@@superkoopatrooper4879 i've lapped my 3950x and waterblock, so i'm fairly familiar with how un-even it is, so i'd not be suprised other metals would be hit and miss in that regard If someone was to try gold, copper or any other metal sheet, it surely would not be me!
Silver has higher thermal conductivity at around 429 W/m·K compared to gold at around 318 W/m·K. Graphene has conductivity in range of 4800-5300 W/m·K .
NZXT also has copper pads on their AIOs. Danger of mixing some metals with an electric current involved, you get electrolysis and the metals transfers and causes issues.
Aren't some coolers actually solid copper with nickel plating to even out the surfaces because of how soft it is? And to reduce the chance of conduction of course lol.
I bought Arctic MX-4 2019 a year ago when it had a contest win $500. Of course I had a tube already but it was 5 years old so the contest just convinced me to get a fresh tube. Donated the 3/4 full older tube to Free Geek, they do a lot of computer building. Doubt I will ever buy anything else.
@@mastaw the gaps are nanometric and they exist in almost any surface you can study. Its unlikely that any solid can fill those gaps at all. Due to the very bad thermal conductivity of any gas in comparison with a solid materials, even a little bit of air can turn into a problem for the heat dissipation. In W/m.k, Al ~ 204, Cu~372 and air ~0,023.
I think you could try this again with a lapped CPU/cooler. Also I think you should be using a higher wattage CPU since the differences show better at higher heatloads. E.g. If paste A performs better than paste B by 2C at 100W, that might be 4C better at 200w as a rough example. Great work nonetheless, hope to see more on this soon!
And the all forgotten rule is, higher pressure on the sheets work better on transferring heat. Gamer Nexus have shown this to be a legit method to increase the heat transfers with sheets as gap filling.
@4:46 No, its not even close. you'd be lucky to even cover 1/4 of the motherboard with that little paste. Not to mention, what about the back of the motherboard?
Right now I'm seeing these available for $15 on amazon with free shipping. That's not too much more than regular thermal paste and this is reusable... seems like pretty good value honestly. Thanks Hardware Canucks!
What about the 0.15 mm sheet? It's 3/4 of the thickness, which means thermal resistance should be 3/4th of the 0.2mm sheet. This could very well improve the results by a few more degrees here!
It didnt, we tried. It actually was worse. It was cut out of the video for length ( might be in a follow up ) but at .15 it seems there isn't enough pressure to push it through the surface imperfections. Making it thicker means bottoming out the cooler gives you slightly higher mounting pressure. enough for a 2-4c difference.
@@Salty.Peasants That's that I want to know. IIRC the carbon sheets don't quite perform as good as paste, meaning this is better. But it looks like this is on par with paste, meaning the conductonaught liquid metal will still be a little better. But this obviously has a huge advantage with clean up and reusability..
@@SnowsLife If you're up for messing around, it would be interesting to see if you could double stack the 0.15mm or even the 0.20mm sheets... I mean, you could just buy thicker ones for a proper test, but it could be interesting ;-)
I'd like to see snows do some hardware cunucks episodes doing review stuff. He's great on boot sequence for the news too and his beats are flipping amazing! Also 1-2 hour min is required for the burn in test to goto steady state. So I'd like to see that be done in the future instead of 15-45 min tests. Also plot the Temps from start to finish to see how they deal with low temp and high temp between the different substrates. Aluminum, carbon nano, copper paste, indium etc all good tests I'd like to see, even though I've seen them all before. It's going to be some what different between amd with its Chiplets and Intel with its single chip package. So maybe a comparison between the two for the top two winners.
One more thing while I'm here, aluminum foil is a household item that is extraordinarily uniform in thickness with deviations in the range of a couple nanometers across the full width of a sheet. Aluminul foil will not introduce ANY imperfections (air pockets) but will only expose those already there.
This video was an incredible surprise. I'm seriously considering trying this stuff out. Thanks guys for the great content. More stuff like this please!
Yeah, we need to integrate the metric system more in the US. It's just better. A full transition would take at least a generation, but we can at least teach both.
