Closed

Thermal Paste Round-up: 85 Products Tested

This much-needed update to our thermal paste round-up now includes 85 contenders. It's one of the most comprehensive comparisons available, so we hope you find it useful!

Thermal Paste Round-up: 85 Products Tested : Read more
92 answers Last reply
More about thermal paste round products tested
  1. *heavy breathing*
    I love these kinds of articles and in-depth super tests!! Thank you so much for all your time, effort and hard work, I appreciate it. I'm sure that I'm going to enjoy reading it.

    Um, do you guys still have a single page or "printable" view please?
  2. Oh, amazing article. I love it a lot.

    Maybe it's because I've used Artic Silver 5 for so many years, but for me it's the best all-rounder compound there is. Plus it's very cheap. I like it more than the MX-2 and MX-4 compound siblings people usually recommends. But I have to say, the "diamond" compounds are indeed better it seems. I had my doubts, but no more with these tests.

    Cheers!
  3. Long story short: apart from esoteric TIMs, all pastes are practically as good as any other for typical uses when applied correctly. That really shouldn't surprise anyone as all pastes rely on the same principle of various particle sizes in silicon oil suspension getting crushed together.
  4. Hang on, I'm sorry.

    Quote:
    Also, very cheap silicone-based solutions like Arctic MX-2 and MX-4, despite being easy to apply and affordable, aren't worth the trouble they cause later as they deteriorate.


    I work with MX-4 almost exclusively. Yeah, it's not $30 a tube, but it's also not "very cheap," are you kidding me? "Very cheap," is the Elmer's glue you sniffed as a kid, repackaged as thermal paste.

    I use MX4 specifically because it doesn't have a burn in period and because it lasts FOREVER.

    No, it doesn't deteriorate. I've seen reports a decade after the fact showing less than three degrees celsius difference from when it was first applied.

    So. Either you're biased because of ignorance, or both Artic's warranty and every long term test done before this has been lying. Gosh, lemme think which is more likely...

    Now, is something like MX4 the best thermal paste out there? Of course not. But it IS way better than a lot of the market, super easy to apply and maintenance-free, and very reliable. If you're going to be a snob about your thermal pastes, at least be accurate about it.
  5. Well, I didn't see that result coming. They are almost all the same. So, why even bother picking? Just get the cheapest stuff from a reputable name. That's a little disappointing that doubling your money gains you a degree or two, at best.
  6. Within the article you talk about the considerations for GPU backplate for augmented cooling. Do you plan to do a review/article for products involved in that? I would be interested to know which thermal pads or shims or pastes you are using to augment GPU cooling. And to see a comparison of different products. I recently replaced a laptop GPU and redid the pads for that. The installation did involve a learning curve and finding products was not straight forward.
  7. Nice job on this article. Do more of this It helps the enthusiast community.

    Looking at your data Thermal grizzly Kryonaut wins as the best non-metal TIM except in low mounting pressure situations. it doesn't seem to matter as long as you have one of the decent pastes but its obvious there are a few to avoid like the Coolplast20 or Amasan T12 for example.
  8. @DarkSable:
    I'm using TIM since over 15 years, not only for Home PC's, but also in the industry. The major problem of this MX-4 are the long Burn-In time to get a better performance and the fast dry-out issue. As hotter a CPU or GPU works, as worse this grease performs (and is drying out). I does a lot of long-term runs with different products and especially this older products (not only from Arctic) were showing this typical behavior.

    If you prefer MX-4, why not? Use it. But please accept, that a test of different products over 4 years can show at the end a completely different picture. :)

    I get a lot of hardware (mostly VGA) with MX2- or MX-4 as replacement of the original TIM from other reviewers in rotation. And I have every time to replace this replacement with better (or original) products to get the original performance back. MX-2 on a VGA card is pure pain. Simply try one time another, better products and you will be surprised.

