Thanks everybody @LTT for the testing! The methodology was fair and it's good you used Noctua's best thermal paste with their own NH-D15 cooler as a benchmark. We've modified the test rig and are now formulating a liquid metal alternative, stay tuned.
Matching an R&D marvel of an experienced, specialized top-tier manufacturer is beyond impressive, and it seems that's just a start. You guys rock and these kinds of videos are SO much more fun and informative than another computer building. More power to you and please more videos like this.
They are. Go have a look at their videos. :D They do all sorts of stuff! Air con to electro-hydro-dynamic drive*. Rum to rocket motors! * As made famous in _The Hunt For Red October._
I'm convinced that LMGs interview process involves a test where the hand a candidate something on camera to see if they drop it. If they do, they get hired.
I don't see the point in not spreading the thermal compound with a spatula though. Wouldn't it be logical if spread is an issue to manually do it yourself? That is the way I do it and it drives me nuts to see that no matter the application there are often dead spots left behind.
This is here for people who don't realise this is a joke, and need an explanation as to Alex's methodology. By using a case, they would be introducing another variable; this way they can control the ambient temperature. If the PC was in a case, other variables would contribute to temperatures, skewing their findings (admittedly unscientific).
I’ve just realised that when Alex is not trying to dumb things down for some lowest common denominator he becomes a much better and more natural host. This ironically makes for much better communication.
To be fair, that guy in the shed who made this isn't normal. He's incredibly intelligent, detail oriented, meticulous, almost obsessive. His video on this popped up in my feed recently, but I haven't watched it yet because his videos take my total attention and his video on the thermal paste is over 40 minutes long.
Dude, the guy has in his "shed" more stuff then some labs have. While he is a bright guy, he has the tools too. Those industrial mixers are pretty expensive.
He did have a serendipidous "accident" when selecting powders for heat conduction, I am willing to bet not many in the "professional" Thermal Interface Material field have stumbled upon it, because it seems so counter-intuitive at first. This is probably what gives their paste an edge, even if it is not perfectly mixed as it could be with proper, but really expensive equipment. Watch both videos of TI would be my recommendation
Grandpa: Builds a complex setup to minimize variables in order to say if stuff is good or not LTT: We don't care about that, we got REAL WORLD *LTT does real world test* LTT: Our setup is so shitty, we have no idea which one is actually better 🤷♂️
@@nlingrel No they were correct in saying that. There is literally a 0.9 degree difference in the average, and the noctua's lowest run was lower than TI's lowest run. Considering run to run variance you cannot conclusively say one is better than the other.
That's kinda the point though. For a real world setting you can expect about the same performance from the two. Doesn't really matter if one is like 0.01% better in a tightly controlled experiment.
They've probably tried harder for consistency than any actual person applying thermal paste to their PC. Goes to show that incredibly marginal gains don't really matter.
I'm really looking forward to getting an order with Tech Ingredients. Even if it would have been worse (withing a few degrees) I would buy to help support their shenanigans.
@@joshuascholar3220 the "easy to make at home" one was glycerine. The high performance one was some specialist silicone oil, yes we still don't know if it will evaporate, but they did say in the video that they tested that
@@joshuascholar3220 I'm not certain, but I don't think glycerin does much until it's heated to its boiling point, much hotter than it would get in this application. Anyway I'm also curious how long it lasts.
Ok, PC Jesus needs to get involved in this. If Tech Ingredients Grandpa was to build a case with an integrated fluid cooling system I suspect it would be the most BOSS cooling system ever.
That... is actually quite likely, given that the thermal paste videos are on his 'Cooling' playlist: ru-vid.com/group/PLzrI14lOlSqeS5pNkRCiV1-yhoBlCmxms
@@chunye215 you don't see many older people use RU-vid. So if you have gray hair you're pretty much a grandpa on here. He probably could be a grandpa too by now tho.
@@Kiritomens You don’t think Gen X folks waste hours and hours on RU-vid? You don’t think Gen X folks don’t have a TON of active channels? If you watch this, I assume you know who Jays2cents is?