@@SirSethery It would probably takes a lot longer than that, even in Canada, both systems are used about as much, and in a wierd way too, by exemple, we'll swim in a pool if it's 25C+ outside, and the pool is at least 70F... And I'll cook chicken at 425F, until the internal temperature is 65C...
reliability over time- does the indium give consistent results over time with lots of heating/cooling cycles? It might even get better after a few burn-ins for all we know
This is extremely interesting to me. Even if it's expensive it's actually quite practical since it's reusable (in theory? Please test how good it is when reused!). Basically even if you upgrade your CPU, barring any unforeseen circumstances like you accidentally rip it while taking it off, you can just reuse the same sheet. Please keep making content like this heheh.
Definitely would like to see a roundup using this stuff. Like different thicknesses, different CPUs, to see how it handles higher vs. lower wattages and even how it works out on a larger die like on Threadripper. Super interesting vid though, would definitely like to see more!
You can buy high pure graphite sheets. You can buy Graphene sheets these days. You can buy Silver sheets. Make part 2 with 5 kinds of sheets, and lets destroy the paste industry! (Paste industry really hates sheets of any kind)
Love this format, it is a much more interesting way to present longform content than how Gamer's Nexus does it. Only question is what happened to testing the 0.15mm Indium pad?
Was thinking long term, 24-7/365. never dries out, or cracks, just helps move heat. Sure as mentioned above it might weld it's self to a cpu, But I've pulled AMD cpus out that had stuck to the heatsink and wouldn't let loose . If you need to service it, it probably needs more than fresh paste .
@@OH-vb6ww true, no worries about dried up thermal paste. There are some server applications where this would probably be ideal. Hopefully no other motherboard components fail though, nor no CPU upgrades needed, then it would be an excellent server solution.
Anti-seize, for car bolts is an EXCELLENT TIM. I've used it for years on PCs with better results than thermal paste. For those of you that don't know, anti-seize is a compound made from fine metal shavings (usually aluminum or copper) and oil. You must treat it like liquid metal because I believe it is electrically conductive. But it's well-worth the performance gain over traditional thermal paste. I saw anywhere from 5-8*C improvement. Try it out.
Fruits may not always be the best way to lose weight, especially if you get the sweeter fruits. I'm sure he's done his due diligence though! Great vid, regardless!
Hey Guys I loved the video. It would be amazing if you tried more of unknown or not so famous stuff. I always appreciate the a different thing to watch even if it is underperforming or if it is somehow expensive (like the Indium Sheets). In any case, it is always nice to see different content.
This proves again that the industry put the concave and the convex surfaces there to sell PASTE! Remember CPU lapping? Yes, Lapping CPU and Heatsink will give you the best results.
In researching Gallium Nitride devices, i have read that using predominantly diamond based cooling paste/pads that it lowers 400C junction temperatures quite well so I’d love to see how well this could work if it was even doable.
It works great but can get very expensive. IC Diamond is better than Kryonaut and pretty close to being as good as Conuctonaut. Problem is that it's very difficult to apply and the diamonds destroy ihs' unless you use a lot of cleaner and acetone while being super careful not to rub or wipe with any pressure, I use a brush. Now back to cost, you may have seen Kryonaut LHe / Kryonaut Extreme Pink pastes on Der8auers YT channel. The reason they perform better and cost more is because the size of the particles used (TGrizz secret blend, not diamond) are smaller and it becomes more difficult to actually make and to mix evenly + keep dispersed and suspended in place instead of working their way downwards from gravity. This is why IC Diamond looks and smells like tar and has larger particles that scratch any surface they touch
There is little point. It's going to be ballpark to any decent paste apparently, but it's unlikely to be identical to any. There is very little difference between the top pastes.
What about long term exposure? Galvanic Corrosion concerns? Also, testing with multiple thicknesses for other materials you try would be appreciated. Definitely interested in seeing other materials!
Due to indium's broader uses I'd like to see how it compares in an LN2 XOC situation when compared to something like Kryonaut. I would also be weary of using it against any copper surface from indium's propensity to diffuse into copper.
tbh, you can hypothize that metals with that high thermal conductivity are in the same temperature. That being said, Indium melting point is ~157°C. Its unlikely that any CPU can achieve that temperature. The problems that would need some testing would be thermal fatigue and creep phenomena. My guess would be no welding
@@myw0rk I don't mean actually welding... But even at submelting points, the soft metal will warp into those tiny deformations on the surfaces and hold them together like glue... So less like a piece of solder.. More like a piece of velcro??