    @JamesSneed
    I have to take, what's in Germany on the market. All pastes were retail and not sponsored samples from the manufacturer. It was my idea to do this under real conditions. But I think it is possible to organize some stuff also from the US or Asian market.
  9. With Ryzen and more so Thredripper I wonder if those will impact application methods due to the multiple dies under the heat spreader? Seems you would want to make sure you have the area the dies are covered with TIM and that area is spread out more with those CPU's.
  10. A great article, thank you! :-)
  11. Interesting, I didn't expect my preferred TIM, Antec Formula 7, to do "poorly" as it did. But that's really only in comparison to the exotics. Compared to the standard TIMs, it is slightly above middle. However, of all the TIMs I've seen or used - the Formula 7 seems to have the best shelf-life. I had a few tubes of various other TIMs around - BeQuiet's included paste for the H7, and old bottle of Zalman (like...9500AT top of the game old), MX-4, and others. Of those, only the Formula 7 hasn't had any sort of separation or drying out issues. The same tube I've had for about 8 years now, and its still just as good as it was when I got it.

    Also, its one of the easiest to apply and get good coverage out of.
  12. @JamesSneed:
    This was really not easy to learn. I tried different methods with my chiller (to get at least one stable factor for a comparison in this tests). It is at the end more important to get the thinnest but perfect film without any bubbles and not the biggest possible area/surface. TR is soldered (thx for that) and the IHS is really stable. You need a good compromise between pressure and the amount of used TIM. Mostly is less better ;)

    My best friends are good torque screw drivers ;)

  13. Strange, I'm not seeing inline images for this article (FF 54.0.1).
  14. I would say if folks are concerned with things as trivial as 'paste burn in' then they really need to re-think their lives.
  15. And yeah...images not showing.
  16. FormatC said:
    @JamesSneed:
    This was really not easy to learn. I tried different methods with my chiller (to get at least one stable factor for a comparison in this tests). It is at the end more important to get the thinnest but perfect film without any bubbles and not the biggest possible area/surface. TR is soldered (thx for that) and the IHS is really stable. You need a good compromise between pressure and the amount of used TIM. Mostly is less better ;)

    My best friends are good torque screw drivers ;)




    Thanks for that.
  17. daglesj said:
    I would say if folks are concerned with things as trivial as 'paste burn in' then they really need to re-think their lives.


    I'm not concerned so to speak but before I set my fan profiles up to keep my computer cool and quite under lower loads I do burn in the CPU with prime 95 for a couple hours just to make sure my temps are what they will be after a burn in. I like to Prime95 a new build to check stability, Overclock it, Prime 95 it, repeat until stable. Then I go in and tweak my fan profiles and other misc features(wait on that just in case I have to do a BIOS rest on OCing). Maybe its just me but how I do it.
  18. Thank you! I've been waiting for just such an update!
  19. Yuka said:
    Oh, amazing article. I love it a lot.

    Maybe it's because I've used Artic Silver 5 for so many years, but for me it's the best all-rounder compound there is. Plus it's very cheap. I like it more than the MX-2 and MX-4 compound siblings people usually recommends. But I have to say, the "diamond" compounds are indeed better it seems. I had my doubts, but no more with these tests.

    Cheers!


    One of my disappointments w/ this article is that it leaves out the winner of the last 80 way test like this and that's Shin Etsu which as far as I have seen has the best performance / price ratio out there. My guess is because it's sourced directly from the manufacturer who caters to industrial customers rather than a vendor who buys from an OEM and repackages for the PC industry.

    https://archive.benchmarkreviews.com/index.php?option=com_content&task=view&id=150&Itemid=62&limit=1&limitstart=12

    Shin Etsu equaled AS5s performance in that last 80 way test but AS5 has 3 major disadvantages.

    a) It costs more.
    b) It requires 200 hours of curing time... at 8 hours a day, that's 3-1/2 weeks
    c) The last I will take from A5s home page and concerns potential mishaps.

    Quote:
    Arctic Silver 5 was formulated to conduct heat, not electricity.
    (While much safer than electrically conductive silver and copper greases, Arctic Silver 5 should be kept away from electrical traces, pins, and leads. While it is not electrically conductive, the compound is very slightly capacitive and could potentially cause problems if it bridges two close-proximity electrical paths.)


    Shin Etsu is available from multiple sources for < $4

    It does seem to increase in viscosity over time when exposed to air. Not an issue on CPUs. But when ya doing a GFC card including both sides of thermal pads and both for the back plate and block at VRMs and memory on GFX cards, it can be an issue. Gelid Extreme is good here which is good as that's what is shipped with EK water blocks.
  20. This was a great article with a lot of information. I've been wondering about the specifics of thermal pastes and many of my questions were answered.