No offense but Tech ingredients deserved a better analysis of their thermal paste, I was excited to see this video but it feels really "thrown together at the last minute" thing, I just feel Gamers Nexus would have done something way better
could be an "ok linus we'll do it your way first and let the comments agree with me that we need a better control environment before we spend $5000 on making a thermal test control room in the warehouse and test for a few weeks with no content coming out of it in the meantime except a build vlog while you pay our wage still" thing, too. honestly though, it's "good enough" (in my opinion) to verify (for a future vid) that maybe testing needs to be done with a de-lidded and liquid-metal-repaste, incase the die-to-ihs tim is bottlenecking both pastes' abilities, and make linus go "OK FINE DELID THAT SUCKER" but then again idk how much a benefit, if any, liquid metal has over the die's tim, since iirc intel uses solder now
Indeed. It's not just the test environment but the script (or possible lack thereof) being very disorganized. I haven't seen anything this bad from LTT since the Anthony vs Linus repair challenge.
Yeah, seems like a rather rushed video, and no real thought put into any reasoning, I can almost predict that the increase across the tests was due to heat soak from the other components/case, looks like the tests were carried out almost imediatly one after the other only stopping to switch paste. Not really a fair comparison.A haphazard video and I would really hope for better from an LTT video.
1:18 kinda supports this statement. GN has the proper testing apparatus and solid testing procedures. The only issue with GN is that they are sponsored heavily by Thermal Grizzly. While I think the majority of viewers, myself included, would trust them, it's still important to consider.
He is great because unlike most he runs down everything in detail, alternatives, and why the alternatives might work or not and why he chose what he did. its less of a "heres how to do it" and more of a "how to do it, and why so you can adjust"
Some thermal pastes lose efficiency over time as they dry up & harden, I would like to see them do a 1-month long "burn-in" test to see if it actually holds up against the competition.
Over the years I have had the opportunity to do that with a few big name pastes, although my tests have all been at least 1 year. Here is the order of the most stable paste, as well as their thermal ranking: 1. Coolermaster MasterGel Maker (3rd) 2. Arctic Silver 5 (4th) 3. Noctua NT-H1 (2nd) 4. Noctua NT-H2 (Tied with 1st) 5. Thermal Grizzly Kryonaut (Tied with 1st) My recommendations if you want the longest lasting paste with very acceptable performance, I have ran chips with Coolermaster MasterGel Maker for 5+ years with no change in performance. Noctua NT-H1 is the best compromise between performance and longevity, it will perform well for 2+ years before it begins slowly getting warmer. Thermal Grizzly Kryonaut and Noctua NT-H2 are interchangeable performance wise, but Kryonaut is extremely unstable and falls behind NT-H2 after about 6-8 months on average depending on how hot the paste is getting. Noctua NT-H2 is not as stable as NT-H1, it is also more prone to pump out (although not as bad as Kryonaut is), NT-H2 averages about 1 year or 1.5 years before a new application should be considered. The most valuable lesson I've learned from my own testing is that Kryonaut should absolutely not be used in a 24/7 system. It's excellent for sub-ambient and bench testing, but in a system that you're actually using for long periods it is a waste of money.
The bigger factor is thermal creep. The silicone oil suspension medium both under pressure and experiencing thermal expansion cycles tends to creep out of the suspension and dissipate. This is a major problem with thermal pastes for aerospace where wild temperature and pressure swings exacerbate the problem but it's still a problem at normal temperatures over time. While there is some evaporation of silicone suspension mediums, most of the drying you see with old thermal paste is due to thermal creep.
I can add MX-4 that works fine for a long time between IHS and a cooler, but once put under the lid and it sees around 90C temperatures, the oil separates from the paste within 1-2 days
Some people (not you Rob) need to learn to lighten up a bit and understand a joke. If Tech Ingredients got upset about it then I'm sure the LTT staff would apologize but it's not meant to be insulting or mean beyond just being a joke. I'm 32, if a15 year old called me Grandpa or old man I'd probably call them a little brat or something, all in good fun (unless they got physical with me or really carried it too far which LTT has not).