@@vivekvs1992 I see. I guess that could happen if there is creep happening. Anyway, its not a few minutes testing that can tell if these things are doable or not. This is a nice subject for a video but a shitty idea and well, its not worth the trouble, either for air gaps or other issues that wouldnt happen with a paste.
I really like the trio at Hardware Canucks! Eber does testing, and have the best voice! Dimitry does peripherals, and has the best humor (and hair!) Mike does experiments, and has the most approachable presentation! Keep this up, it's great to watch all three of you (and Snows is great too ;)
Just the science behind this is really (pardon the pun) cool. I'd love to see more videos like this. It makes me wonder what else can be used like this tbh. Like, what would happen if you put .2 mm of high purity copper? Might be expensive, but interesting!
Think about it this is actually very good! Thermal paste starts to dry up after a year or so, so therefore these sheets will be alot more reliable and stable for workstations and alike!
The theory behind the medium-low melting point allowing it to soften (but not melt) at high loads and imprint on the imperfections on the IHS and cold-plate is definitely an interesting idea.
As for your question of should we test more products, I would vote yes. I actually have a couple small sheets of something similar or the same from a few years back and I may just give it a try on my GPU. Now if there was a good replacement for those pesky silicone pads that would be something.
I have an IC thermal carbon thermal pad myself. I used it on my main system for about a year before going back to a paste. Only because I was just getting center point contact from the cpu and cooler. :/ But I use that thermal pad to run tests all the time. It has saved me soo much money in thermal paste. I used to go through tube after tube. I now get just one tube every couple months. I might spring for this and try it out. Maybe the little extra thickness over the carbon pad will make the difference. thanks
Our boi got thicc! Snows, I just lost all my COVID weight. I started having a small bowl of cereal and maybe a banana and definitely milk for dinner. I have also increased my activity levels (gym and biking). I’ve lost 22 pounds (10 kilos?) since January.
The problem with aluminum sheets, it's that by the time you apply it, it's no longer pure aluminum. It will have a layer of aluminum oxide that have a way lower thermal conductivity. Aluminum oxidize extremely easy with the oxygen in the air.
Been using gallium/indium on my gear for six years. Granted, it is liquid. Started with CLU, but it had a hardening problem that isn't as bad with Conductonaut. I'd think think that these sheets wouldn't fill the little pits and crevices that LM or even paste will unless it is really soft. Oh, and maybe "heavy duty" aluminum foil would be the better choice rather than that discount stuff because the indium was 200 micrometers thick.
I definitely would enjoy a more in depth comparison between same metals with different thicknesses, and also different metals altogether. I really believe this study could change how we handle building a PC from a maintenance part because as you said : when you build a PC for a friend, you do not want any maintenance tasks to happen, and that is basically the remedy to it. I also think a review using these for some time could be interesting, because I know that some metals can react with other, like the IHS of the CPU for instance, so this could also be interesting to see in the long term.
I think that the main advantage is it doesn't dry out. Ever had an old tube of thermal paste and saw some oil come out of it with the paste? That oil is what likely dries out leaving that disgusting chalky and now mediocre paste behind. When you buy a graphics card odds are that you have dried up paste after a while if it is not already dried. If this foil is on par with a good paste in cooling performance that is actually worth considering. Now if you ever take the cooler off you can't reuse it. You should not reuse old paste either other than an application fresh from the tube.
I think that aluminum foil was full of oils at that point, but hey, not a bad result. The performance of the Indium was impressive. Paste is easier to use, I should think, but it dries up, and removal requires alcohol. A sheet of metal alloy like that won't dry and won't stick too much. Really, though, it's rather shabby of Intel to not make proper heat spreaders. I saw another video where sanding it even cut the temperature with like 5 degrees. That's crazy.
I love me some experimentation. Maybe check them pads or other thermal solutions with a notebook as well? I remember Linus doing that with liquid metal and it was pretty sweet. So why not with some other materials? Who wouldn't want improved cooling for their notebook if it's easy to apply (like a pad). xD
One note about indium, it would work much better if it had higher pressure, but pressure from common cpu coolers is not enough to get the most of it. So you should try to find a version with holes in it, that works better apparently since it gets higher pressure (lower area). That would be an intersting test, but I am very glad that you tested indium in first place :)
I would love to see this tested, including the different thicknesses. As for removing the sheet afterwards, that would make for a nice set of experiments too. The razor works, but what about dental floss? Sewing thread? I'm leaning towards some kind of thread being a safter method of doing it.
