Hi,
I'm currently having issues with the stability of a build using this board, which is temperature related and wanted to ask a few questions to see if it was the board that was faulty, or whether it's down to something else.
Spec is
Intel 6600K
Corsair Air 240 Case
Corsair RMX650 PSU
GA-Z170MX-Gaming 5
Corsair Vengeance 2x8GB 3000Mhz
Intel H100i GPU Cooler (front radiator push/pull intake 4x SP120)
Gigabyte Windforce 390X
1x Rear 120mm and 2x noctura 80mm
So having built this machine at the weekend I found I was getting issues with system freezes when under heavy load (GPU). I originally thought it was because the GPU was overheating, but under Furmark (before crashing) it maxes out around 82C which is on the high side but not enough to cause a crash.
So to test this I tried out another GPU, this time a 970 GTX, and even though this ran cooler, around 70C the same occurred and the system still hung after 2-3 mins of furmark. So I was guessing it was heat related. To test it out I took the side of the air 240 off and ran the benchmarks again. Ran Furmark for over 30 mins and all stable... so seemed to point to it being heat related.
After trying various different configurations of my fans it didnt make a difference and the same was happening. One thing i did start to do was look at the temperature of the motherboard. I started to notice a pattern that when the 00 and 01 temperature probes reached ~50c the system always crashed. When the side of the case was off the max value of these was in the high 40's around 47-48, and was stable as I mentioned.
To test my theory i added a bit of extra heat into the system whilst the case was open, right above the GFX and where the RAM is, and as predicted when the temperatures of those two sensors reached 50c the system crashed, meaning it only takes a couple of extra degrees to push the system over the edge, and as it's winter at the moment and the ambient temperature low, it's not allowing much room for any increases!
So as this has happened with two GFX cards it can't be down to the GPU overheating (infact I did some other gaming tests where the GPU was around 55-60 with the 970 for around 20 mins and it still crashed when the probes on the mobo reached 50C). The system is stable before the overheating, so I can't think it being an issue with the PSU or the CPU (which is liquid cooled anyway and on a full prime95 test for over 4 hours is around 45C under full load).
So that leaves either the RAM or the motherboard. My question is, what are these two probes measuring, I'm assuming it's around the GPU to get so high. And also is this an acceptable temperature for these probes, or is it crashing because it is indeed too high for the location and components on the board and I just need better cooling (i.e another case with more room). Or is it a fault with the board?
Thanks