I totally believe that TI has made good performing paste, but what I'd really like to see tested is the longevity. I don't recall what maker it was, but paste performed well for about year and then temps began to increase dramatically. Opened up the machine and paste had completely dried up. I've never seen this tested.
@@barnett25 That's excellent thing to hear and I'm really glad they've considered this issue. Gotta check out their video. They definitely earn and need all the views they can get. 👍👍
Yeah, shelf live and longterm stability. Those are way more important than 1-2°C for the first few weeks (aside from extreme overclockers). Best example is the thermal paste that OEMs and Intel are using - by far not the best in terms of performance, but they still do their job even after 5 years.
@@ABaumstumpf I fully agree with that. I don't want to open laptops etc. to replace paste. Writing this comment with older business series Lenovo laptop that was bought from reputable distributor. Sales information said that they change thermal paste for models they sell. That's now been about year ago. Either they didn't change the paste and it's now failing or they've used some really poor quality one. Anyway looks like I have to do it to this myself 🙄
That would be thermal grizzly. That's why once in an interview with Steve, Derbauer w̶a̶n̶t̶e̶d̶ recommended you to re-apply your thermal paste every year.
Agreed, PC Jesus and Co is a much better choice to go with testing. More than disappointed with the condescending tone. TI's "shed" is much better equipped than LTT from what I've seen.
He's still unboxing the blower test bench crate, probably needs to be calibrated again on site for verification the shipping and installation didn't screw it up.
Problem is, the comments would've probably imploded with people telling them spreading TIM would create air pockets. It's just easier and more consistent to do it the way they did.
@@hamajangz9586 It is easier and probably more consistent because of it, but not necessarily ideal. I've seen very little spreading due to heat in TIM myself.
@Mario Kart Manual spreading was there before heatspreaders and it is still preffered for direct die. (now mostly gpus). With direct die if you don't get all the corners pasted you risk hotspots. (and 20years ago cpus didn't throttle or turn off they just released magic smoke and died). And if you take off the cooler to check the spread and plonk it back the chance of trapped bubbles is probably higher than with carefull spreading.
Yeah. I mean, I wasn't expecting as good testing as GN would do here but this video was... like someone said "hey you go in the back room and make this video we don't want to make." Then rush released it. I was excited to see this video. Now I'm disappointed.
So every time someone does a thermal test these days there is a dogpile of people that come from n to say "Steve would have done it better". Please stop. You aren't helping Steve is many things, an excellent and forceful advocate for consumers, and an intelligent and thoughtful journalist first and foremost. But he is absolutely not to everyones taste, and frankly I find Alex and Collins video more entertaining and more compelling than the video Steve would have produced. Yeah, it's less scientific, yeah there isn't as much data as there could be, but I can also say as a field engineer that their test more closely mimics the reality an end user of the product would likely see. It's fun, and it meets the LTT audience where it's at, and doesn't demand that I engage in rigorous examination of the methodology. There is 💯 a place for that kind of content, and I am great ful for the rigor that Steve puts into holding manufacturers to their claims, but not everything needs to be addressed that way, and constantly chiming in on other channels to tell them that GN does it better does not reflect well on Steve's audience. Commenters like you are a large part of the reason he has to go out of his way to tell his audience not to go review bomb other RU-vidrs.
