Question All screens go to standby and system locks up during gaming

Apr 27, 2019
3
0
10
Hello everyone,
so I have the following problem with my PC. When I try to play a game - it doesnt matter what game, after some time (usually between 30 minutes and 2 hours) of playing all the sudden both my monitors go to standby with the "no signal" showing on screen and all audio stops. I do also know that the whole system is locked up at this point, as pressing the power button does not result in a shutdown as normal and the reset button doesnt do anything either. So the only way to get out of this state is to press the power button for a while until the powersupply shuts off all power. After a reboot everything works again.
Now this only happend after I needed to change my motherboard as I accidentally damaged my old gigabyte 990 fx ud 3. Now Im using a asrock 970 extreme 4.
So heres the rest of my systems specs:
CPU: AMD FX 8350 Black Edition
RAM: 2 * 8GB DDR3 Kingston Fury
GPU: PNY Nvidia Geforce GTX 970
Storage: 2 1TB HDD + 1 256 GB SATA SSD + 1 512 GB M.2 NVME SSD
PSU: Be quiet Power Zone 750W
OS: Windows 10 running on the nvme ssd.

So I think it might be a problem with some chipset on the mainboard overheating, as shortly before changing the mainboard I installed a custom water cooling loop. So both CPU and GPU dont get above 50°C with the CPU usually staying around 30°C and the GPU usually peeking at 40-45°C during gamplay or benchmark load.
I thought about the chipsets as since the water cooling is installed all the fans are running quite slow and both chipset heatsinks get so hot, that you cant touch them for more than a few seconds. Also my old mainboard had bigger heatsinks on the chipsets.
I dont have an IR-Thermometer so I cant give you a exact temperature reading for these heatsinks.
The big question is, is it really one of those chipsets overheating or is the problem caused by anything else? If its the chipset, which one? If not, what else could be the cause of the problem?
Of course if I had the money I would simply add the chipsets to my watercooling loop, but thats expensive and I dont want to spend that money if its not neccesary.
 
Apr 27, 2019
3
0
10
The mainboard is a Asrock 970 Extreme 4 and the PSU is a few years old I dont know the exact age, but I would say its between 3 and 8 years. The cooling loop is as follows: reservoir - pump - GPU - front radiator (2 * 120mm fans installed) - CPU - top radiator (2 * 120mm fans installed) - reservoir (again). But here you have a picture:
PC.jpg
 
Apr 27, 2019
3
0
10
So further testing revealed, that my guess with a thermal issue cannot be far off. When I turn all the case fans up to their maximum speed, it takes significantly longer for this problem to appear and sometimes it doesnt appear at all. I have reproduced this multiple times now. Using different games and different fan settings. It will always take longer for the problem to appear, when the fans are running at maximum speed.
So the questiuon that remains is: What part is overheating? Is it a chipset or something else? What chipset or other part exactly?

Any guesses for what part it would be?
I would guess maybe its north bridge? As according to wikipedia, north bridge is responsible for driving the pcie lanes and therefore communication with the graphics card. This would also make sense thermally, as it pretty much in the center of the pc and therefore furthest away from any case fans.