This also makes me wonder something: at which point does the IHS becomes the bottleneck in the whole process? After all, it's way more thick than the indium sheet or even the layer of paste after it's squished down
I would like to see longer term test with the indium on both nickel plated and bare copper base plates to see if there’s any chemical reaction between the dissimilar metals like how galinstan liquid metal TIM reacts into copper and gets absorbed and etches nickel plating.
They got it completely backwards on where the pressure was on that indium. The pressure was on the outer circle, and the patch in the middle was less pressure.
Really interesting to see the comparison between different cooling solutions and different TDP cpu's. Cuz by the looks of things there's no reason to further use thermal paste unless the IHS or the cooler has a dank shape.
Would be nice to see the results of indium an other materials on heatsinks that have direct pipes contact. The gaps in between will maybe act against due to air pockets?
Cool stuff. Id go with the paste still. Thou in my mind the pad would cover the whole area better without it leaking. I seen some horror story's of liquid metal. It drying up making your cpu heat up. but I bet most peoples applications are doing just fine. Whens that MX-5 coming out.? gotta see the results for that stuff.
I want to see the difference Lapping the CPU and Heatsink would make using this same testing methodology. And like others would like to see Copper, Gold, and Silver. Or a Copper Indium alloy; Copper Gold alloy; Gold Silver Copper Indium Alloy?
The Verge - the gift that keeps on giving.
These guys look like they're from Verge with those gay masks on.
@@mrniusi11 Unfortunately, while I agree with you on the whole mask thing, Canada is a fucking police state right now.
@@thegirthquake8574 should have shot back when they banned guns :/
@@mrniusi11 oh man I didn't know masks had a sexuality attached to them. I hope you don't get sick :)
you know, Copper is soft...i'd love to see how a extremely thin copper sheet could work as TIM, or even gold!
@@superkoopatrooper4879 i've lapped my 3950x and waterblock, so i'm fairly familiar with how un-even it is, so i'd not be suprised other metals would be hit and miss in that regard
If someone was to try gold, copper or any other metal sheet, it surely would not be me!
Silver has higher thermal conductivity at around 429 W/m·K compared to gold at around 318 W/m·K. Graphene has conductivity in range of 4800-5300 W/m·K .
NZXT also has copper pads on their AIOs. Danger of mixing some metals with an electric current involved, you get electrolysis and the metals transfers and causes issues.
Test the indium sheet of different thickness and also on higher TDP cpu and GPUs ❤️
Aren't some coolers actually solid copper with nickel plating to even out the surfaces because of how soft it is? And to reduce the chance of conduction of course lol.
"It's way thinner, seriously Mike?"
"The things I start pulling out over here"
"I'll be seriously impressed if it just beats rawdogging it."
Something about thermal paste just feels right to me
Well, thermal paste is a liquid
And its job is to fill gaps, so a liquid feels like it would fill more gaps then a solid
I bought Arctic MX-4 2019 a year ago when it had a contest win $500. Of course I had a tube already but it was 5 years old so the contest just convinced me to get a fresh tube.
Donated the 3/4 full older tube to Free Geek, they do a lot of computer building. Doubt I will ever buy anything else.
@@elinor8569 But both expand under heat, so that should be a non-issue
@@mastaw I was trying to explain why it "feels right" rather than how it actually works
@@mastaw the gaps are nanometric and they exist in almost any surface you can study. Its unlikely that any solid can fill those gaps at all.
Due to the very bad thermal conductivity of any gas in comparison with a solid materials, even a little bit of air can turn into a problem for the heat dissipation. In W/m.k, Al ~ 204, Cu~372 and air ~0,023.
I think you could try this again with a lapped CPU/cooler. Also I think you should be using a higher wattage CPU since the differences show better at higher heatloads. E.g. If paste A performs better than paste B by 2C at 100W, that might be 4C better at 200w as a rough example. Great work nonetheless, hope to see more on this soon!
And the all forgotten rule is, higher pressure on the sheets work better on transferring heat.
Gamer Nexus have shown this to be a legit method to increase the heat transfers with sheets as gap filling.