@@isl1ngt0n The reason isn't taste, it's rigorous process and equipment. In regards to this stuff specifically, its because GN actually have tailor made, proper lab test equipment for this. That enables testing with far more variables under control. Meaning the results you get _much (!)_ more closely represent the merits of the actual product you're testing, instead of having tens or even hundreds of other variables affecting your results, many of which you have neither control over nor data on. Also if you don't like GN's or Steve Burke's "style" / persona / presentation, I would like to point out that Gamers Nexus was a _written articles_ hardware testing and news site long before becoming a RU-vid channel. And more importantly, _they still are._ If you would like the GN testing data/results, or a review or hw news piece, but don't like their RU-vid stuff - GN still publish almost all of their content in written article format too. Good articles, not just video version script copypasta. I would suggest giving one of them a chance if you are inclined to such reading, there's definitely gonna be some interesting ones on Alder Lake, Z690 and DDR5 the next handful of days. While I personally enjoy the YT stuff, I often check their articles when wanting/needing a review's test data. For that, I personally find the article format to be more easily digestible. The huge clusterf**k of uncontrolled variables is the challenge with these kinds of uncientific, "real world" comparison tests. Practically no one else is gonna have an identical setup as yours. Here, we have a huge amount of variables you cannot account for, and _a sample group of one._ If you want to get reasonably clear data from "real world" tests of something individually applied, assembled, configured and operated by users, you certainly can - but you need gigantic sample groups. Only one sample of a "real world" test like this, is the measured and statistical accuracy equivalent to the seat belt analogy, here: "Seatbelts serve no purpose and make no difference, because _I_ drove to the store _once_ and did not crash". Ps. Please forgive the typos, I'm sure there's several. And have a great day.
For test results to have value, you have to establish what you are testing for. This is a test whether or not it matters if you use A or B in a real world scenario. It doesn't. I have never spend money on after market paste, because ever cooler I have bought came with good paste. If I need after market paste, I'm going to spend 8 bucks, and not worry too much about whether another paste might give me a result that is 2 degrees Celsius lower,
Yeah. I mean, I wasn't expecting as good testing as GN would do here but this video was... like someone said "hey you go in the back room and make this video we don't want to make." Then rush released it. I was excited to see this video. Now I'm disappointed.
@@mokahless The effort behind this video was actually kind of embarrassing to watch for being a follow-up to those Tech Ingredients videos. LTT couldn't even reproduce their own numbers for the noctua paste, so their conclusion was a shoulder shrug and "meh good enough." This level of phoning it in is more fitting of a teenager flipping burgers at mcdonalds.
These tests mean nothing if the stuff degrades over time and that would take years under load to know. Manufacturers have to be conservative and not manufacture paste that will shit the bed within a few years.
Honestly i most probably wouldn't have known about this unless LMG made a video. I love gamers Nexus but honestly i don't really watch unless I am interested in a product.
I think this was as much effort as needed to understand how it would perform in the real world. It's all fine and good to test this in isolation, but in reality none of us are going to intentionally replicate even this level of consistency in our own homes. It might be the best thermal paste _in theory_ in a well controlled environment. But in real world use, it is at best at least as good as the best thermal paste on the market, and at worst offering less value per unit than its competition because TI doesn't have the resources to produce the volume themselves that would compete with the big players.
Steve's tester is probably about the same as Tech Ingredients, so if TI said theirs was better then so will Steve. But doing a real world test like this one puts that in perspective. Does measured a little better matter? Real world test says "no."
@@UndeadSoulOfPopeBenedict Did you not notice the low quality of the editing, possible complete lack of script and direction in addition to the bad testing methodology? It appears like a rushed video, ala the Linus vs Anthony repair video. I hope LTT can get their shit together and not keep releasing low quality content.
Can you test it being a bit more generous? I mean you can totally account for less spread with more paste, right? You wouldn't test it against a pea-sized piece of thermal pad either ;)
After all these years how does LMG not have a temperature controlled booth for thermal tests like this? Make it sound proof too to do noise testing. It only has to be like closet sized, you have plenty of space.
The name Linus MEDIA Group explains it all.The entire purpose of the company is to make videos.They dont have the space or reason to get that kind of equipment.Its kinda obvious when you see how much space and equipment they dedicate to video production.If you are watching ltt videos expecting they will do a proper job explaining and comparing things then you are watching the wrong channel.
@@pesho9971 They produce tons of content whenever they upgrade their facility. Any upgrade can be made into a video. Temperature controlled booth ("testing upgrade") could easily be one of them.
