GIGABYTE X299 AORUS Gaming 3 Motherboard Review (Page 10)

Thermal Imaging and Power Consumption

System power is measured at the wall with an AC power meter.

Note on Thermal Images: In the temperature section, we use our Seek thermal imaging camera to capture the surface temperatures of major components on the board. I look at the VRM and then all other things that light up the screen. If there is something to worry about, then I will state it. Otherwise, I will just show the hotter running parts of the board for fun. Unless some component is over 80-90C, then there isn't anything to worry about.

All systems will act differently, so I will look for commonalities, such as how far from the VRM the heat spreads through the PCB and the difference in temperature between the front side and backside of the PCB. Keep in mind, the majority of the heat from the VRM goes into the PCB as it is a giant soldered on copper heat sink. A larger difference in temperature between the back and front of the PCB points towards a more effective heat sink.

Thermal Testing at Stock Speeds:

The image on the left is always at idle, and the image on the right is at load. During ALL TESTS, fans above the VRM that cool the CPU cooler's (Corsair H110i) radiator are turned on to high (12v).

Full frontal.

Up-close of the front of the VRM.

Up-close of the back of the VRM.

The VRM here is very solid regarding quality, but the cooling isn't anything magnificent. That being said, temperatures on the back and front are almost identical at stock, and they are actually low, which is great news for those who don't plan on pushing their HCC (high core count) CPUs to the maximum.

Low to moderate airflow 4.6GHz 1.75V VCCIN OCed VRM Thermal Imaging:

Temperature readings are taken after 40 loops of Intel Burn Test have been run (with AVX). Pictures of the setup are on the Test Setup Page. The two radiator fans (120mm Corsair) of the H110i blow in the general direction of the motherboard and VRM from the side (that is why the right side is slightly cooler in the first pic), so there is airflow (less than a case but more than a test bench with no airflow). The VRM here is cooled by a single heat sink, and there is no heat pipe.

There is also no backside cooling plate and not much surface area. The results here are not the worst and not the best. You should be fine with 4.5-4.6GHz with little airflow (like on an open test bench), anything higher and you will want to put the system in a case with decent airflow.