@4:46 No, its not even close. you'd be lucky to even cover 1/4 of the motherboard with that little paste. Not to mention, what about the back of the motherboard?
Right now I'm seeing these available for $15 on amazon with free shipping. That's not too much more than regular thermal paste and this is reusable... seems like pretty good value honestly. Thanks Hardware Canucks!
What about the 0.15 mm sheet? It's 3/4 of the thickness, which means thermal resistance should be 3/4th of the 0.2mm sheet. This could very well improve the results by a few more degrees here!
only if it doesn't have any voids in it from lack of depth to seal the contact.
It didnt, we tried. It actually was worse. It was cut out of the video for length ( might be in a follow up ) but at .15 it seems there isn't enough pressure to push it through the surface imperfections. Making it thicker means bottoming out the cooler gives you slightly higher mounting pressure. enough for a 2-4c difference.
@@SnowsLife How is this, vs thermal grizzly conductonaut? How is it for long term use, and are there any metal reactions I should be worried about?
@@Salty.Peasants That's that I want to know. IIRC the carbon sheets don't quite perform as good as paste, meaning this is better. But it looks like this is on par with paste, meaning the conductonaught liquid metal will still be a little better. But this obviously has a huge advantage with clean up and reusability..
@@SnowsLife If you're up for messing around, it would be interesting to see if you could double stack the 0.15mm or even the 0.20mm sheets... I mean, you could just buy thicker ones for a proper test, but it could be interesting ;-)
"Why are you calling me on my day off"
This is one of the most interesting PC videos I've seen in some time!
I'd like to see snows do some hardware cunucks episodes doing review stuff. He's great on boot sequence for the news too and his beats are flipping amazing!
Also 1-2 hour min is required for the burn in test to goto steady state. So I'd like to see that be done in the future instead of 15-45 min tests. Also plot the Temps from start to finish to see how they deal with low temp and high temp between the different substrates.
Aluminum, carbon nano, copper paste, indium etc all good tests I'd like to see, even though I've seen them all before. It's going to be some what different between amd with its Chiplets and Intel with its single chip package. So maybe a comparison between the two for the top two winners.
One more thing while I'm here, aluminum foil is a household item that is extraordinarily uniform in thickness with deviations in the range of a couple nanometers across the full width of a sheet. Aluminul foil will not introduce ANY imperfections (air pockets) but will only expose those already there.
This video was an incredible surprise. I'm seriously considering trying this stuff out. Thanks guys for the great content. More stuff like this please!
i would love more content like this
I love both your guys' videos. So much so, that it makes me hate that they teach us Farenheit instead of Celsius in the US still.
HAHA!
Yeah, we need to integrate the metric system more in the US. It's just better. A full transition would take at least a generation, but we can at least teach both.
@@SirSethery It would probably takes a lot longer than that, even in Canada, both systems are used about as much, and in a wierd way too, by exemple, we'll swim in a pool if it's 25C+ outside, and the pool is at least 70F... And I'll cook chicken at 425F, until the internal temperature is 65C...
reliability over time- does the indium give consistent results over time with lots of heating/cooling cycles? It might even get better after a few burn-ins for all we know
It probably gets better as it is soft with a low melting point, that could make it fill the gaps as time passes.
This is extremely interesting to me. Even if it's expensive it's actually quite practical since it's reusable (in theory? Please test how good it is when reused!). Basically even if you upgrade your CPU, barring any unforeseen circumstances like you accidentally rip it while taking it off, you can just reuse the same sheet.
Please keep making content like this heheh.
Definitely would like to see a roundup using this stuff. Like different thicknesses, different CPUs, to see how it handles higher vs. lower wattages and even how it works out on a larger die like on Threadripper. Super interesting vid though, would definitely like to see more!
You can buy high pure graphite sheets.
You can buy Graphene sheets these days.
You can buy Silver sheets.
Make part 2 with 5 kinds of sheets, and lets destroy the paste industry!
(Paste industry really hates sheets of any kind)
Awesome content Canucks ! You guys rock! I wanna see you guys lap a cpu and try it .
Hell yea! More content like this. And with Snows. Don't leave Snows out 😂
Cool twist on the theme. No pun intended. Thanks guys!
last time i was this early pb&j is considered thermal paste
Love this format, it is a much more interesting way to present longform content than how Gamer's Nexus does it. Only question is what happened to testing the 0.15mm Indium pad?
