Thermal Grizzly Carbonaut Review vs. IC Diamond Thermal Pad & Thermal Paste Benchmarks
ฝัง
- เผยแพร่เมื่อ 1 มิ.ย. 2024
- Reusable thermal pads are all the rage. We're reviewing the Thermal Grizzly Carbonaut pad versus IC Diamond's Graphite Thermal Pad, including benchmarks of Carbonaut vs. thermal paste.
Sponsor: MSI Z490I Unify Mini-ITX (Amazon geni.us/j7G9A)
Support GamersNexus directly via the store! We have PC building modmats, PC component mouse mats that are desk length, X570 chipset posters in the design of a metro, and more. All store purchases go straight toward supporting our research: store.gamersnexus.net/
Watch our IC Diamond Graphite Thermal Pad Review for more information on how we do this testing: • Thermal Paste vs. Reus...
Our CPU cooler review methodology is here: • Why Most Cooler Tests ...
Or written form here: www.gamersnexus.net/guides/35...
Scanning Electron Microscope photos taken by JOHANNES WEHNER.
This testing covers the Thermal Grizzly Carbonaut pad, which is primarily a competitor to the IC Diamond Graphite Thermal Pad. These two interfaces are alternatives to thermal paste, mostly marketing on claims of reusability, endurance (no paste dry-out or pump-out effect), and conformity to uneven surfaces. We're testing all of that here, using an uneven/unflat surface in the form of the Corsair A500 to benchmark the last claim versus traditional pastes, which have a lower minimum thinness than these pads can achieve. We're also looking at durability and reusability, thermal performance as tested in an extremely controlled environment, and value versus traditional paste solutions. Watch our IC Diamond review for methodological notes on the testing solution. As we said before, "real" computers -- even using heavily controlled test benches, as described in our CPU cooler methodology above -- are simply too variable and difficult to control for this type of testing. In order to accurately benchmark thermal interfaces, custom-engineered dummy heaters must be used in order to produce a meaningful difference. A "real-world" solution would either (A) produce unusable data due to variance, with no ability to trust said data, or (B) produce results that, in the absolute best case, are all within error. With our custom solution, which is a dummy heater in the exact shape of an Intel i9-9900K IHS, we're able to narrow-in results to produce meaningful data between the solutions. If you want to know what the best thermal pad or thermal paste is, this is the best way to produce that answer. We're working toward it still, but we're starting with a slew of graphite and carbon thermal pads.
Find the Thermal Grizzly Carbonaut pad (32x32mm) on Amazon: geni.us/cwLnKrB
Find the IC Diamond Graphite Thermal Pad (40x40mm) on Amazon: geni.us/56kk27 (we used one size-appropriate for the Intel DT IHS)
Find Thermal Grizzly Hydronaut on Amazon: geni.us/Fsray
or Arctic MX-4 on Amazon: geni.us/k3kT6
TIMESTAMPS
00:00 - IC Diamond & Paste vs. Thermal Grizzly Pads
01:52 - Why Paste is Almost Always Thermally Better
02:55 - Test Bench Decisions
04:13 - How Thermal Grizzly Carbonaut Works & Microscope Photos
07:05 - Thermal Grizzly Carbonaut & IC Diamond Pad Marketing
09:54 - Thermal Conductivity (W/mK) Claims Incomparable
11:07 - IC Diamond Graphite Pad vs. Carbonaut Thermal Benchmark
13:38 - Pressure Testing Notes & Hitachi HM01 Thermal Pad
14:32 - Unflat Coldplate Testing vs. IC Diamond & Carbonaut
18:20 - Conclusion: Paste vs. Graphite Thermal Pad & Carbonaut
Like our content? Please consider becoming our Patron to support us: / gamersnexus
** Please like, comment, and subscribe for more! **
Links to Amazon and Newegg are typically monetized on our channel (affiliate links) and may return a commission of sales to us from the retailer. This is unrelated to the product manufacturer. Any advertisements or sponsorships are disclosed within the video ("this video is brought to you by") and above the fold in the description. We do not ever produce paid content or "sponsored content" (meaning that the content is our idea and is not funded externally aside from whatever ad placement is in the beginning) and we do not ever charge manufacturers for coverage.
Follow us in these locations for more gaming and hardware updates:
t: / gamersnexus
f: / gamersnexus
w: www.gamersnexus.net/
Editorial, Testing: Steve Burke
Video: Keegan Gallick, Andrew Coleman - เกม
Watch our IC Diamond Graphite Thermal Pad Review for more information on how we do this testing: th-cam.com/video/niAQs8dZohE/w-d-xo.html
Our CPU cooler review methodology is here: th-cam.com/video/fmTOJP4KOyk/w-d-xo.html
Or written form here: www.gamersnexus.net/guides/3561-cpu-cooler-testing-methodology-most-tests-are-flawed
Support GamersNexus directly via the store! We have PC building modmats, PC component mouse mats that are desk length, X570 chipset posters in the design of a metro, and more. All store purchases go straight toward supporting our research: store.gamersnexus.net/
what about testing a pure indium foil? personally I'd like to think that a CPU at full load would have enough heat to partially melt the foil/pad filling gaps while also having better conductivity then impure liquid metal "galinstan" mixes. with the added benefit of not leaving a stain due to the lack of gallium. but I could be wrong.
