Power or gfx issue need advice.

B14cKs0L

Was playing X4 foundation one day and my system turned off. I had to unplug the psu for about 20 sec then it would boot up. I could still play less gfx heavy games. Then even games that I had no issue with like civilization 6 would cause a full power shut down. Also during load screens I would get a slight screen flicker.

Thinking it was a psu problem I bought a new psu but the issue is still there. On reboot my screen image will flicker. Can also notice it on dark screens but is fine while just in the web or watching video.

Now is it my gfx card or MB?

I also get this error

"Event 41 Kernal-power"

"The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly."

Giving the current state of things is not like I can just run out and get a new card.

or could it be the MB?

How to tell without testing every part in a friend's computer?

Ram tested fine.

No heat issues

clean dirt free system.

Bios and drivers updated.

No coming.

Ryzen 5 3600

Windows 10

Tuf b450 pro gaming

32 gig Ripjaw 3600 ddr4

MSI MPG Series, A650GF, 650 Watt, 80+ Gold Certified

Gtx 1060 6gig.

Thx fir any advice.

Vrika

If you have occasional flicker that does not crash the computer or result in any error messages, it's most likely GPU, monitor, or the connection between them since a problem anywhere earlier than GPU usually means a crash.

If your computer is occasionally crashing, then it's unlikely to be monitor or the connection between monitor and GPU since problems with those don't crash the computer.

I'd say that a problem with GPU is most likely cause.

But a situation like yours is always guesswork. Do you maybe have some old computer you could use for testing? If you have, you could try to test moving your GTX 1060 to that old computer and trying to see if you can reproduce the problem.

EDIT: If you don't have old computer for testing, you might want to try stress test programs that stress GPU and CPU separately. If the computer is ok during CPU stress test but crashes during GPU stress test, it's further indication that GPU is likely causing it.

B14cKs0L

New psu in place. Like the loading screen for civilization 6 I get image flicker and power only shuts off when game gfx load.

Vrika

B14cKs0L said:

New psu in place. Like the loading screen for civilization 6 I get image flicker and power only shuts off when game gfx load.

If it's consistent then that could also be something caused by power delivery, but since you already switched PSU and that graphic card takes most of its power from its own power connector I'd still guess that it's likely to be graphic card.

Though with computers it could also be anything else. The problem is that there's just no info. Even your error message "Event 41 Kernel-power" basically means that based on logs your system was running, and Windows has no idea why it stopped. It can't really be used to determine what went wrong.

Vrika

This is unlikely, but: If you're living in North and your problems coincided with cold weather, you might also want to check that you don't have your computer plugged behind same fuse as heater that uses a lot of electricity. Something like that could cause power problems, which could cause electronics to behave unreliably.

It's not very likely, it just came to my mind due to the season.

Ridelynn

First off, if anything is overclocked anywhere, put it all to stock. In my experience the number one cause of misbehaving hardware is an overclock; even if it's been working fine for years and never had any problem before.

I've seen spontaneous reboot problems be a lot of different causes in different machines - they can be very hard to track down.  It will help some. One thought would be to run a few different benchmarks that target different areas of the computer. Start with those benchmark programs one at a time, see if you can narrow it down using that.

Prime95 is a great test for the CPU, and tests the RAM pretty well as well. If it crashes or hangs on Prime95, it's probably one of those two components.

Furmark is "decent" at just the GPU. If your computer hangs or crashes during Furmark, odds are it's just in the GPU.

MemTest86 just tests RAM, and does a great job at it. If that crashes or hangs - the problem is probably just in your RAM.

I'd give each one at least a good 30min-hour to make sure it's stable

And outside or inclusive of these 3 programs, PSU, motherboard, and installation can always be suspect, so you just replaced the PSU, that probably isn't it. Loose wiring is remotely possible, but you've probably been over that already. I've seen something as benign as a loose screw stuck under the motherboard, shorting SATA power cables, and improperly mounted CPU heatsinks all cause the same problem as well -- all things to check.

Wouldn't hurt to pull your GPU out and clean it out really really well - it's amazing how much dirt can hide under that shroud, and it could be as simple as that.


It's also very remotely a bad or malfunctioning software issue. Rolling back GPU  drivers isn't a bad idea to try - if a driver update for the GPU suddenly starts it to reboot, an older driver may behave better... usually for something like the GPU it's just a crash or reset, not a reboot, but it's possible. The nuclear option is a clean install on a new drive that would eliminate most any software issue, but that isn't always the easiest thing to check.

A step of last resort that, if everything above fails to help point to a cause, is pull the computer completely apart and rebuilt it from scratch. That's how I've found the lost screw under the motherboard, or the shorting SATA cable, or some other similar hard-to-find problems. It also clears up loose connections (power cables, PCI cards, etc) and can clear corrosion on connectors if any has happened to build up over time.

Nanfoodle

When is the last time you replaced your thermal paste?