Thermal compound comparison review

[Mouse-Potato]

I haven't found anything comparing multiple thermal compounds on the LinusTechTips youtube channel.

 

I think it would be interesting for Slick and Linus to do a comparison about some of the most popular thermal compounds on the market.

 

Air and liquid coolers are only as performant and efficient as the thermal compound transferring the heat.

 

I was wondering if any of you would be interested in a benchmark comparing thermal compounds.

 

 

If so what thermal compound do you use and how good do you think it is?

 

 

 

I use Prolimatech PK-3 and I think it is one of the best.

 

[NeatSquidYT]

I use MX-2.

P.S. Usually doesn't matter between thermal pastes, like 5 degrees max afaik.

[Oshino Shinobu]

The results would be boring. There's little difference between different TIM. I use MX-4 personally. 

[Raudi_]

I just got some Antec Formula 7 its non conductive which is what you really want for GPU's and CPU's and it spreads pretty well, its definitely better then the AS5 I was using for my GPU's and CPU's but its all personal preference.

[Zeke]

#MX4masterrace

[Mouse-Potato]

I just got some Antec Formula 7 its non conductive which is what you really want for GPU's and CPU's and it spreads pretty well, its definitely better then the AS5 I was using for my GPU's and CPU's but its all personal preference.

I agree on the non conductive compounds, AS5 used to be some of the best but it's been a long time. I've never tried the Antec F-7, I wonder how it compares to other compounds though.

[Raudi_]

I agree on the non conductive compounds, AS5 used to be some of the best but it's been a long time. I've never tried the Antec F-7, I wonder how it compares to other compounds though.

Compared to other stuff its normally the best on charts but IC Diamond is the best out there still as far as pastes go.

[Toddwjp]

mx4, i mean its cheap and one of the best. 

[JamGorby]

It's been done. http://www.tomshardware.com/reviews/thermal-paste-heat-sink-heat-spreader,3600.html

[Mouse-Potato]

I use MX-2.

P.S. Usually doesn't matter between thermal pastes, like 5 degrees max afaik.

 

 

The results would be boring. There's little difference between different TIM. I use MX-4 personally. 

 

Yes but wouldn't 5°C make your fans run slower and make your system quieter?

[Gofspar]

NT-H1

[Mouse-Potato]

It's been done. http://www.tomshardware.com/reviews/thermal-paste-heat-sink-heat-spreader,3600.html

That Coolaboratory stuff looks pretty cool (no pun intended), has anyone had any experience with it?

[JamGorby]

That Coolaboratory stuff looks pretty cool (no pun intended), has anyone had any experience with it?

I'm....afraid....of it. lol 

[Mouse-Potato]

I'm....afraid....of it. lol 

I would be too, some people have complained that basically I "welds" your cpu to the cooler permanently..

[JamGorby]

I would be too, some people have complained that basically I "welds" your cpu to the cooler permanently..

 It's basically just a tube of gallium. It becomes liquid at around 30c. It eats aluminum and is very conductive. Not for the faint of heart. 

[Oshino Shinobu]

Yes but wouldn't 5°C make your fans run slower and make your system quieter?

Well, between 75C and 80C, maybe. Between 55C and 60C, no. 

 

Also, we're talking 5C between the best of the best and then mayonnaise and toothpaste.

[JamGorby]

snip

The main problem with this (or any) comparison is that the compounds aren't given time to cure and there's no follow-up on their longevity.

If Arctic Silver's claims of a cure time  for as5 are true, you'd have to consider that, for example. 

I have tried Gelid GC-Extreme because of this article and I can say the stuff gives great results, at first. 

But, after a few months it turns into dried poop and temps start to go up. 

[Mouse-Potato]

Well, between 75C and 80C, maybe. Between 55C and 60C, no. 

 

Also, we're talking 5C between the best of the best and then mayonnaise and toothpaste.

Well I was running 100°C TDP on my laptop and it kept thermal throttling in all my games so I took everything apart and cleaned the GPU and CPU and applied some high end TIM and basically dropped to 85°C max.

 

So yeah, 5°C in some cases but I don't know if I should take in consideration factory "thermal compound"...

[Mouse-Potato]

The main problem with this (or any) comparison is that the compounds aren't given time to cure and there's no follow-up on their longevity.

If Arctic Silver's claims of a cure time  for as5 are true, you'd have to consider that, for example. 

I have tried Gelid GC-Extreme because of this article and I can say the stuff gives great results, at first. 

But, after a few months it turns into dried poop and temps start to go up. 

 

Maybe Linus and his crew should run different types of TIM on all their rigs so that they can compare short and long term results over a certain period of time. They could use the TIM's with the best results and see how they fair in the long run.

 

Plus they have all their computers in the same room so the air temps are basically the same. I don't know if they're still running the whole room water cooling system or not.

 

But if they took the initial temps with the ambient air and the compared them to the long term test and corrected the ambient temperatures from before and after then we'd be able to get an accurate reading.

[Thony]

@Mouse-Potato They did this back when Linus worked at NCIX. No new revolutionary thermal solutions has been made since then.

It was few years ago and usually there is 1-3 degrees celsius difference.

[VSG]

Without a proper hotbox and multiple mounts, it makes no sense to even try.

One other metric no one really tests for is performance vs time. Some pastes degrade quickly in terms of viscosity, others dry out with time.

[Mouse-Potato]

I wonder if the video about changing out the TIM in GPU's was related to this thread?

 

It's just weird that we were talking about it and then a video pops up on the channel about it.

 

https://www.youtube.com/watch?v=qDLQ7FjPMf8&feature=youtu.be&a