Can you use alluminiun foil"tinfoil" as TIM???
why do you use hydronaut rather than Kryonaut @Gamers Nexus ?
is this you? steve? th-cam.com/video/QTahVLSJz_k/w-d-xo.html
@@UKgamer87 have you used one on an Xbox 360? Do you know what size pad you would need for that?
That intro B roll shows the typical CPU ritual done before testing :D
This is actually a GN secret to get the most accurate testing.
@@GamersNexus you didn't let it complete. OCD 9000
Roman probably does it every time before deliding. :D
I thought it was the next step to checking out of the box thermals. You put it out of the box, onto the table, apply thermal paste and check temps.
🤣🤣🤣🤣
You're supposed to use 2 slices of pad with thermal paste in between to create a Grilled Carbonaut sandwich. Then place more thermal paste as needed.
Run prime 95 for 1h and it's ready to serve
I thought you were looking familiar🤪. You also recommended using a big dot of thermal paste for every processing core. If you ever do a Threadripper 3990X build, please use one of these pads🤣.
What do you need to do all that? A table, a Swiss army knife hopefully with a tool that you can use. What else?
@excelsior ink You probably should recognise the OP as a meme and their comment as a joke. I don't believe paste with thermal pads is a good idea at all. But i am curious whether shim (and paste) or carbon pads behaved better for you.
It's just good PC building practice.
Commenter: "Omg thats so much paste, don't do that"
GN: "makes a ring of thermal paste around a cpu just for some b roll"
Pretty much!
Ice the cake baby
It's a common practice to summon the PC gods! (source: probably the Verge) 😂
The Verge: "You would usually have 1 tube of thermal paste, but usually it's not quite enough. You need a second. Smear with confidence"
at least Steve isn't putting it in CPU Sockets anymore 🤣🤣🤣🤣🤣
I was recommended this channel when I asked for info about how to build a new PC. And WOW, this is *monumentally* detailed info! I had no idea anyone even cared about this level of detail for thermal paste and competitors. I am fascinated!
Welcome to Tech Jesus 👌😁
You have now been sucked into the world of always trying to improve. I am addicted to making my PC run cooler and faster.
More importantly they take great care into their testing methodology. It's fine to watch reviews from Linus and Jay but GN is great for tests that compare competing hardware head to head for many different use cases. This is especially useful when putting together a parts list for your new PC. A PC you mainly use to play Fortnite has different requirements than one used for video editing.
Though a channel that can be recommended if you're on an extremely tight budget is LowSpecGamer. He has the best tips and tricks to run demanding games at an acceptable frame rate on very low end hardware like a dual core Athlon APU using integrated graphics. Don't expect games to look pretty though. That kind of hardware actually isn't often worth buying if you live in a western country but can be very useful in countries where the average income is relatively low and PC hardware is relatively expensive without an abundant 2nd hand market (like Brazil for example).
I watch other tech channels when I want to be entertained. I watch Gamers Nexus when I want data, and no punches pulled information. HAIL TECH-JESUS
U were lucky u didnt discover Verge first...
Unbelieveable. Tech jesus actually mentioned my previous comment about thermal pad for laptop. What an honour!
I made a comment about half on Intel's Core series being 14nm, and they mentioned it. Best feeling ever.
@@DrearierSpider1 Same here. Thank you tech jesus for making my day.
🙌
Congratulations, we are all proud of you guys 👏.
As an atheist who chooses to believe in facts and proof I believe in tech jesus
@ 0:02 That's not how to apply thermal paste! You need to drill a hole in the IHS, and inject the paste in there with the syringe till it won't hold any more!
Pentiums used to have the hole pre-drilled for you. Talk about service!
Wait, why does it actually sounds like a half decent idea?
@@StevenZephyc because you don't have a clue probably
I want someone to benchmark that...
@@StevenZephyc It would be for Intel Coffee Lake. The original TIM between the die and the IHS was crap (they were good processors though).
It's like a cpu blanket, tuck it up all warm & cosy! 😂
But you could slather it in hydronaut and have your wicked way with it!
Emphasis on warm lol
@Optical Claritystill won't be as good as just using the paste, bigger gap less heat transfer. \also you would combine both disadvantages, bigger gap, and having a paste that needs renewing.
@@Machistmo we know. :D
@@user-yv2cz8oj1k Don't be so cheap... Kryonaut is where it's at
One of the things about the pads is that you don't have to wonder if you put too much or too little grease on. They are basically always consistent with the "dosage".