    I wish the author would have discussed what their favorite thermal paste to use was and why. The numbers seem to indicate that the Thermal Grizzly Kyronaut paste typically performed the best and was subjectively ranked high up in the usability category. So in conclusion.......... is this the paste that I should buy for my next PC? Or is there a reason I wouldn't want to? Similarly, the Titan Nano Blue consistently fell to the bottom of the charts. Should I avoid that paste at all costs. The article never really says.

    It may seem like I did a lot of complaining, but I want to reinforce that this article was very helpful. Thank you.
  21. @AGENTLOZEN

    I was equally impressed by the performance of the Grizzly Kyronaut, then I saw it was around $12, which is one of the most expensive pastes here. So, you need to ask yourself if spending $4-$6 extra is worth 2-3 degrees. For some, that is trivial, but if you need a lot of it on hand because you build PCs, then that premium starts to really hurt.
  22. InvalidError said:
    Long story short: apart from esoteric TIMs, all pastes are practically as good as any other for typical uses when applied correctly. That really shouldn't surprise anyone as all pastes rely on the same principle of various particle sizes in silicon oil suspension getting crushed together.


    This test repeats the benchmark reviews conclusion that, again avoiding the esoteric stuff, the range from best to worst is about 4-5C. Yes, you can lose the advantage w/ a poor application. But you can also waste the advantage by a cooler upgrade

    A $30 Hyper 212X delivers a temp of 78C in TPUs latest test ... upgrading to a $90 Cryorig R1 nets 72C.... a 6C gain. As we can see from the test you can lose 4C or 2/3 of that gain and 2/3 of that $60 investment, by choosing a poor thermal paste.

    Love seeing the data, and agree that you can lose any advantage with a poor application .... but w/o Shi Etsu products in the mix, its kinda like talking about best boxers, w/o having Ali and Mayweather in the list.

    Monumental effort .... Hopefully it can be expanded over time.
  23. AgentLozen said:
    The numbers seem to indicate that the Thermal Grizzly Kyronaut paste typically performed the best and was subjectively ranked high up in the usability category. So in conclusion.......... is this the paste that I should buy for my next PC? Or is there a reason I wouldn't want to?.


    Its about 3 times as expensive than Shin Etsu.... bit harder to apply... but also sometimes hard to find and out of stick when ya do. Fitting it into the tables... Shin Etsi G751 fits between Gelid and Kryaonaut.... so about 0.2 advantage to Kryonaut.


    daglesj said:
    I would say if folks are concerned with things as trivial as 'paste burn in' then they really need to re-think their lives.


    For me it's the time investment.... I have a box on my workbench, I really don't want to wait 3 weeks and have the user bring the machine back so I can do the final dial in the overclocks.... Day 1 w/ AS5, that 4.9 GHz OC may break the users desired max temp limit....21 days later after curing at lower temps it may not. So with two choices

    a) More expensive paste w/ capacitance and curing issues
    b) Less expensive paste with no capacitance or curing issues

    That's an easy decision. That's likely we see AS5 giving such a poor showing here compared to other tests. When doing tests like these, usually they will do from 3 to 5 mounts to eliminate a paste from being mis ranked..... that's 600 - 1000 hours of curing time.

    So yes, if you are looking at TIM performance and rankings ... ignoring the gain in curing time is a significant oversight as it can knock the product down from near the top to the middle of the pack or worse. I fa do a build and have the patience to wait 3 weeks before dialing in ya OCs, as long as you take this effect into your TIM selection, not an issue. But if you doing 1-3 builds a week, that's a "whole 'nother thang".
  24. JackNaylorPE said:
    Yuka said:
    Oh, amazing article. I love it a lot.

    Maybe it's because I've used Artic Silver 5 for so many years, but for me it's the best all-rounder compound there is. Plus it's very cheap. I like it more than the MX-2 and MX-4 compound siblings people usually recommends. But I have to say, the "diamond" compounds are indeed better it seems. I had my doubts, but no more with these tests.

    Cheers!