Really educational video, big thumbs up to you guys, wish there's more videos like this in the future!!!
Damn, this video was actually pretty good! I've never heard of that stuff, and seems like a possible solution for specific applications.
Was thinking long term, 24-7/365. never dries out, or cracks, just helps move heat.
Sure as mentioned above it might weld it's self to a cpu, But I've pulled AMD cpus out that had stuck to the heatsink and wouldn't let loose . If you need to service it, it probably needs more than fresh paste .
@@OH-vb6ww true, no worries about dried up thermal paste. There are some server applications where this would probably be ideal. Hopefully no other motherboard components fail though, nor no CPU upgrades needed, then it would be an excellent server solution.
Aweseome format, I had no idea Boot Sequence was related to you guys 👍
Anti-seize, for car bolts is an EXCELLENT TIM. I've used it for years on PCs with better results than thermal paste. For those of you that don't know, anti-seize is a compound made from fine metal shavings (usually aluminum or copper) and oil. You must treat it like liquid metal because I believe it is electrically conductive. But it's well-worth the performance gain over traditional thermal paste. I saw anywhere from 5-8*C improvement. Try it out.
Fruits may not always be the best way to lose weight, especially if you get the sweeter fruits. I'm sure he's done his due diligence though! Great vid, regardless!
Yeah too much sugar. A meal on a plate 1/2 should be veg, 1/4 protein, 1/4 low fat carbs. Works for me anyway!
Hey Guys I loved the video. It would be amazing if you tried more of unknown or not so famous stuff. I always appreciate the a different thing to watch even if it is underperforming or if it is somehow expensive (like the Indium Sheets). In any case, it is always nice to see different content.
This proves again that the industry put the concave and the convex surfaces there to sell PASTE!
Remember CPU lapping? Yes, Lapping CPU and Heatsink will give you the best results.
In researching Gallium Nitride devices, i have read that using predominantly diamond based cooling paste/pads that it lowers 400C junction temperatures quite well so I’d love to see how well this could work if it was even doable.
It works great but can get very expensive. IC Diamond is better than Kryonaut and pretty close to being as good as Conuctonaut. Problem is that it's very difficult to apply and the diamonds destroy ihs' unless you use a lot of cleaner and acetone while being super careful not to rub or wipe with any pressure, I use a brush. Now back to cost, you may have seen Kryonaut LHe / Kryonaut Extreme Pink pastes on Der8auers YT channel. The reason they perform better and cost more is because the size of the particles used (TGrizz secret blend, not diamond) are smaller and it becomes more difficult to actually make and to mix evenly + keep dispersed and suspended in place instead of working their way downwards from gravity. This is why IC Diamond looks and smells like tar and has larger particles that scratch any surface they touch
Can you test this vs all thermal paste? Thermal grizzly paste, TG liquid metal and other top brand pastes?
There is little point. It's going to be ballpark to any decent paste apparently, but it's unlikely to be identical to any. There is very little difference between the top pastes.
What about long term exposure? Galvanic Corrosion concerns? Also, testing with multiple thicknesses for other materials you try would be appreciated. Definitely interested in seeing other materials!
how about different material sheets? Copper? Silver?
Due to indium's broader uses I'd like to see how it compares in an LN2 XOC situation when compared to something like Kryonaut. I would also be weary of using it against any copper surface from indium's propensity to diffuse into copper.
I guess the greatest thing about this indium solution is that it doesn't need replacement or maintenance, definitely a great tp alternative
I'm going to hypothesise that over time the indium is going to weld the cpu and the cooler together..
tbh, you can hypothize that metals with that high thermal conductivity are in the same temperature. That being said, Indium melting point is ~157°C. Its unlikely that any CPU can achieve that temperature. The problems that would need some testing would be thermal fatigue and creep phenomena. My guess would be no welding
@@myw0rk I don't mean actually welding... But even at submelting points, the soft metal will warp into those tiny deformations on the surfaces and hold them together like glue... So less like a piece of solder.. More like a piece of velcro??
@@vivekvs1992 I see. I guess that could happen if there is creep happening. Anyway, its not a few minutes testing that can tell if these things are doable or not. This is a nice subject for a video but a shitty idea and well, its not worth the trouble, either for air gaps or other issues that wouldnt happen with a paste.
