PC Shutting Down And Rebooting In Games

[sean2109]

This started happening to me about a month or 2 ago. At first it was rare and didn't bother me very much. However, at this point it is happening multiple times a day. My PC completely shuts off and then instantly restarts when I'm playing a game, intensive games like PUBG and Battlefield 1 in particular. No blue screen or anything, just a black screen and then a few seconds later I'm taken back to the windows login screen.

From my research I gathered that temperatures are a common cause of this problem. I am confident that my CPU temperatures are fine.My CPU ranges from 40-60 degrees in all games.
On the other hand, I'm not too sure about my GPU. It is a Sapphire RX 480. As soon as I launch a game the GPU rapidly rises in temperature until it meets the threshold (90 degrees C.) In some games it sits around 70- 80 degrees but I see it rising to above 85 in intensive games. Is this too hot? I've heard some people say 90 degrees is fine as long as it doesn't stay there for long, and I've heard other people say that the GPU should never reach 90 degrees. I have a custom fan setup in the Radeon Watman. I have set the fan to reach 2600 RPM when under load, and this did not seem to effect the temps at all. Is this whats causing my problems? I have had this card for about 2 years and this problem has only just really started over the last few weeks.

UPDATE: I'm pretty sure GPU temps are not the cause as I capped the temp to 85 degrees and my PC still shut down on PUBG.

Another thing worth noting is that my CPU (i5 6600k) was overclocked to 4.4ghz for about 4 months. the temps were fine so I assumed the OC was safe. The overclock was fine until I turned it off the other day in order to try and find out what was causing my PC to shut down. The shutdowns seem to happen less often without the OC, but my PC did shut down on PUBG today with default clock settings.

I'll list my specs

i5 6600k @ 3.5ghz (was overclocked for a while)
Cooler Master hyper 212 evo cooler
Gigabyte z170x Gaming 3 mobo
Sapphire RX 480 8GB
16gb DDR4 3000mhz RAM
Corsair CX750 PSU


I really want to fix this issue as it's starting to become really tedious. I want to play games with my friends but apparently my PC doesn't want me to have fun. Any help or advice is appreciated.

Thanks, Sean.

 

[Roland Of Gilead]

I was about to say it sounds like an unstable OC, until I saw that it happened at stock.

Radeons run a little hot anyway. As long as it's not extended periods at 90c then I wouldn't worry too much. But you can look at additional case fans for more airflow, maybe thermal paste on both CPU/GPU.

You could also look at changing the fan curve for the GPU so that when it hit's a certain temp the fan kicks in earlier and see if that reduces temps a bit. Obviously you have to be tolerant of the fan noise.

Is the GPU OC'ed, or was it only CPU?

 

[ElectrO_90]

I would like to ask, do you have a second hard disk or a spare one?
The reason I ask, is you could install windows 10 - on a second drive, boot up from that, install the game you want to play and this would prove if it was Hardware or Software.
New Windows 10 and if it still crashes, the chances are then, that it is hardware.
Finding hardware solutions is a pain, if you dont have spare parts laying around or a friend who has a computer you can swap parts over with.

 

[sean2109]

I was about to say it sounds like an unstable OC, until I saw that it happened at stock.

Radeons run a little hot anyway. As long as it's not extended periods at 90c then I wouldn't worry too much. But you can look at additional case fans for more airflow, maybe thermal paste on both CPU/GPU.

You could also look at changing the fan curve for the GPU so that when it hit's a certain temp the fan kicks in earlier and see if that reduces temps a bit. Obviously you have to be tolerant of the fan noise.

Is the GPU OC'ed, or was it only CPU?

I've never touched the GPU clock or VRAM speed. The only thing I messed with was the temperature settings.

 

[Roland Of Gilead]

I was about to say it sounds like an unstable OC, until I saw that it happened at stock.

Radeons run a little hot anyway. As long as it's not extended periods at 90c then I wouldn't worry too much. But you can look at additional case fans for more airflow, maybe thermal paste on both CPU/GPU.

You could also look at changing the fan curve for the GPU so that when it hit's a certain temp the fan kicks in earlier and see if that reduces temps a bit. Obviously you have to be tolerant of the fan noise.

Is the GPU OC'ed, or was it only CPU?

I've never touched the GPU clock or VRAM speed. The only thing I messed with was the temperature settings.

how did you manage to change the temp settings? What program did you use?

