Seeing that the weather is about to heat up, our computer heat dissipation pressure gradually increases, in addition to the pile fan, replace a better radiator, the CPU and the graphics card used for a period of time, may also need to consider the processing of replacing the thermal grease, otherwise the cooling fan is strong, can not quickly conduct heat to the heat sink is useless. But how to choose silicone grease? How different are they? This is the problem, today we will take a look at the relevant tests.
The first is to find a bunch of typical silicone greases that are more common on the market, from gray to gold, silver, from large bottles to small tubes, from an average of only 3 cents per gram to nearly 40 yuan per gram, and the thermal conductivity from 4.x to up to 14.x... has even added the legendary toothpaste that can be used as silicone grease. So, under such a huge parameter gap, how much difference does their actual effect really make?
The test method is a bit special, instead of using the CPU directly, it uses a simulated heating device to understand the thermal conductivity through the temperature gap between the upper and lower surfaces of the silicone grease. In order to better distinguish between actual capabilities and the size of the CPU, the top cover of an Intel LGA1200 CPU was flipped over, and the groove was filled with test grease, which was definitely much thicker than the thin layer on the surface of the CPU. Therefore, it should be noted that the temperature difference between the upper and lower surfaces of the silicone grease in actual use is certainly not as good as the test at 40 ° C.
So, can this test method reflect the real situation? Come to a few models of random inspection, the use of Core i7-12700K and high-end air-cooled radiator to try, to the CPU built-in temperature sensor given data prevail, you can see that its effect and the above test ranking, gap ratio is basically consistent.
Comparing the two tables, we will find that the price of silicone grease is a little tempting, but the effect is well, not only is it far from other products, but even worse than toothpaste, it seems that too greedy and cheap is really not OK. However, we can also see that among the several products with better results, in addition to the expensive models, there are also some models that are close to the people and are not very powerful. Here is also based on its reference price made a cost-effective chart, you can refer to the next.
This test also revealed a problem, that is, the nominal thermal conductivity is not necessarily reliable, such as the nominal thermal conductivity of the best performing products is even relatively low, and some of the models with high nominal thermal conductivity systems, although the performance can not be said to be poor, but compared with the price, parameters is a bit .... The suggestion given by Xiao is that in the case of a relatively reasonable price, the pursuit of a high coefficient of 10+ is no problem, but because of this very high price of the product, or look at the evaluation and then decide.