@@myw0rk yeah... And not practical for us individuals to risk our only set of hardware on...
I really like the trio at Hardware Canucks!
Eber does testing, and have the best voice!
Dimitry does peripherals, and has the best humor (and hair!)
Mike does experiments, and has the most approachable presentation!
Keep this up, it's great to watch all three of you (and Snows is great too ;)
Aww, Thanks dude :)
Just the science behind this is really (pardon the pun) cool. I'd love to see more videos like this. It makes me wonder what else can be used like this tbh. Like, what would happen if you put .2 mm of high purity copper? Might be expensive, but interesting!
Please do more of this kind of testing. Great stuff!
Why did you not compare it to something like the thermal pad? Seems like the logical upgrade path to determine if one is better.
Think about it this is actually very good! Thermal paste starts to dry up after a year or so, so therefore these sheets will be alot more reliable and stable for workstations and alike!
Great video. I would love to see more of this type of content, also great job with the videos mike!
The theory behind the medium-low melting point allowing it to soften (but not melt) at high loads and imprint on the imperfections on the IHS and cold-plate is definitely an interesting idea.
As for your question of should we test more products, I would vote yes. I actually have a couple small sheets of something similar or the same from a few years back and I may just give it a try on my GPU. Now if there was a good replacement for those pesky silicone pads that would be something.
Very nice! It actually did a bit better than I thought!
As someone doing electrical engineering in cryogenic environment (from 10mK to 4K), we use indium foil for thermalization of the chips we test.
Thanks for the aluminum foil test! It's the first thing that popped into my head when I started watching.
I have an IC thermal carbon thermal pad myself. I used it on my main system for about a year before going back to a paste. Only because I was just getting center point contact from the cpu and cooler. :/
But I use that thermal pad to run tests all the time. It has saved me soo much money in thermal paste. I used to go through tube after tube. I now get just one tube every couple months. I might spring for this and try it out. Maybe the little extra thickness over the carbon pad will make the difference.
thanks
Actually innovative video on this hiatus of exciting tech (or available tech, for that matter)!
You need to be on camera more often Mike, this is the best content I've seen on Hardware Canucks by far!!
Try lapped cpu ihs and heatsink plate to see what the best scenario would be.
This was a super interesting watch.
Our boi got thicc! Snows, I just lost all my COVID weight. I started having a small bowl of cereal and maybe a banana and definitely milk for dinner. I have also increased my activity levels (gym and biking). I’ve lost 22 pounds (10 kilos?) since January.
The problem with aluminum sheets, it's that by the time you apply it, it's no longer pure aluminum. It will have a layer of aluminum oxide that have a way lower thermal conductivity. Aluminum oxidize extremely easy with the oxygen in the air.
Been using gallium/indium on my gear for six years. Granted, it is liquid. Started with CLU, but it had a hardening problem that isn't as bad with Conductonaut. I'd think think that these sheets wouldn't fill the little pits and crevices that LM or even paste will unless it is really soft. Oh, and maybe "heavy duty" aluminum foil would be the better choice rather than that discount stuff because the indium was 200 micrometers thick.
You really succeeded in making this interesting
I definitely would enjoy a more in depth comparison between same metals with different thicknesses, and also different metals altogether. I really believe this study could change how we handle building a PC from a maintenance part because as you said : when you build a PC for a friend, you do not want any maintenance tasks to happen, and that is basically the remedy to it. I also think a review using these for some time could be interesting, because I know that some metals can react with other, like the IHS of the CPU for instance, so this could also be interesting to see in the long term.
WE want more.
Just curious why didn't you do a test on the thicker sheet? To have a comparison on how the thickness affects the thermal conductivity.
I think that the main advantage is it doesn't dry out. Ever had an old tube of thermal paste and saw some oil come out of it with the paste? That oil is what likely dries out leaving that disgusting chalky and now mediocre paste behind. When you buy a graphics card odds are that you have dried up paste after a while if it is not already dried. If this foil is on par with a good paste in cooling performance that is actually worth considering. Now if you ever take the cooler off you can't reuse it. You should not reuse old paste either other than an application fresh from the tube.
Yes, it'd be nice to see other channels testing niche products, other than LTT, NG and the usual suspects. Thanks!
This was interesting AF
Might as well test silver now that Snow's curiosity is sated and silver was highest on your chart of metals.