    One of my disappointments w/ this article is that it leaves out the winner of the last 80 way test like this and that's Shin Etsu which as far as I have seen has the best performance / price ratio out there. My guess is because it's sourced directly from the manufacturer who caters to industrial customers rather than a vendor who buys from an OEM and repackages for the PC industry.

    https://archive.benchmarkreviews.com/index.php?option=com_content&task=view&id=150&Itemid=62&limit=1&limitstart=12

    Shin Etsu equaled AS5s performance in that last 80 way test but AS5 has 3 major disadvantages.

    a) It costs more.
    b) It requires 200 hours of curing time... at 8 hours a day, that's 3-1/2 weeks
    c) The last I will take from A5s home page and concerns potential mishaps.

    Quote:
    Arctic Silver 5 was formulated to conduct heat, not electricity.
    (While much safer than electrically conductive silver and copper greases, Arctic Silver 5 should be kept away from electrical traces, pins, and leads. While it is not electrically conductive, the compound is very slightly capacitive and could potentially cause problems if it bridges two close-proximity electrical paths.)


    Shin Etsu is available from multiple sources for < $4

    It does seem to increase in viscosity over time when exposed to air. Not an issue on CPUs. But when ya doing a GFC card including both sides of thermal pads and both for the back plate and block at VRMs and memory on GFX cards, it can be an issue. Gelid Extreme is good here which is good as that's what is shipped with EK water blocks.



    I did know that all of that (including curation, which is something I tell everyone that asks me what I use), but unfortunately, I've never really trusted thermal paste that is not available "everywhere". I know it's a poor excuse and I'm not denying you are indeed correct, but AS5 is great for what it offers and you can grab it from Amazon with no issues in the same day.

    I've never really wanted the "absolute best" and I know there's such a thing, I want a good thermal paste that can handle most scenarios well. AS5 proves that in this review quite well.

    I would love some GPU tests using all of these, like some other people has mentioned. I have applied AS5 to some of my GPUs (sans 7970 and current RX480), but haven't really noticed much difference in the long run. I want to know what I'm missing and an article on that would prove very beneficial.

    Cheers!
  25. Nice article, it shows that for the most part, Thermal compounds are fairly close.

    On the other hand, the paste I'd recently purchased, Cooler Master MasterGel Maker Nano, seems to perform consistently well regardless of mounting pressure. I was completely out of thermal paste, so decided to try something new for future builds. Managed to get mine for $9.99 on sale at Newegg last week. Going to be using it on the system rebuild for my mothers computer, Ryzen 3 1300x with MasterLiquid 120 cooler. And I'll probably use it when I get a new case for my build.
  26. I put man small dots of paste on the cpu, smear it around with a tooth-pick until everything is covered. Most of the paste is so thin, it's translucent. Then I place the heatsink on the cpu and while lightly pushing down, give it some back-and-forth twists. If I lift the heatsink up and look at it, there is a very very thin layer spread evenly across the CPU.

    I'm not sure if there any downsides to this method. Could air pockets occur?
  27. JackNaylorPE said:
    c) The last I will take from A5s home page and concerns potential mishaps.

    Silicone oil which is used as the foundation of most thermal greases has a dielectric constant of about 2.8, which makes all greases based on it about 2.8 times as capacitive as the air that would otherwise be there. The exact amount can also vary depending on the conductive and dielectric properties of the particles used in the paste.
  28. What about IndigoXtreme? One of the most interesting types of next gen thermal interfaces around. I would like to see that type of product reviewed.
  29. With such a comprehensive analysis, you really needed to give out some Tom's Hardware Recommended/Approved awards.
  30. JackNaylorPE said:
    but w/o Shi Etsu products in the mix, its kinda like talking about best boxers, w/o having Ali and Mayweather in the list.

    Shi Etsu... I hope you know, that this company and other big chemical brothers like Dow Corning are the manufacturer of a lot pastes I've tested here ;)

    It is the same as a comparison of an Arctic AiO vs. a NZXT or others. You get the same Asetek stuff inside ;)


    @turbotong:
    Thermal grease is difficult to recommend, because the skills and the preferences of each user are very different. Also the financial aspect. Better cheap and usable or expensive and excellent? It is so difficult to say what is the best in a special environment. One wrote here, that I'm biased. No, that's not right. And exactly this is the reason why I never wrote about my personal favourites. :)
  31. Formatc I don't think many realize how few actual manufactures of TIM there are. I wonder how hard it is to note who really makes it like we do with PSU's? I assume Shin-Etsu and Dow Corning make a large percentage of the pastes we see on your 80 way list. I also suspect that is why we see large groups that have almost the exact same results.
  32. Good read.