Reset whatever you've changed, keeping the OC off the CPU and test again.

It could be your PSU, I guess.

In the meantime, run a game and have HWMon/Info running along with it. Alt-tab out of the game and take a screenshot of the info for the GPU, so we can see max/min values for various usage, along with temps for everything.

 

[sean2109]

I was about to say it sounds like an unstable OC, until I saw that it happened at stock.

Radeons run a little hot anyway. As long as it's not extended periods at 90c then I wouldn't worry too much. But you can look at additional case fans for more airflow, maybe thermal paste on both CPU/GPU.

You could also look at changing the fan curve for the GPU so that when it hit's a certain temp the fan kicks in earlier and see if that reduces temps a bit. Obviously you have to be tolerant of the fan noise.

Is the GPU OC'ed, or was it only CPU?

I've never touched the GPU clock or VRAM speed. The only thing I messed with was the temperature settings.

how did you manage to change the temp settings? What program did you use?

Reset whatever you've changed, keeping the OC off the CPU and test again.

It could be your PSU, I guess.

In the meantime, run a game and have HWMon/Info running along with it. Alt-tab out of the game and take a screenshot of the info for the GPU, so we can see max/min values for various usage, along with temps for everything.

Using the Radeon software I changed the target temperature to 70 degrees and set the target fan speed to 2600 RPM. That was what I meant by temperature settings.

 

[sean2109]

I was about to say it sounds like an unstable OC, until I saw that it happened at stock.

Radeons run a little hot anyway. As long as it's not extended periods at 90c then I wouldn't worry too much. But you can look at additional case fans for more airflow, maybe thermal paste on both CPU/GPU.

You could also look at changing the fan curve for the GPU so that when it hit's a certain temp the fan kicks in earlier and see if that reduces temps a bit. Obviously you have to be tolerant of the fan noise.

Is the GPU OC'ed, or was it only CPU?

I've never touched the GPU clock or VRAM speed. The only thing I messed with was the temperature settings.

how did you manage to change the temp settings? What program did you use?

Reset whatever you've changed, keeping the OC off the CPU and test again.

It could be your PSU, I guess.

In the meantime, run a game and have HWMon/Info running along with it. Alt-tab out of the game and take a screenshot of the info for the GPU, so we can see max/min values for various usage, along with temps for everything.

I put everything back to stock settings including the Watman settings and took screenshots of HWMonitor while playing Battlefield 1. My computer didn't shut off but I only played for like 10 minutes. Sometimes it can go for a whole day without shutting off, but the next day it wont let me play at all.

The link below is 2 screenshots so I could show all the info. The only reason it says my GPU temp is currently 65 degrees is because I had to tab out to take a snip and the temperature dropped. My GPU was at about 80-85 the whole time I was playing. Also, the game seems to run significantly worse since I took off the CPU OC, so I hope I can put it back on at some point.

https://drive.google.com/open?id=1G8ET70KuPQIUdHYIMGVmMwkwrntvW3Vl

 

[Roland Of Gilead]

To be honest, everything looks exactly like it should. Temps, Voltages, power consumption are all where I would expect them to be.

It seems very random how it crashes. I guess it's possible it could be the PSU. Have you another one you could try or perhaps borrow from a friend to test?

 

[sean2109]

To be honest, everything looks exactly like it should. Temps, Voltages, power consumption are all where I would expect them to be.

It seems very random how it crashes. I guess it's possible it could be the PSU. Have you another one you could try or perhaps borrow from a friend to test?

Is there any other way to test if my PSU is faulty? I could swap it with a friend's if it came to it, but the cable management in my case is really tight and stressful. I don't want to take it all out if I don't have to. Thank's for all the help so far .

 

[Roland Of Gilead]

No worries. hmmm, i'm not an PSU expert at all. Only go on what I learn and read. I know you could prob measure at the wall socket with a digital tool, perhaps others could provide an accurate answer on that.

Other than swapping them out, I guess there's not much else other than buying a new one as a solution. But naturally I wouldn't advise that until you are pretty sure it is the PSU. Your PSU although 80+ Bronze has very mixed reviews, with some saying it doesn't output even 600w and it being a CX750. Still with your system your not going to be drawing more than 450w even with a CPU OC. So, although it possible its the PSU not supplying enough juice at specific heavy load situations, hence the crashing, it's likely in my mind it's the PSU thats the culprit. All the other sensor readings from HWMon look as they should, roughly speaking.