Yes as someone who never put thermal paste on and was putting together a new pc for the first time, this is why I went with it. I was sure I would have to take it right back apart to troubleshoot, so doing thermal paste 3 times didn't sound fun. Luckily it ran fine first assembly but I didn't test parts so the new mobo could have been doa. Long term under pressure it gets rough and I wouldn't really put it from one cooler to another, but for straight reassembly I think it's fine. I might have tightened it too much and I did the second one less when I upgraded my cpu. We will see in another few years if it's right. If I remember correctly thermals went down when I tightened it more but it stopped at a point and I went a little further for testing so when you see it stop improving I would leave it there.
Oh I had to swap the cpu cooler for one 20mm shorter a few weeks/months into the first assembly to fit side fans. I wasn't sure about the clearance, got the tall one and had to wait a while to get the second one when I could. A weird use case for "reusability" but SFF is getting big right now and chasing every mm of clearance when new parts come out. Mine was an old matx mini tower case with side fans above the gpu and cpu and the 155mm I got first left me 7mm not even enough for the slimmest 12mm fan so cpu cooler had to change.
just spread the paste, no way to mess it up
I love how Steve explained the nanomaterial aspects for both products (especially in the pad).
I can definitely see these being useful in things like iMacs, where they’re basically glued shut. So you don’t want to open the damn thing more than once or twice. So just adding a “fire and forget” pad and not needing to worry about thermal paste is seriously damn cool.
That’s a great point actually!!
And if the user launches old minecraft in java we have a nuclear meltdown.
Until you have to dust it out....
@@bam6693 True
I've used IC's pads over thermal paste for most of my PC builds for the last five years. I am very happy with their performance, which is so close to optimal paste, while also reducing the chance of screwing things up with a bad application, that I feel it's the best option overall.
I been trying to decide on Carbonaut vs Cryptonaut Extreme for about an hour now and this is the best review I came across. Also, having it 'mapped' with chapters is a very, very nice touch. Loved the presentation and the scientific approach. Didn't get into unnecessary detail and didn't try to any obnoxious humour. Well done!
Engineer: Boss, we have imperfections on the cut process
Boss: No one will notice them
Steve: So we grabbed the microscope...
Edit: Image credit to Johanness Whener
Johannes Wehner took those photos! Make sure he gets credit, not us!
@@GamersNexus edit done.
Great content btw, keep up the good work.
@@ijustsawthat The joke was still very much appreciated! And thanks!
To me, the best use case for Carbonaut is consoles, where you really don't want to open them up and repaste them, and their longevity is typically longer than a standard PC. Console users generally would have no idea about repasting.
I recently had to open up and clean out a Phat PS4. I certainly agree with this comment. It would be a perfect application for one of these thermal pads.
I never had to repaste in my PC and its 8 years old now. Still same temperatures.
Xirpzy but consoles use cheap paste which is notorious for drying out. This has been something ongoing with consoles for a long time
Still would use paste.
@@Xirpzy Same here, using a generic cooler master paste from my 212 evo cooler on a 10 year old I7 860, the paste is 7 years old still runs cool even if the paste is most likely to be dried out and crusty.
I LOVE the way you guys test, its not just put it on and take it off and done. You guys look at almost every scenario. GOOD JOB!
So guys, first of all I love your videos. I've been here for some time now, and I am really thankful to know a tech reviewer with such strong believes in testing methodologies, wanting to deliver simply quality content, pointing out how things are. For me personally, I was always most impressed by your statements and philosophy regarding bias, and how to properly deal with it.
Now, to be really clear: I do NOT believe your reviews to be biased in general - and I think this one here neither. But... you guys know how, especially in this area, it is often difficult to keep things apart. You said some time ago, that for this exact reason you don't like to take techy sponsoring deals and prefer those mostly unrelated to your content. With all that in mind: How did you forget to mention, that alot of your other videos are sponsored by Thermal Grizzly, a company founded by Roman aka der8auer who is actually (for all I know) a friend of yours? I mean he even gave you an early sample of the Carbonaut thermal pad personally. I can remember that; I was actually really looking forward to your review, because I do believe that you can keep things apart, despite all odds. And I still think of it that way, otherwise I wouldn't even be botherd to write this comment. So why didn't you mention this at all, although your audience could benefit from this information?
I guess one could argue that probably most of your audience knows about these facts - your friendship with Roman isn't exactly a secret, neither is your sponsoring deal.
But that definetly won't be the case for all of your viewers, and even some of your subs and supporters might not be aware.
I think to briefly mention this fact at the beginning, to inform your audience of possible bias, would have actually benefitted your content. And surely it would've earned you a lot of additional trust within the community. Now I don't really know how to feel about this.
Maybe I'm just overreactig on this, maybe I'm all wrong here - but I really wanted to point this out.
Nonetheless, great video as always - and I'm looking forward to (maybe?) more thermal compound reviews! Thanks for listening
Here's a possible fun test case for your new high precision CPU heater that I've been curious about for a few years:
*Tinting* ... I remember a few years back I *actually* read the instructions that came with a tube of Arctic Silver (for some reason).