    Anyone interested in Liquid Metal, take listen to this segment of the review carefully.
    Quote:
    "Liquid metal is suitable for more experienced power users; its application is difficult to master and you may run into trouble with hardware warranty claims, since these "pastes" can never be completely removed without some sort of leftover residue. Given the challenges posed by highly conductive pastes, we'll discuss them separately.


    I used Liquid metal for a while. It performed well. I still use it in my laptop. But it is extremely difficult to clean off after it sits for a while. It also caused me some issues with an RMA once because the markings on the IHS weren't legible. That's the one I think isn't worth the trouble.
  33. the thermal paste for intel is chiller, winter, snow... 2500k 35ºc 3770k same thermal compound 50º my next cpu will be from amd need to be soldered. why the intel don't sell that <language edit> cpu with out the heatspreader

    Mod Edit: Please watch the language
  34. mapesdhs said:
    Strange, I'm not seeing inline images for this article (FF 54.0.1).


    We're looking into this. Upon opening in FF I also don't see the images, but refreshing brings them up. Apologies as we look at this issue.
  35. Xy-graphs plotting the results against price/gram (or price/application) would have been nice.

    As it stands, it's very difficult to get a grasp on the value proposition of all the products, if you're not familiar with them all (and know their prices and package sizes by heart).
  36. Gillerer said:
    Xy-graphs plotting the results against price/gram would have been nice.

    As it stands, it's very difficult to get a grasp on the value proposition of all the products, if you're not familiar with them all (and know their prices and package sizes by heart).


    It would for shops that do a lot of builds as it could add up. The most expensive non-metal one is around $12 a gram about 3-CPU's worth, so I'm not sure it warrants a price breakdown to be honest.
  37. IInuyasha74 said:
    I used Liquid metal for a while. It performed well. I still use it in my laptop. But it is extremely difficult to clean off after it sits for a while. It also caused me some issues with an RMA once because the markings on the IHS weren't legible. That's the one I think isn't worth the trouble.


    This is from one of my friends, he is a shareholder of a cooler company and is every time totally annoyed if they get such so-called RMA cases due liquid metal:

  38. I simply go with what ever tastes the best. Works for me. :p

    Kidding aside, very informative article.
  39. Ive been using arctic silver 5 since early 2000. I only ever bought 1 tube, and it only just ran out at the end of 2016.

    I was considering buying a new tube for my ryzen build. But, in the end i just decided to use whatever came with the noctua nh-15d i bought.

    I didnt have any problems with temps so i didnt give the matter anymore thought till now.

    Good to see that it really didnt matter. The stuff a quality heat sink comes with is good enough to not bother buying something else.


    Thanks for the article. And i love seeing the toothpaste on there!
  40. I am curious what the results of no thermal compound would be. Is it possible to add that to the results?
  41. evdjj3j said:
    I am curious what the results of no thermal compound would be. Is it possible to add that to the results?

    The result would be thermal trip before managing to get core temperature readings. Without TIM of any sort, heat transfer is extremely poor the moment the mating surfaces are anything other than perfectly matched. Even mayo and toothpaste are significantly better than nothing.
  42. FormatC said:
    The major problem of this MX-4 are the long Burn-In time to get a better performance and the fast dry-out issue. As hotter a CPU or GPU works, as worse this grease performs (and is drying out). I does a lot of long-term runs with different products and especially this older products (not only from Arctic) were showing this typical behavior.

    If you prefer MX-4, why not? Use it. But please accept, that a test of different products over 4 years can show at the end a completely different picture. :)

    Funny you should say that. I built a 130 W system exactly 4 years ago, using MX-4. No signs of dry-out, so far.

    I applied using the "blob" method. Also, the cooler was lapped (but not the CPU).
  43. The animation on page three is not real, accurate or proportionate and it triggers any noob on matters related to polished/manufactured surfaces. If you dont know what a polished surface is or the actually blob size on that scale.. then refrain from installing cpus or replying.
  44. FormatC said:
    @DarkSable:
    I'm using TIM since over 15 years, not only for Home PC's, but also in the industry. The major problem of this MX-4 are the long Burn-In time to get a better performance and the fast dry-out issue. As hotter a CPU or GPU works, as worse this grease performs (and is drying out). I does a lot of long-term runs with different products and especially this older products (not only from Arctic) were showing this typical behavior.

