Question 7 Year Old PC Randomly Ceasing Operation

[tarias]

Hello! I have a desktop build that's been functioning properly for 7 years, but just recently has been experiencing random critical failures. Without any other symptoms, it will suddenly become unrecognized as an output to my monitor and has been observed to reconnect with the Windows boot screen 5-30 minutes after the initial disconnection. A manual power cycle of the system will make the PC boot as normal until the problem occurs again.

I've run a memory test using Windows Memory Diagnostic tool, and it didn't seem to output any memory error messages, but I'm not entirely sure how to interpret the Event Viewer messages. My GPU drivers are up-to-date. The issue persists when connected via a DVI or HDMI cable to another monitor.

Additionally, I've just recently (past 5 minutes as I'm writing this post) consistently been getting a black screen with a large system message saying "Please Power Down and Connect the PCIe Power Cable(s) for this Graphics Card" whenever I power the PC on.

What would be the best process to troubleshoot this issue?

Here are my PC specs:
CPU: AMD FX(tm)-8350 Eight-Core Processor 4.00 GHz
GPU: NVIDIA GeForce GTX 970
RAM: 12.0 GB DDR3
PSU: Rosewill Capstone-550 80 Plus Gold
Motherboard: GA-990FXA-UD3
OS: Windows 10 Home

Please let me know if any additional information is required!

 

[SpikeSpiegel]

I assume that PSU is part of the original build? I just had a similar issue with my 7 year old desktop and after some troubleshooting it turned out that my PSU needed replacing. Afterwards everything went back to normal. Just something to look into.

 

[tarias]

Yes, it is the original PSU! I will keep this as a potential on my radar.

 

[CANINE_RAPPAH]

Are you running your OS on an SSD or an HDD? Have you run any tests on your boot drive?

But from what I'm reading, this sounds like either a PSU issue or a failing GPU, but don't take my word for it.

 

[tarias]

I am running the OS on an SSD that was pretty full last I checked, had less than 10 GB available. I haven't run any tests on the boot drive. I can potentially check for that but I'm not sure now that it's giving me a perish message when I boot.

 

[tarias]

So the way it's lookin right now is I'll have to order a replacement PSU or GPU and see if that fixes the problem, I'm guessin. Good think I have a new PSU and GPU comin in for the build that'll replace this one. Does anyone think that the issue could potentially damage new components? Or should it be a safe troubleshooting procedure to test the system with theoretically functional new parts?