@@barnett25 I watch ltt more for the hey this exists gamers Nexus has set a high bar for testing and it just seems ltt is more focused for the casual but they seem to not really be interested in even trying to meet thebbat for testing like come on how hard is it to do better testing the have the manpower and equipment they just need to put in the effort. This glosses over the cost for the man power and time but I don't expect them to change they have their segment of the tech space and Steve has the other segment but they have halfway decent projects that are cool for time to time.
They don't take it seriously that's all. If they were enthusiastic about it, Linus would do it and do it better. but obviously here no sponsor so.. LTT is so big, they've forgotten their beginnings.
Focusing on spread and intentionally applying too little paste seems like a weird test to me. You can always just apply more, it's totally fine. With proper mounting pressure, you really can't over apply it.
True. I'll never understand why people add so little paste it's never covering the whole IHS. 100% coverage always dissipates more heat than just a tiny spot in the center.
As Steve from GN has shown us a couple of years ago, even before the whole Verge PC building debacle (where people derided Stefan for the application, it wasn't pretty but it was serviceable).
@@tim3172 I have no idea what sort of series I had. But I remember it was the last model before they phased it out to get the new models in. Strange though, what I heard about AMD at the time was completely opposite for me. Those were the ones that had a lot of thermal issues, that's why almost everyone preferred intel. It was 3rd hand info, but that's what I heard. I did have issues with Radeon cards though. Every single time I buy one to replace an old one, I keep getting re-reminded why I had Nvidia instead. It was always a massive source of headache for various reasons.
@@noiJadisCailleach it depended on which P4 you got. The Willamette cores were decent and Northwood were the refined later gen P4 chips after the disaster known as Prescott.
Three pastes for the Elven-kings under the sky, Seven for the Dwarf-lords in their halls of stone, Nine for Mortal Men doomed to die, One for the Dark Lord on his dark throne In the Land of Mordor where the Shadows lie. One paste to rule them all, One paste to find them, One paste to bring them all, and in the darkness bind them, In the Land of Mordor where the Shadows lie.
Back when Athlon CPUs were new and overclockable with a pencil, Arctic Silver 3 (i think this was the version) was the premium thermal compound, The prevailing procedure was to apply tiny tiny amounts, with the theory that it was just there to fill in the (air) gaps between surfaces, and heat transmission still relied predominantly on metal to metal (well, metal to chip die, technically) contact, not on a distinct layer of paste. it worked, and this was before heat pipes were in air coolers. Now we smear it on like we're icing a cake... interesting how things change.
Because the smarter voices finally prevailed over the ignorant ones. Metal-metal contact will happen because the mounting force will squish out the extra thermal paste. The tiny amount of extra metal contact you get by using microscopic amounts of paste is outweighed by the benefits of full IHS coverage you get by applying a lot of paste in an X and the drawbacks of manually spreading the paste and introducing air pockets.
@@angolin9352 not saying current methods are bad (or good), just commenting on how they have changed. At the time, I distinctly remember seeing tests that showed thicker amounts of paste led to poorer performance. I know paste and cooling tech has changed a lot since then, not to mention the presence of an IHS.
ikr! I watch LTT from like beginning and remember themselves tried diy thermal solutions and they failed horribly. come to think of it, maybe that was their point. to fail and show these company products are better (they're been sponsored anyway).
@@noiJadisCailleach Maybe he needs work then. Let's hope videos like this are getting proper internal feedback and Linus is simply pushing his employees to practice making videos in the real world, rather than this being an example of some kind of LTT decline in quality. This video really was poorly put together, not just for a lack of testing methodology. That Linus vs Anthony video was rough.
Wait... they signed that letter as Eric and Alex? So I guess we finally found out the presenters' names? Assumedly the main presenter is Eric and the oft-cameraman and occasional second presenter is Alex.
The paste properties seem to be intrinsically tied to the mesh sizing of the Aluminum powder added. Specifically it needs to be a combination of 3 and 5 micron Al beads. I may have misremembered the exact mesh sizes however the idea is that you have one effect from the 5 micron and another effect from the 3 micron, and then the 2 sizes talk (ie. exchange heat) with each other. The most interesting thing is the emergent properties that arise from sizing control.