It suggested "Tinting" the IHS and cold plate by applying a tiny amount of paste to them then buffing it back off.
At first I thought it was just a clever way to get people to use more paste, but as I thought about it, it started to make sense.
By buffing in a microscopic layer of paste, you can apply more direct pressure to the surfaces, filling in the tiny imperfections with more force than just the mounting pressure of the cooler, so when you apply the paste to like usual, and mount the cooler, you'd theoretically get better contact.
There would be no way to detect any improvement by using a real CPU, but your test rig may actually have the resolution to check this.
Just a thought.
Amazing guys, appreciate the work to build this testing method, awesome video and thanks for the information as always!
At last the Thermal Grizzly Carbonaut review that I'm looking for haha
u wont a get a review like this from the person who made the pad
I appreciate the thoughtfulness behind your testing methodology. The logic is sound, you mention potential pitfalls, and it's never dumbed down.
Excellent review and side-by-side comparisons. Thanks so much.
0:00 I can’t believe how much paste you waste! Don’t you know there are starving chips in Xboxes!
Been looking for a comparison of these pads, good review!
Thanks GN! I learned about the thermal pads from you. I did some research and replaced the thermal pads on my 2012 i7 Laptop. I used TG minus pad 8, Arctic cooling mx4 paste for CPU and GPU, removed the dust carpet from behind the radiator, soft brushed the cooler and the results are amazing!!! From constant throttling with 80-100 Centigrade, down to 60-70 Centigrade! My SSD is safe now....
Right down to the electron microscopic details. I love this channel.
what is that intro b roll lol
It's a waste
Bro ur awesome, you've giving me so much advice on topics that I didn't even know I needed help with. Thank you dude
Dear GN team, can you test also the usual amazon "go to" inexpensive stuff since 90% will buy them ?
Thanks ... nice job !
Agreed, I have been in the PC industry nearly 20 years and cheap or bundled pastes have always been suspect.
@@Knirin cheap stuff will be used in the workshop, office pc etc ... I'm talking about MX-4, Cooler Master and others (I know there are many versions). there are also the same priced items with really low W / (mK) and don't trust Worty. From experience, ultra-cheap paste is only good for the office: after 70w it is horrible
P.Š. sorry now native speakers
@@ElettroStef92 I've been using the same thermal compound bundled with my Hyper 212 from 2011 and it always worked flawlessly, especially considering the price.
@Advocatus Diaboli Also it's not conductive or capacitive
@@Knirin With bundled paste it does depend on the cooler manufacturer. Using the included Noctua TIM is fine.
I just got a Carbonaut pad for my 3400G build that lives behind my TV. I probably won't replace or upgrade it for a decade if Windows doesn't end up needing far higher end hardware, I just use it to play very light old games it stream from my main machine. Main machine and servers get paste, and that gets replaced every few years if the machine isn't being upgraded. I got the Carbonaut because I didn't want to have to worry about the graphite pad potentially breaking and having conductive flakes get somewhere on the board.
But yeah, for my 3400G with a Noctua NH-L9a, the temps of MX-4 vs. the pad were very nearly identical, although the Carbonaut pad seemed to heat up faster...
ah, it's like a cloth then, I was curious what kind of structure it has, interesting
I would say its like very thin silk since i have one.
@@TheNewCrap is it fragile? If so, how much?
@@internetexplorer6824 Can´t say for sure but I wouldn´t try ripping it. I handle it with two fingers only since I didn´t want to damage it by mistake.
Dont know if you cover this, and i dont want to forget, but I've noticed that carbonaught gets 'gummy' when hot, you have to let it cool completely off before removing the cooler or it might tear, or tear when pulled from the IHS
Also, i want these pre-installed in OEM desktops and servers, i dont like having to re-paste hundreds of machines
Thermal pads would be excellent for servers. I bought a used server off eBay, and the thermal paste on the two CPUs was easily over a decade old, and it was hard, dry, and flaked off. The thermal performance was terrible. On the upside, I put some Kryonaut on there, and the improvement was drastic. Even the noise went way down because the fans weren't having to work so hard.
You are sir, what I would call an actual reviewer ! As always, awesome video. Precise and concise yet detailed where it matters !
I'm satisfied with seeing "uploaded 46 seconds ago."
That's strange, mine says 19 minutes.
I'm satisfied with seeing "uploaded 46 minutes ago", huh
Your level of analysis here is off the charts. ;) Awesome work folks.
Waiting for the Elmer's Glue brand Thermal Pad. Likely will be a sponsor in The Verge's next PC build.
Great, now all I want is for Steve to test Elmer's paste/white glue as a thermal interface
WABBIT season
@@sopcannon Duck season
@@dikbozo Wabbit season
Well done Steve! Great stuff as always!
Thank you Gamers Nexus and also thanks to Beve Sturke for your awesome Videos.
I dont know any other channel exept of igorsLAB that puts so much effort in his Videos and testing.
Great video like always.
