Complete shutdown (reboot?) when gaming

mjustind12

Commendable
Oct 3, 2016
4
0
1,510
First time post here, I went through the motions with manufacturers and even lurked here for a while and couldn't find a solid answer.
So I am not new to PC building, but I've had the most frustrating time with this new budget gaming build for myself. While gaming, within 5-10 minutes of gameplay (not even intense games) my pc will shut down. If i'm lucky it will reboot within a couple hours and be fine, if not and I press the power button to turn it back on my pc wont get past the Windows 10 splash screen. It will just show the rotating dots until I hard reset, then it will boot, but I will have to unplug and replug my network adapter back in to get wifi, and everything just seems way more sluggish than it was previously.

Specs:
CPU: AMD A10-7860K 3.6Ghz (not overclocked)
RAM: G.SKILL Sniper Series 8GB DDR3 2400MHz
Mobo: Gigabyte F2A68HM-H
1x WD 1TB HDD
1x WD 160GB HDD
PSU: Apevia Venus Power Series ATX VS450W

No discrete graphics, (had to RMA because of freezing) so i'm using the onboard R7 Graphics with up to date drivers (16.7.3)

Like I said before it will shut down for any game. I can watch youtube videos full screen HD just fine, it's literally just games. Not even intense games, it shuts down within 5 minutes in Roller Coaster Tycoon 2. (side note: I did get to play left 4 dead 2 for over an hour before it died once)

Temperatures are weird... every program I tried gives me a different reading. MSI Afterburner shows average -3C, HWPerformance shows about 70C at idle, and BIOS shows about 40C at idle so I don't know who to believe. Using the stock red cooler it came with, I turned off Cool n' Quiet thinking overheating was the issue, turned on full speed and nothing changed.

Sorry for the long post, Thanks.
 
Solution
Probably a damaged or just defective cpu chip, my A10-5600K stay alive until now with stock cooler and it is still functional even though it has 100W TDP and goes beyond 65°C on full load.
However, liquid cooling isn't the best solution at all, even though your tdp goes beyond 95 Watt. Hyper 212 evo still a better choice than a liquid cooler.

MCID47

Distinguished
*Temperatures are weird
Thats the point, in this case.
AMD's cpu temp seems icky to catch and its hard to find a perfect software for cpu core temp. The best software sometimes included on CD driver and sometimes it shows on UEFI Bios screen but wont shows up on 3rd party utility software like MSI afterburner and Speedfan.
 

mjustind12

Commendable
Oct 3, 2016
4
0
1,510


That's why its so hard to tell if I'm overheating or not. I wouldn't rule out my PSU either tbh. You think it could be a false alarm shut-down?
 

MCID47

Distinguished
I think so,,,
Easiest way you can do is touching the cpu heatsink. If it was too hot then it should be a cpu overheating. If not, then there must be a problem with your RAM or the Entire Motherboard. I dont think so if your psu falling apart, or even losing any power output.
 

mjustind12

Commendable
Oct 3, 2016
4
0
1,510
I ran memtest for 24 hours and found no errors, and I'm not entirely sure how to test the motherboard, but I am at least 87.82679% sure it's a overheating issue with my CPU. I'm sure it has something to do with the GPU kicking in and bringing the temps up. Back when I had a dedicated GPU it would just freeze. Removed that and now it's shutting down. Strange thing is that I've tried 2 different heat sinks (both stock coolers) thinking the problem was with the stock cooler it came with, and the problem persisted. I'm going to try the Cooler Master Hyper 212 Evo and see if that solves the problem. If not then maybe a defective chip? Buddy of mine that works at geek squad suggested Liquid Cooling for all APU's but I don't imagine EVERY person who bought this apu is having the same issue and requires Liquid cooling. I'll update if that solved the issue.
 

MCID47

Distinguished
Probably a damaged or just defective cpu chip, my A10-5600K stay alive until now with stock cooler and it is still functional even though it has 100W TDP and goes beyond 65°C on full load.
However, liquid cooling isn't the best solution at all, even though your tdp goes beyond 95 Watt. Hyper 212 evo still a better choice than a liquid cooler.
 
Solution

mjustind12

Commendable
Oct 3, 2016
4
0
1,510
So I found out that the only accurate reading I can get with an AMD APU is by using AMD Overdrive and reading the Thermal Margins. (Max Temp - Core Temp = Thermal Margin) And readings are looking normal from what I can tell. Thermal Margin is about 60C idle, and during Prime95 stress testing it reached about 40C. So ~10C Idle and ~30C Load. I understand that it only tests the CPU so the temps are going to generally be higher factoring in the integrated graphics. (I'd guess ~30C Idle and ~60C Load) Long story short, probably not an overheating issue. Most likely a faulty PSU. I have a 450W Apevia power supply that ill admit was cheap and I've accepted that I need to upgrade soon.