@@deelkar Yeah, it definitely looks like a size/mechanical regime of thermodynamics. I wonder if someone could convince thermal paste to grow dendrites? That would be cool, waiting for a battery reaction to activate your thermal compound.
@@deelkar Also, silane solutions are very interesting. The VAPSA research I was involved with was heavily into the stuff. Not my department though. We were all about the ASAP (Accelerated Surface Area Porosimetry) research. Nothing like loading a steel ball into an instrument, wait 10 minutes, take ball out, wait 10 minutes, repeat until you are completely mad and there you go it's Calibrated. Stupid ASAP instrumentation!!!
It’s not that hard to come close to the top consumer brand paste. Buy a 1kg can of thermal interface material from a chemical company such as Shin-Etsu, repackage it into syringes and there you have it. Dirt cheap and works extremely well. Consumer brands these days are all about fancy packaging and marketing, there isn't really much R&D involved. In fact I know some of those top brands actually only do repackaging.
Why wouldn't reducing the amount of IC diamond you put in the center and then adding a little extra IC diamond closer to the corners outperform the Noctua paste, if the only factor is spread?
Or using the credit card method. It sounds like for Noctua, they've got the single dot application down but it's not as taking a minute to spread the paste is a big deal if it can improve your temps.
Spread isnt the determining factor in performance, thinness of the layer is. For the same volume of paste, and compression force, spread is an indicator of how thin the layer is though. It basically works out to "the paste stops flowing outwards at x PSI of pressure" If a paste is so viscous that it doesnt squeeze into a thinner layer, then boosting coverage by using multiple dots wont be as significant as switching to a less viscous paste that can spread into a very thin layer. I.e. if a paste has twice the bulk thermal resistance, but makes a layer 1/4 as thick, it can have half the actual thermal resistance of the original "better" paste, because there is less of it in the way. But to have a thinner paste, you need more carrier oil and less thermally conductive filler, so its all about balance. I.e. as said in the video.
It's midnight in India and I am a literature student, yet I'm watching this. I don't know a thing about computers or chemistry. My laptop is kinda not working and I miss it and the guy who was supposed to come and fix it has been putting it off on the next day for like three days now. I am SO pissed at him rn but I can't say a word because that minx is the only one who can fix it. FUCK On that I have to take an online test tomorrow and my parents have planned an outing EXACTLY tomorrow and when I asked them to postpone it, they clearly denied saying God ain't gonna wait for you-they were taking us to a temple which is good, but not on the same day when you've got two novels to read, a test to take and also an important class which won't even be recorded. So, ya! Thermal paste to stick or idgaf melt glass... WHATEVER ig i need a diary
The spreadability is kind of interesting. Back in the day, it was all the rage to use a credit card to spread out the compound as thin as possible. Knowing what I now know about applying thinset for ceramic tiles, I think a micro notched trowel could be used to deliver the absolute best, most consistent spread every time. Someone needs to make a miniature notched trowel. :) :)
I don't know about you, but a consistent 2C difference is about the result I would expect. After testing this stuff on laptops, I can confirm it results in the lowest temperature by a very small margin compared to things like kryonaut and syy-157. Very pricey for the amount you get though. Can't beat gd-900 in value.
Woo! I was waiting for this. It's a really impressive result! Funny how one is made in a highly optimized factory and another was ground up by hand with a mortar and pestle.
If you aren't overclocking to the point you have to worry about thermals with a cheap aftermarket cooler, thermal paste isn't that important. I'm running a 4770k oc to 4.4ghz, slapped a hyper 212 Evo on it. The thermal paste I have is shitty and degraded and I still don't ever go over 80C, and my fans never spin up to be noticable over the sound of a HEPA fan on low I keep in my room. The majority of all this technology stuff is trying to make you think you are missing out if you don't buy the new thing, these videos make you feel like you are behind the curve if you don't spend a bunch of money. But even in places here the tech is important, like digital photography, even the professionals aren't using the newest stuff.