This is incredibly good testing. Thankyou
Really useful, next time I have to redo the paste on my One X, I'll be renewing the surrounding heat pads...
Id try this with an older CPU but theres no way Im trying these things with an i9 or something. But anyway. The amount of editing, attention to detail, the animations.. EVERYTHING in this video is just top of the line. You guys have earned my trust 100 times over at this point. If you tell guys me to jump, Im gonna ask how high.
Comprehensive, clear and easy to understand. Hallmarks of Gamers Nexus at this point. Love it!
As someone who loves playing around with retro PCs, I absolutely love my IC Graphite Pads. They save so much time and effort, and flipping CPUs back and forth takes only seconds.
Glad to see there's more development in this space, gonna have to try the Carbonaut next time. :)
Good video, thanks guys!
Hope you could review MSI's MAG Core Liquid 240 or 360 AIO with the pump in radiator design, with your usual deep-dive and tear-down. I'm interested to learn who makes their AIO coolers. Thanks for all your hard work!
Nice. Saw the intro and had to watch the whole thing.
No real “wow” moment until that 0.11Nm “loose AF” test. WOW the low pressure wasn’t even that bad for paste, but wrecked these pads.
I'd love to see an additional follow up using some sort of hot-spot test, to see if the IC Diamond provides some bonus with it's horizontally aligned carbon.
Another niche use for the pads might be fanless cases that clamp heat pipes to the walls of the case. Those often need a ton of paste, and dripping/mess could be a major issue
I'd be interested in knowing if gold leaf (you can get thicker varieties), being relatively soft yet solid.
Also, I'm curious to see the hotspots of different CPUs (Ryzens being wonky) and hot much the IHS actually spreads the heat. Take Ryzen again, CPUs with only CCS have the bulk of the heat on one corner. That must have an impact on heatpipes? Maybe embed a series temperature probs in the surface of a heatsink or waterblock that you can rotate 90 degrees to map wear the heat is located. I think a thermal camera would be a great idea if you can find a way to read the source not the reflection.
Can't wait for a comparison/ review of all the TIMs on the market. 😉
(Or just the popular ones.)
YES! I've been curious about using gold as a TIM as well!
I've kind of let my imagination go crazy with ideas of lapping the IHS and cold plate down to a perfectly leveled surface, then plating them each with gold, then wring them together to cold weld the surfaces together. A lot of work, but more of a thought experiment.
the top 5 best conductors are 1Diamond, 2Silver, 3Copper, 4Gold, 5Aluminum
Silver leaf would be a real interesting experiment.
@@BRUXXUS the issue is that the level of "flatness" required to cold weld something of that size isn't likely attainable at home, and it needs to be done in a vac chamber. The possibility of getting a full weld on the entire surface with no free space isn't that good. Though gold leaf as a TIM is something to test.
Would be an interesting test for sure. While at it, they should also test an indium sheet as well
Damn it Steve, I was just about to go to bed! Oh well, guess I’ll watch it (2:31 EDT as of this comment).
Okay, now bedtime. Good night, and I learned a little more about these types of thermal pads :)
For CEST its perfect. :D
Steve, his team and his associates are the bane of lazy engineers everywhere. The moment a flaw is overlooked, they jump on it and rip it apart.
That's why I love this channel. Who else goes that in depth into their reviews and testing?
Recently moved to Kryonaut from using an IC pad for the last two years. My 8700k is OC'd to 5.2 GHz at 1.36v. While using the pad temps would hit the mid 80s. After switching to Kryonaut I ran CBr20 for two straight hours. Temps shot up to 79c initially and settled down at 74c after my Celcius S24 was saturated. Currently have a PC-o11 Dynamic with the s24 mounted for exhaust at the top with three 120mm fans for intake on the side. All fans are FD X2 dynamic PWM fans all running at a static 1000 RPMs.
0:02 OK, so thermal grease is running circles around carbon pads; got it! Tldr - great job 😁
Great content, as always. Asking the absolutely most ridiculous question - would thermal pads perform better if you lapped the CPU/Heatsink? Granted, at that point we're already talking an extreme enthusiast who would likely want something better; but now my curiosity is piqued if they'd provide better thermal transfer on a perfectly flat, polished surface. Any chance you might try that out for science?
thanks for doing the work. Once I discovered it can conduct electricity I swapped back to paste. I keep it as a backup for emergencies.
It's been awhile since someone did a serious review of how Lipstick, Mustard, Mayo and toothpaste perform. Your channel now seems uniquely positioned to do this accurately if you ever find yourself with too much time on your hands. Keep up the great reporting, thanks!
serious review- it sucks don't do it
Thanks for the excellent content as always.