I would have liked to see a copper ($12 for .5 mm sheet) or even silver leaf (gold too they're both even cheaper).
I used alcan's best in an AMD A10-6700 machine. It was okay. Towards the end I had to limit the top frequency so FreeBSD's audio wouldn't suck
Crazy high quality content! I was smiling & laughing the whole video 🤣
Yes more of these videos!
I think that aluminum foil was full of oils at that point, but hey, not a bad result. The performance of the Indium was impressive. Paste is easier to use, I should think, but it dries up, and removal requires alcohol. A sheet of metal alloy like that won't dry and won't stick too much. Really, though, it's rather shabby of Intel to not make proper heat spreaders. I saw another video where sanding it even cut the temperature with like 5 degrees. That's crazy.
Perhaps try it on a GPU. Would be interesting to see it compare to the stock pad on reference AMD cards for example.
You can use peanut butter, in a crunch. Or chewed bubble gum.
You know it's real when Snows stops everything he is doing on his day off just to come to the office lol!
as we know thermal paste will dried up over time, I wonder how long can this Indium last?
very interesting idea that was, yes test it more further
This video was really interesting.
Anyone else suddenly beset with the urge to stick random metals in places to replace TIM? I'm thinking tin and lead for the lols 🤣🤣🤣
So what if you hit the the heat sink with a heat gun after applying the Indium? Heat it up and see if it conforms better to the heatsink IHS.
I love me some experimentation. Maybe check them pads or other thermal solutions with a notebook as well? I remember Linus doing that with liquid metal and it was pretty sweet. So why not with some other materials? Who wouldn't want improved cooling for their notebook if it's easy to apply (like a pad). xD
One note about indium, it would work much better if it had higher pressure, but pressure from common cpu coolers is not enough to get the most of it. So you should try to find a version with holes in it, that works better apparently since it gets higher pressure (lower area). That would be an intersting test, but I am very glad that you tested indium in first place :)
I love this channel, everyone is great. Cheers from Argentina 🇦🇷👌
Woah, I didn't recognize Snow in the first couple seconds, I was like: He sounds very familiar I don't know why..... Oh that's why.
as you said it gets softer so what about overheating it a bit to the verge of throttling to see what happens?
I would love to see this tested, including the different thicknesses.
As for removing the sheet afterwards, that would make for a nice set of experiments too. The razor works, but what about dental floss? Sewing thread? I'm leaning towards some kind of thread being a safter method of doing it.
Raw dog 😂 well done boys
Great collaboration, guys this was super interesting
This also makes me wonder something: at which point does the IHS becomes the bottleneck in the whole process? After all, it's way more thick than the indium sheet or even the layer of paste after it's squished down
0:44 totally taught that was my OP xD
Would like to see this using the Thermalright anti warping frame.
I would like to see longer term test with the indium on both nickel plated and bare copper base plates to see if there’s any chemical reaction between the dissimilar metals like how galinstan liquid metal TIM reacts into copper and gets absorbed and etches nickel plating.
Should've thrown in some Conductonaut results as well, just for reference
Love to see that Indium on a Higher power CPU, such as a Threadripper/Epyc or 10900K
They got it completely backwards on where the pressure was on that indium. The pressure was on the outer circle, and the patch in the middle was less pressure.
Really interesting to see the comparison between different cooling solutions and different TDP cpu's. Cuz by the looks of things there's no reason to further use thermal paste unless the IHS or the cooler has a dank shape.
That shirt should say
"Dude, where's my card"
More testing please! 😎👍
If you want to test something very efficient in cooling CPU and GPU then test Honeywell PTM7950 phase-change thermopad.
Would be nice to see the results of indium an other materials on heatsinks that have direct pipes contact. The gaps in between will maybe act against due to air pockets?
Cool stuff.
Id go with the paste still. Thou in my mind the pad would cover the whole area better without it leaking.
I seen some horror story's of liquid metal. It drying up making your cpu heat up. but I bet most peoples applications are doing just fine.
Whens that MX-5 coming out.? gotta see the results for that stuff.
I want to see the difference Lapping the CPU and Heatsink would make using this same testing methodology. And like others would like to see Copper, Gold, and Silver. Or a Copper Indium alloy; Copper Gold alloy; Gold Silver Copper Indium Alloy?