The LAST Thermal Paste you'll ever need??

[HimymCZe]

cut it and use on direct CPU die and/or GPU

[Zerim]

5 minutes ago, mariushm said:

These pads you guys tested are probably rebranded pads from some other manufacturer like Panasonic.

Yep. At $0.23/piece for a 30x30mm square, it could also be t-Global Technology. Panasonic lists one at 1850 W/mk, so I hope people can tell that W/mk isn't the number to look at--test results are.

[Professor Chaos]

So, umm is this better than Indium corporation Heat-Spring®? Could make another video in which you compare these two TIMs?
http://www.indium.com/thermal-interface-materials/heat-spring/

[Enderman]

Are you guys going to switch all your testing to these reusable thermal pads because they perform better?

Or stick to thermal paste because that's what 99.9% of people are actually using?

 

Also yeah, as other people have mentioned graphite thermal pads have existed for a few decades, I'm surprised it actually worked better than thermal paste because usually they are crap on anything that isn't perfectly flat. Nice to see it actually work.

 

Still worse than liquid metal though

[Navi Retlav]

I'm curious if it can be applied on delidded CPU and stuck on top of each other to not even use the CPU heat spreader at all ? 

At the end they should make a full heat spreader out of the nano-tubes that touches the core without any need for thermal paste at all even internally. 

[Progressor]

Man, you guys are stretching these videos like a motherfucker. Had to skip thought the video(felt more like a vlog with you rambling about random stuff rather than a tech video) to get to the interesting bits. Anyway, really interesting product, as others have already mentioned, it would be cool to compare it to other thermal pads(other good ones, not the shitty thick thermal tabs gpu and laptop makers put on memory and VRMs). Also try cutting this pad and use it on a delided cpu but without using the IHS, like bare chip > thermal pad > heater.

[aezakmi]

we all know the clickbait is there but

 

WHERE IS THE ARROW?

[redsquirrel0249]

I don't understand why people think this video is clickbait...

 

Yes, it's a pad. That's not the point of the video. The point is the application: you can use it like paste. Most pads can't do that. It's a change in functionality.

 

However, I feel like the product claims themselves are clickbaity. The stats read wayyy better than IC Diamond, but the test came out with negligible differences. What gives?

 

  Sold out of amazon, anyway.

[tikker]

Haven't watched the full video yet (scanned through it), but maybe a comparison with the liquid metal pads would also be nice? Different league of course, but still, now that the application hassle is out of the way maybe a value comparison.

http://www.coollaboratory.com/product/coollaboratory-liquid-metalpad/

 

[paddy-stone]

2 hours ago, Minibois said:

When installing a larger cooler, the case kind of acts like a third hand/arm that hold the motherboard so you can access both sides of the it more easily.

Depends on the case, on all of my cases I cannot do a cooler swap with the mobo inside the case, I don't have long enough arms as I like the dual chamber cases personally. Would need arms like an orangutan to do that with those and even then I doubt enough pressure could be maintained for some of the cooler designs to catch properly. 

[compromise]

Nothing new and fairly expensive. https://uk.rs-online.com/web/p/thermal-gap-pads/7124030/

[BluJay614]

3 hours ago, Minibois said:

It's not a paste, this is clickbait. Good day sir.

/s

 

Cool video, interesting product.

The first think I thought was "huh, seems like this is a small thermal pad if you're using Threadripper or similar size CPU's", just like Linus mentioned later.

 

Mine was more "I just bought a used laptop that I'll probably want to re-paste. Now there's this, and I don't have to worry about messing up the application."

[mariushm]

1 hour ago, redsquirrel0249 said:

Yes, it's a pad. That's not the point of the video. The point is the application: you can use it like paste. Most pads can't do that. It's a change in functionality.

But you can use most pads as paste replacement, they just don't work as well.  But then again you can also use toothpaste or margarine instead of thermal paste and they just won't work as well. You get the idea? The simple fact that it works BETTER than other pads, shouldn't classify this as a "paste" replacement, and it's not a change in functionality.

And some would argue that being conductive would make them worse than thermal pastes, and consider being non-conductive more of a feature than thermal conductivity.

 

For example there's thermal conductive pads with 12 W/mK thermal conductivity, which is better than some thermal pastes... that doesn't this worthy of being "paste replacement" : http://www.tglobaltechnology.com/datasheets/TG-X.pdf

 

It's silly, it's not a change in functionality.