    If you prefer MX-4, why not? Use it. But please accept, that a test of different products over 4 years can show at the end a completely different picture. :)

    I get a lot of hardware (mostly VGA) with MX2- or MX-4 as replacement of the original TIM from other reviewers in rotation. And I have every time to replace this replacement with better (or original) products to get the original performance back. MX-2 on a VGA card is pure pain. Simply try one time another, better products and you will be surprised.


    Are you sure you're not thinking of Arctic Silver 5? I've heard a lot about AS5 drying out and losing cooling efficiency after a few years, and being difficult to remove, and its widely known to require a long burn-in. I kind of think you may have mixed up the brands here when writing this. I've used MX-2 for years, and haven't noticed any issues with it's long-term durability, nor have I heard about others having issues with those compounds. And all the marketing info about MX-2/4 states that it requires no burn-in, and should remain durable for over 8 years, which I kind of think other reviewers would have called them out on if it wasn't true. My guess is that the heat sinks you are getting that have dried on thermal compound have Arctic Silver 5, not Arctic Cooling MX-2/4. They're entirely different brands, though some years back Arctic Cooling shortened their company name to just Arctic, which might lead to some confusion with Arctic Silver. At the very least, it seems a bit odd that you would make questionable claims about one specific brand of popular thermal compound without really mentioning many issues about others, or testing the durability of the compounds in any scientific way.


    Anonymous said:
    I was equally impressed by the performance of the Grizzly Kyronaut, then I saw it was around $12, which is one of the most expensive pastes here. So, you need to ask yourself if spending $4-$6 extra is worth 2-3 degrees. For some, that is trivial, but if you need a lot of it on hand because you build PCs, then that premium starts to really hurt.


    The price difference is significantly more than that, when you consider the tube sizes. Many popular pastes with relatively good performance are available for under $6 for a 4 gram tube, while with Thermal Grizzly Kryonaut, you're paying twice as much for only a 1 gram tube. Or in other words, it costs around 8 times as much, at least at these smaller tube sizes. You might get a dozen applications out of a tube of another brand, but probably no more than three out of a tube of Kryonaut. So, where it might cost less than 50 cents to reapply many other popular thermal compounds, this one can set you back around $4 each time. Again, that might be a perfectly reasonable price for someone heavily overclocking a high-end CPU, but considering that this only gets you a degree or two of better cooling performance over options costing far less, many will likely consider it not to be worthwhile.
  45. ummm. It seems to me the tests show The Grizzly Kryonaut is the best solution for non-permanent metal liquid options. Yes, its only a couple of degrees, but it clearly is superior to most other pastes.

    I agree that most TIM compounds are very similar, the evidence shown here is proof of that. I can see where some pastes are overpriced, and with an article like this its hard to determine long term effectiveness. Also some solutions require a lengthy burn in time, notably Arctic Silver 5 which requires periods of high use and down time over a two week period to set in.

    But clearly there are some superior products which deserve a premium price.
  46. buzznut said:
    with an article like this its hard to determine long term effectiveness.

    For pastes based on silicone oil and zinc oxide, aluminum oxide, diamond, carbon and other chemically inert heat transfer media, the working life is practically unlimited as long as mechanical shock doesn't fracture the paste after it packed down over the course of a few years. (That's where you see people complaining about their CPU termperatures suddenly shooting up after dusting off the CPU fan, changing RAM, replacing the GPU, etc. because they knocked the CPU while doing so.)
  47. Hmm good roundup, could have used this a month ago when I was shopping for paste. Based on other roundups and pricing/availibility in my country, I went with Coolermaster MasterGel Maker NANO and it still looks like a good choice here.
  48. Can you add the stock performance to GPU charts?
    I realize manufacturers would use various different solutions, but they are likely to end up in a similar quality bracket or we would have already heard about one company out-cooling others significantly...
  49. If you're not seeing the images in this article then try a different browser.
    No images were displayed when I used Edge, but they all displayed correctly in Chrome.
Ask a new question

Read More

CPUs Thermal Compound Components DLD