Video card causing power supply death or other way around?

Mar 19, 2018
3
0
10
Hey!

Tldr; Video card stopped giving signal during AC4 and a few days later the power supply went out. Now I wonder what caused what? And what do I begin to check?

Really unfortunate stuff has happened with my computer and now I am unsure how bad it is. It began when the video card stopped giving signal during gaming, mostly Assassins creed 4 – Black flag. I believe (though I don’t remember 100 %) it happened with another game at some point early on but after that only Ac4. In the middle of the game, the screen would go black, and then go searching for input and after a couple of seconds go to sleep. The only way I would get the monitor back on was to turn the computer off, using the power button, and, at least I think (and later concluded might be significant), turn off power to the computer. At first it happened very infrequently an in conjunction with specific actions in AC4 so I figured it was a bug with the game that somehow messed with graphics drivers. It sounds stupid now, but I didn’t think too hard into it.

However, after a few days, the issues started happening over and over again with AC4 and not in conjunction with the specific action. I tried turning all graphics options down as much as possible which did seem to prolong the amount of time I could play before the issue occurred, but no more than that. I checked the temperature of the GPU every minute or so until the issue occurred several times and it never reached the specified “during load” temperature, it was 61-62 degree Celsius at most. After googling other’s experiences with similar issues my take on it was that the graphics card was slowly dying and there was nothing I could do. I gave up on AC4 for now and used the computer for other games and surfing and no crashes happened. But 2 or 3 days later the computer randomly just died. Like there had a split-second power outage but nothing else had gone out. When I tried turning it on it turned on for a fraction of a second before dying. Now, nothing happens if I try to turn the computer on and I assume the power supply has gone to meet its ancestors.

So now I am trying to figure out, what came first? The video card issue? Or the power supply issue? Are they unrelated or did one cause the other? I hope that the video card was acting up due to power issues and is otherwise fine. But maybe the video card has been damaged due to the power issues? Or maybe it’s more likely that the power supply died due to a faulty video card eating too much power? The video card is less than two years old, but I bought it preowned, from someone who seemed reliable and stated it hadn’t been heavily used, but who knows. On the other hand, I reused the power supply from an earlier build, I believe it was at least five years old, and it wouldn’t be too surprising if it died on its own.

I could borrow a power supply from another build, but I am worried that the video card is a fault and will damage the new power supply too. What do I start with here? Any suggestions? Unfortunately, I don’t have the budget for a new graphics card at the moment, so buying a new one is out of the question currently.

Sorry that there is so much text. I wasn’t sure what was significant. Thanks in advance for any help!


Build specs:
Power supply: Antec 620 W: http://store.antec.com/highcurrentgamer/hcg-620m.html
Video card: MSI GTX 970: https://www.msi.com/Graphics-card/GTX-970-GAMING-4G.html
Motherboard: Asrock Z77 Extreme4: https://www.asrock.com/mb/Intel/Z77%20Extreme4/
CPU: Intel i5 3570k: https://ark.intel.com/products/65520/Intel-Core-i5-3570K-Processor-6M-Cache-up-to-3_80-GHz
RAM: Corsair Vengeance 8 gb DDR3: https://www.corsair.com/us/en/Categories/Products/Memory/Memory-High-Performance/Vengeance%C2%AE-%E2%80%94-8GB-Dual-Channel-DDR3-Memory-Kit/p/CMZ8GX3M2A1600C9
Hard drive: Seagate Firecuda 2 tb: https://www.seagate.com/gb/en/internal-hard-drives/hdd/firecuda/#3-5-inch-specs
 
Solution
Are you overclocking and/or changing voltages/timings?

The likelihood of a video card being that messed up that it could potentially cause the power supply to fail is next to none.
The info that you have given does not rule out CPU, MB, or Mem issues either. If overclocking, your OC could just be too high. Reset everything to default and try again. Also clean out all your fans and reapply your CPU cooler with new thermal paste.
Are you overclocking and/or changing voltages/timings?

The likelihood of a video card being that messed up that it could potentially cause the power supply to fail is next to none.
The info that you have given does not rule out CPU, MB, or Mem issues either. If overclocking, your OC could just be too high. Reset everything to default and try again. Also clean out all your fans and reapply your CPU cooler with new thermal paste.
 
Solution
Mar 19, 2018
3
0
10


Thanks for your reply! I appreciate it.
I haven't done any tinkering with the build, I actually have no experience with overclocking. Though, as I mentioned, I bought some parts used. Noob question, but could previous overclocking still be in effect?
Iirc the temperature for other components were good when I checked the temperaure for the GPU though I am not sure and will double check. Did dust cleaning 2 months ago and not much dust accumulated since then. I applied new thermal paste this summer so hopefully it's still good, if there is anything wrong with the CPU temp I will reapply it at once.
What aside from temps should I check for other components?

Edit: I booted up with another power supply: CPU idle temps around 30-35 and about 55 for loads of 85-95 %. CPU voltage: 0.856-1.166. MB temps: 30-40 except for TMPIN3 that spiked to 54 Celsius at some point. These temps were recorded during an hour of light usage, with the case open and ambient room temp of 21 C. Will make more testing tomorrow.
 
Since you have a second power supply (hopefully good enough to support your whole system at max draw?), it's time to do some test gaming.

A previous overclock could have damaged the components, yes. Sometimes motherboards, CPUs, video cards, and memory don't fail outright but slowly become more and more unstable over time due to damage from overclocking and overvolting.

If it's still unstable with the test power supply, and the test power supply is good enough, go 'borrow' a video card from your local store (e.g. confirm that it can be returned, buy it, return it after testing).
 
Mar 19, 2018
3
0
10


Update: I have tested quite a bit now and it seems the fault was with the power supply. No crashes have occured with the other power supply. I will invest in a new power supply and hopefully things will be fine. I have and will continue to keep and eye on the temps, they are somewhat higher than I would like on mb and cpu but still below what's considered problematic from what I can tell.

Thanks for your assitance, I appreciate it! Have a nice day!