Finally a tech channel that come to their senses and uses actual hot plates with a defined, measured Wattage. So many others use benchmarks, in-game FPS and other stuff that is ALSO affected by a bazillion other variables and isn't fluctuating enough within one second, on one system, without any other changes. Not with a normal CPU or GPU that self-regulates its thermal output and runs three hundred computation threads that will never line up the same way after repeating the test. Especially not when the previous test, overclocking, undervolting led to a system crash - nobody knows if the system works properly after that or if it not behaves differently for whatever change the software did in response to the crash. A synthetic heat transfer test is the only thing that can finally put all the speculation to rest about what interface material is better. Test results need to be validated on real CPUs with real PCs of course, but only after we know what the results should look like on bare metal. Actual metal.
I had one of those low mounting pressure situations with an Intel Xeon D-15xx high-integration processor in an embedded device with a passive aluminum heat sink that used stiff springs wrapped onto screws to apply the mounting pressure. In that situation, I measured a very slight advantage for IC Graphite of slightly less than 1C over Carbonaut.
During the testing, I found Carbonaut to be quite fragile while simultaneously being finicky requiring more retouching. The stiffness of the IC Graphite allows for easy repositioning by nudging any edge, while the flimsiness of the Carbonaut sometimes required multiple touches to remove wrinkles induced during repositioning. However the Carbonaut did tend to remain where it was placed much better than the seemingly slick waxy surfaces of the IC Graphite which caused it to unexpected slip out of position. I did not try any "wetting' work-arounds to temporarily increase adhesion.
I tried the LTT thermal conductivity test of a thin sliver of IC Graphite by holding one end while bringing the other end into a flame. In a fraction of a second it conducts the heat to the other end forcing one to drop it. Interestingly, the flame didn't noticeably damage the flame end, though I know that end is probably not quite the same anymore. Given the wild X/Y axis thermal conductivity, I suspect that small imperfections such as creases are largely swamped by the surrounding material absorbing the extra heat along the imperfection, though a full tear might have a meaningful effect due to an actual gap in the material. I suspect that would largely depend upon thermal any involved heat spreaders.
Comprehensive and reliable results. Thank you!
You've got a hand model. That level of attention to production quality detail, on top of the impeccable journalism.. I know it's cringey to gush & fawn over people like a Japanese school girl over their favorite pop band, and I know that it may be a bit eye rolling that *that's* the part I'm commenting on and not the sweet-ass TIM testing methodology, but *damn* fellas. You're leaving every single other PC hardware TH-cam channel in the dust. I know it's not a competition, I'm not trying to engage in tribalism, and I greatly respect & enjoy the content of several other tech TH-camrs, but y'all are just on another level.
Did I put enough disclaimers there to give the genuine compliment a fighting chance?
Haha, thanks for the support! It's actually Andrew. He does a lot of the b-roll, but doesn't normally get too hands-on in the shot.
@@GamersNexus Oh, hahah, I primarily noted the very fancy looking nails. Looked professional. But further kudos to you that you achieved that in-house!
@@snozzmcberry2366 I'm sure he'll be flattered!
Now that you have such a good setup, could you maybe have a page/chart with every single thermal paste you can get your hands on? That would be an invaluable resource.
0:04 So that's how you apply thermal paste properly, i've done it wrong my whole life . Thanks for the tip
GN team, in this test you are using the Intel block that has all the heat in the center with an AIO cooler that prioritizes the dead center spot. Paste is expected to win here. Can you test a chiplet design with the same aio and then add a good aircooler that has multiple heatpipes to understand how the horizontal properties of the different cooling interfaces affects overall cooling? Depending on the results it could either make it non-relevant for future tests or expose specific combinations where performance differs.
Lastly, in my N=1 testing, I used both paste and ic thermal pad on my 3900x. Both performed similar for me so I decided to use both at the same time. I did three super tiny dots over the chiplet locations and then placed the pad on it. My noctua 12s performance was significantly better (like 5C) compared to paste alone or pad alone.
Edit: also, great work building the testing rig!
Paste wins in both.
@@GamersNexus Video or it didn't happen. Also, in my crazy case of using both, I was using 1.3V 4.25GHz static overclock.
Awesome review as always. But I wonder would these perform better on non CPU pc component such hot NVME SSD controllers or cut it to pieces and use it on graphic Card VRMs with something like the new Morpheus 8057
Last time I was this early the Pentium was a premium brand.
I think it has to be said that although these pads are not as good as one of the market leading pastes, it's still very impressive to what we had a few years ago (nothing) and can be done very quickly. Plus looking at the benchmarks the pads clearly beat medium and low end pastes.
And it's much harder to screw things up with a pad.
i love these pads, adding it to a pc or laptop for a non technical family / friend is great, wont have to worry about it after
Awesome, can't wait to see some baseline thermal for liquid metal.
This carbon sheet looks like a perfect laptop and phone solution, where you are unlikely to open it up to replace the paste and they do deteriorate using paste.
I use these in my retro builds when I grab a couple parts off the shelf and throw something together for a couple of days to play around with. And given how much I do this I enjoy the fact of having no mess to clean up and the fast application of the pad as well. And mine seem to be reasonably durable as I have gotten about 20 applications out of it so far, but I am sure the ones I use are not as good in performance as the brands in this video, and they are also not as good as TIM. But I also find them to be good enough as I don't overclock and my systems run open air typically. If I build into a case I will use TIM as it is expected it will be assembled for a longer period of time.
