle

[aborchert19]

Baseball card:

CPU: AMD FX-8320
GPU: NVIDIA GeForce GTX 960
MOTHERBOARD: ASRock 970DE3/U3S3
RAM: G Skill F3-12800CL9-4GBXL (x2)
POWER SUPPLY: Antec HCG-620M (620W)
STORAGE: Seagate ST1000DM000-9TS15E
OS: Widows 10

My computer isn't that old, but recently it's been shutting off seemingly everything except the fans while running games. I've noticed that it only shuts off while running graphically intensive games, so I tried turning the gpu fan up to 100% in open hardware monitor, but it didn't help. I'm thinking it could be the power supply? Don't those fail after a certain time? I'd like to know whether or not that's the issue before I invest in a new one if possible.

 

[Karadjgne]

Is the pc actually shitting down, or are you just getting a black screen, forcing a reboot to fix? Shutdowns can be psu related, motherboard related (overheating VRM's or chipsets usually), could be temp related or even a shorted wire can cause one. Blackscreens are usually about the gpu.

 

[aborchert19]

Is the pc actually shitting down, or are you just getting a black screen, forcing a reboot to fix? Shutdowns can be psu related, motherboard related (overheating VRM's or chipsets usually), could be temp related or even a shorted wire can cause one. Blackscreens are usually about the gpu.

I get a black screen, which made me question the GPU, but I also have a light-up keyboard that stops working during a shutdown, and sound also stops working.

 

[dudio]

Your PSU is respectable. What about temperatures?
Use HWInfo (free) and see what you see

 

[aborchert19]

Your PSU is respectable. What about temperatures?
Use HWInfo (free) and see what you see

I have OHM (lets me easily put gpu fan speed at 100%), Speedfan (has a handy chart function), and HWInfo (information overload) on right now and I'm currently playing. Here is the Speedfan chart to give an idea, in Celsius. The blue line is core temp.

https://imgur.com/a/7XMQh

 

[smorizio]

in the bios make sure the 12v rail is over 11.8v. with hardware info on logging and sensor watch the 12v rail.

 

[Karadjgne]

If those SpeedFan results are typical, I'd honestly believe it's a mobo issue. You seem to be hitting mid 80's on the gpu for extended periods of time. That means pushing the limits of the Northbridge chipset, which is now on the pcie controller. It sounds like the chipset is overheating and shutting down the mobo to prevent damage. Many of the old GeForce mobo's had that exact same issue, and similar results. It's an easy check, pull the side off the pc and blow a desk fan or mount a fan on the hdd cage with zip ties, blowing across that chipset heatsink. Then game as usual and see if you get the shutdown.