Bold move from Linus to do a Square space sponsership, knowing that they fund a military occupation that tear down cemeteries to build a theme park.... Didn't expect that from you, Linus.
edit because i dont like how YT bolds text inside asteriks: < looks at the 5ghz core lock on all 10 cores on his 10900k < looks at the coolermaster h100i cooling it What a beast!
Super proud of seeing Tech Ingredients in LTT content ! :D Well deserved they're doing amazing quality content ! Anyway, am I the only one using the credit card spread technique on thermal paste or what ?
It's a lot better than most. And Steve's set up is probably too similar to TI's - this puts "measures slightly better in a calorimeter [or whatever it's called]" into context.
5 micron grinding stone dust and glycol, The dust does nothing but stop the glycol from leaking. pull 5ml out of the car radiator and boil/steam the water out, add the polishing dust to whatever viscosity it is that the cooler surface works best with.
I disagree. LTT should play with this but... WTF was with this video? Rushed? Low quality? Oops our results weren't as tightly contained as expected release anyway?
Engineers note: The thermal gradient between cpu and cooler depends on a lot of factors. Thermal conductivity tends to be pretty unimportant compared to the two main factors: 1. Thickness of the thermal interface. If your paste has a worse thermal conductivity, but spreads significantly thinner, it will easily outperform. 2. Adhesion. As shiny as the surfaces may look (cpus are usually everything but shiny), there are a LOT of microscopic gaps formed between the thermal paste and metal. Certain additives and a fine tuned viscosity make all the difference by eliminating these air bubbles and maximizing surface contact. An extreme example: Graphene foil is a spectacularly awful thermal compound, despite the FAR better thermal conductivity than any thermal compound. The issue is the interface between the graphene and metal (and also the thickness of the foil). It makes very poor contact, only touching a fraction of the cpu and heatsink surface.
So glad I checked both the description and the top comment. I knew there HAD to be a Tech Ingredients video on the making of this thermal paste. I love those guys!
You guys need to revisit the pads I am a converted I used a thermal grizzly pad and the thicker pads to redo my laptop for a 3rd time. It is so much better than the pastes I tried before.
When we're talking any thermal paste, there is mostly 1 degree difference between the best non elec conductive stuff and an average one. You need really stringent testing to get good results on this stuff.
Yeah that's just blatantly fkn wrong lol. From Guru3D's article on it earlier this year than can be a nearly 10c difference between various brands and TIMs.
I'm disgusted by LTT's double standards. they could've done a better job to test this, if this was a sponsored product they would've done plethora of testing and get it right. didn't see an ounce of respect towards another youtuber who idolise them enough to send their best product aside from the constant mockery. shame on you Alex and LTT.
I know this is an old comment, but. There is no double standard here. A test with perfect mounting pressure and temperature and all will never be the real world. The reals world usage will not always have perfect thermal spread over the IHS, or perfect TIM spread. Just like benchmarks with synthetic loads, the testing they could have done would never be achieved in the real world, creating something that is not what will actually be seen and experienced in the real world use cases.
Odorless mineral spirits makes an excellent surface cleaning solvent! It will strip off stuff that alcohol can't touch, never leaves any residue and takes only a tiny amount to be useful. By odorless, it means that no odor remains once it has evaporated. It has that 'summer barbecue' smell...mineral spirits are often used as charcoal lighter fluid! Be sure to use gloves, as this stuff will remove skin oils.
Alex's testing methodology was so poor it had margins of error greater than the differences between pastes. I can't believe they would post a video like this. Doesn't do Tech Ingredients any justice.
All the more reason to use the spreader method of putting on thermal compound. You just don’t know how well a certain compound will spread if you just do the dot or x.
When I first started building computers 25 years ago, we didn’t just rely on the pressure from mounting the cooler to spread out the compound. We spread a thin layer of compound across the die with a flat item.