So in a way I am the target market for these things.
There was this old TIM that was basically a sheet of some Gallium alloy that melts around 70C, you installed it and then unplugged your CPU fan and let the processor idle the whole thermal solution up to around 70-75, then plug the fan back in and you've basically soldered the CPU to the heatsink. Worked damn well and was pretty easy to remove afterward, not reusable and very expensive though. It was one of those "I'm building a whole PC anyway I'll try the $20 shiny" purchases. I don't remember what it was called and can't find it now though, sadly...
I definitely like this for troubleshooting purposes. Seems incredibly convenient.
For my new built i finally went with the Carbonaut Pad for the first time after decades of thermals paste and i will probably never go back.
The 2-3 C° difference comapred to top tier thermal paste don't really matter to me and it still outperforms most pastes out there or is at least on par., but the great thing is that you never have to worry about cleaning when chaning CPU/cooler or if you are using too much or too little etc.
Can you redo this with your new pressure imaging/mapping?
But, can you use this indefinitely as opposed to the paste? If so, this might be better for pre-builts/off the shelf.
*ah you covered this.
I want to see a base test of no material, see how bad the cooling is just from the heat sink by itself.
It will just shut down. Maybe you can get in the bios, but after some minutes cpu will be too hot.
@@simoSLJ89 The beauty of a dummy heater is that it doesn't have BIOS and won't shut down.
@@GamersNexus But Steve! How can you overclock it then? Awww!
@@conza1989 just plug it into a 220V outlet and watch it glow with power
Could get flat and smooth enough surfaces to eliminate the paste?
@Gamers Nexus, after watching your review on the thermal pastes and pads, if one was wanting to redo the paste/pad of a slightly older laptop that seems to run a tad hot which would you recommend then? It has been a while since I've last looked into this and I really enjoy watching your videos and what you guys do for the computer world with your reviews. Thanks.
Now we need a review like this for those blue and pink and other color thermal pads used on GPU memory chips, power MOSFETs on motherboards, and etc.
Genuinely blown away by the complete consistency and quality from this channel. GN belongs on Discovery channel with their own show, not TH-cam.
I've used both, and I much prefer the IC Diamond pad. The Carbonaut part tears very, very easily. One time handling it a tiny bit too much and it tears. And that happens easily when taking the pad off a CPU, or when it's stuck to a cooler cold plate which happens quite a lot. Also, I often have lower quality coolers with sharp edges and/or old coolers with a fairly small cold plate. I feel the Carbonaut pad performs somewhat better though, seems I was correct at that.
Funny how they say now that u really shouldn’t reuse it and it’s notoriously easy to break because it’s graphene.
Great review!
Having watched this, for 90% of PC builds I'm going with a Carbonaut pad from now on. Easy, no mess, shave a couple of minutes and some stress off build times. For 45 and 65 watts CPUs in a media pc it's not worth bothering with paste for the sake of 1 degree delta, they'll never get above 70 anyway. I'll probably start putting them in workstation builds too, since so many of those get shoved under a desk and never maintained, ever. Thanks GN.
This is the scientific rigor I demand in my PC build decision making! I'd love to see liquid metal and stock generic thermal compound for reference.
15:50 Was someone playing with the fonts on this chart? Nice to see times new Roman on screen.
This has really made up my mind for me. I'm using a laptop that is starting to run at 99 C after 2 years. I've never applied thermal paste before, so I think I'll be going with the paste-but not a metal based one. I'll save that for when I have a better idea about what's going on :D
first considering the nature of pads i'm actually impressed,i thought they would of been much worse than they actually are,2nd "when you buy a new cooler every few years" i don't buy a new cooler that often,shit i still have the original cm hyper 212 here that i gave to my younger brother,i myself have a noctua D9L (i have a case that has a cooler limit of 130mm) and i highly doubt i'll be swapping it any time soon.
pads would be good if you don't trust/expect the person to be able to re apply the paste over time themselves (like your mum or grandparents etc)anyways appreciate the in depth review.
Still using my 212 EVO. It's been through 1 AM2+ MoBo and CPU, a hybrid AM2+/AM3 MoBo with 2 different CPU's, and my current FX 8350 + Asus M5a99x MoBo.
Will probably be buying the AM4 bracket and continue using it when i finally upgrade to Ryzen and AM4.
It's a frigging workhorse.
Certainly on a CPU, paste is almost always the preferred option. Unfortunately, I ran into issues with using paste on my reference RX5700 with an Arctic accelero xtreme iii. Temperatures would start out very good with normal paste(Noctua H1, in thise case) but would slowly rise after several thermal cycles.