 

Well, most of the time, you can, but sometimes a product is designed in such a way that it relies on thermal paste doing heat transfer and the properties of the pad don't work.

For example, that pad may be 0.5mm thick and the heatsink on a video card may be designed in such a way as to leave 0.75mm of space between chip and heatsink when the screws are completely closed - they may rely on automated machines to drop a big enough blob of thermal paste that would be squeezed and stretched across the chip when the heatsink is mounted.

 

Or another example ... consider the Vega56 and Vega64 cards with the VEGA chip and HBM memory on an interposer, and where you have something like 40um difference between vega silicon and hbm memory silicon and depending on where the chips are installed on an interposer you may have bare dies or you may have the dies covered in some molding. If you just use a flat graphite pad that doesn't have adhesive, the HBM memory may not be heatsinked at all... thermal paste is a requirement here:

 

 

If you replace with these thermal pads, unless the pad has some adhesive on one side, the pad could simply float in the space between the heatsink and gpu chip and even fall out, because it's not thick enough and the heatsink doesn't place any pressure.

And note the difference in height between the chips - so unless the manufacturer makes sure all the chips come from only one factory (that puts chip and hbm memory on interposer), they have to make the heatsink leave enough room (0.1mm or more) to account for differences between chip heights. 

So you come with your 0.25mm or 0.5 mm thermal pad and you may be surprised there's not enough pressure on the pad ... the paste can be applied in bigger quantity or thicker pads can compress better than graphite.

 

 

[asus killer]

50 minutes ago, compromise said:

Nothing new and fairly expensive. https://uk.rs-online.com/web/p/thermal-gap-pads/7124030/

that's gigantic dude, 18*11cms that covers all the motherboard 

 

 

 

could we use it to delid a CPU?

[redsquirrel0249]

@mariushm That's interesting. I'd never heard of pads being used as CPU thermal interfaces before.

 

As for the unmolded design thing, it seems like unmolded is just worse for thermals in general and doesn't seem to be an excuse for the low performance compared to IC Diamond according to the advertised specs. And I'm not even sure that applies to contact with an IHS and CPU cooler. They don't have variance in height on the IHS. If you're talking about the chips beneath it, it seems like that would affect both paste and pad...

[mariushm]

15 minutes ago, redsquirrel0249 said:

@mariushm If you're talking about the chips beneath it, it seems like that would affect both paste and pad...

for chips like vega56 :

[ravenshrike]

4 hours ago, Zerim said:

According to this paper, Dow makes thermal compound 4x better than Arctic Silver, and don't lie about it.

Um, no... just no. Properly designed and mounted heatsinks are effectively at contact mounted resistance, not bulk resistance. In which case the AS5 is better than the DC paste. That paper is for applications where there is a larger gap between surfaces.

[redsquirrel0249]

42 minutes ago, redsquirrel0249 said:

I'm not even sure that applies to contact with an IHS

@mariushm That's a Vega chip with HBM and GPU, not a CPU IHS. Different use case.

[AndPeroty]

How much it cost?
It is Currently unavailable. on amazon!

[AndPeroty]

5 hours ago, Enderman said:

Are you guys going to switch all your testing to these reusable thermal pads because they perform better?

 

 

It perform like a IC Diamond, 1ºC  worst actuly.

I prefer they use it over liquid metal!

[Ham the Terrible]

If it's only graphite, then it should be relatively inexpensive.  But as sheet material that thin (probably 1/64" or less) it will be pretty fragile and crack easy in a bend.  I have a ton of this material at work as scrap, used for mechanical seals.  The problem is that the stuff I have is 1/32" or larger.  I should try it on one of my junkers.

[maxichrome]

You should definitely try them on GPUs and delidded CPU die 

[AndPeroty]

39 minutes ago, donglefan said:

You should definitely try them on GPUs and delidded CPU die 

The problem with a delidded CPU may be that ir is conductive.

[Zerim]

3 hours ago, ravenshrike said:

Um, no... just no. Properly designed and mounted heatsinks are effectively at contact mounted resistance, not bulk resistance. In which case the AS5 is better than the DC paste. That paper is for applications where there is a larger gap between surfaces.

Do you have a link to some numbers? 0.0009" doesn't seem unreasonable, and at that scale the Arctic Silver is already twice the resistance of the Dow TC-5022. I'm seeing numbers from Arctic Silver that the average is closer to 0.003"-0.005". 

[AndPeroty]

9 hours ago, WereCat said:

RIP vertical mounting of the cooler

Did you apply thermal paste on vertical?