I measured as much as a 12c difference in Firestrike stress test load temps(20 loops) comparing day one performance to one month later. I suspect uneven mounting pressure from the heatsink's massive weight across the relatively small die of the 5700 caused a portion of the paste to pump out on one side. I replaced the paste with an IC diamond pad and the temperatures have stayed consistent over time. Not as good as day-one paste(~5c worse), but much better than day-30 paste. Looking at your testing, maybe I should've spent a bit more for Carbonaut. At the time, the only Carbonaut in stock were sizes much too large at double the price and I didn't want to risk cutting the pad with scissors myself.
Will you be testing the Panasonic SoftPGS? One of the big benefits of SoftPGS is it's heat distribution capability in XY plane. It would be interesting to see if you notice any measurable difference with a multi die/chiplet setups.
It would be nice if you could run an extended test at the higher wattage (say... weeks) to dry out the thermal paste to show how its performance can degrade over time--where the pads should remain relatively consistent.
I've been using the IC Diamond in my PS4 for over 6 months, the PS4 is now so quiet, whereas before it was like a jet engine probably due to the old paste drying out, was good to see the comparisons but I'll stick with the IC as I plan on getting a ps5 sometime in spring next year when there is plenty in-stock
So around the temps where I hang, there's one degree between them. I bet I could get better results by messing with the thermostat a little.
Entertaining as always guys.
You're tearing me apart, Corsair! 🤓
I did not rip her, i did naaawt
Oh hi benchmark!
Even though I hate this, why do I feel so hot?
Id like to see a big Review with the best Thermal Paste at the Market benching with Ryzen CPU`s. KIngpin, Artic, Grizzly etc...Would be interesting since not many TH-camr do that. Specially not big channels
Already done ...
I think at low torques in your testing thread surface variation in both the screw and receiving thread is going to make quite a large variation in actual clamping force.
Mounting and remounting many part's like technicians do they would have access to cheap decent paste volume purchases that would likely come out cheaper than any of those pads. Also with those pads being conductive and someone working under time pressure its just easier to squirt some paste without having to worry about anything shorting out, especially after we have seen time and time again that to much paste doesn't matter anyway.
I want to propose a use case with low mounting pressure:
Retro PCs. Any high clocked 486, 586, 686, or the K6 line of CPUs. They all use clamps to mount the cooler, and these can be very loose. Also the engravings on those CPUs are extremely deep compared to modern IHS's.
Having a slightly lower thermal conductivity won't hurt at all in such cases, and the potentially better longevity is very welcome.
Phil's Computer Lab seems to really like the IC Diamond Pad. I wonder if he accidentally stumbled upon the mounting pressure anomaly that GN discovered. Could be why he was so impressed, he might have seen issues with Carbonaut if he ever tried it.
Those CPU's didn't produce much heat though but could improve the noise production if you'd solder a resistor in the fan wire.
For beginners, techs, long term use, and interestingly SFF/MFF where clearances are at a premium and constantly changing with new better CPU coolers, larger GPUs, and smaller cases, I would absolutely recommend these. Great for test fitting but they should really not be swapped around if they have been really mounted down for any long time. For maintenance as long as its the same cooler, its in one piece and you're careful I would put it back no worries. Theurr cheap enough to replace with a shiny new part like cooler or cpu. They cool great even with (reasonably) uneven heat pipes on otherwise strong coolers and should do so indefinitely. 4.5/5
I was building my first diy and never done paste before. I went with a smaller matx case with 2 side fan slots. Wasn't sure about the cpu cooler/side fan clearance and went too tall. Also wasn't sure the pc would even work first try but I got lucky with no doa parts or install errors. All this and this video lead me to the TG Carbonaut. I had to swap to a shorter cpu cooler when I could a few months later. I probably overtightened it but it was only a ryzen 1400 at 3.6 ghz. Later I upgraded to a 5700x and got a new one and tightened it less, we will see when I swap the mobo from b450 to b550 eventually how that pressure is. And then I will leave it forever unless i upgrade to AM5 in the same case but I'm gpu size limited in my current case. I put the older pad on the 1400 stock cooler equivalent it had originally since the paste was only on the cooler, it was apart for over a year just hanging in the old case, and it was from 2017 in a pre-built. If it ever needs work or I put my tall cooler on it, I should probably replace it but its still just a 1400. The edges on the am4 recommended 38x38mm stick out around the ryzen 1000-5000 and if you tighten it too much it will pinch or even start to cut it. Try to not tighten it that far but if it does go ahead and take off the edges because it is electrically conductive. It will stick after mounted down and heat cycled, with luck and skill you can keep it on the cooler side and this is better as that will be the side with uneven heat pipes.
"Technically superior" needs to be on GN merch
Lmao, Steve with the bait right out of the gate. Love it.
Impressive Work! Lovin ur videos! I would love if u Could test Liquid Metal Pads. And i would be really interrested if u could test a tubes of old Thermal Paste. Say i buy a tube of Hydronaut and i'd like to use it for my next PC after 3-4 years. Does it loose performance?
Tnx in advance
I've been waiting for this video for some time lol
This is exactly what I go on yt for
Bless Papa Steve.