How to Build a PC Using the Most Efficient Direct-Die Waterblock in the World! (i9 14900K)

[@jimbodee4043]

The block and design looks very impressive and professional, unfortunately I am an AMD person so will have to wait to see how your AMD block development gets along.

[@emcoolinc]

Hey! Appreciate the thoughts! Building an AMD block for their next-gen chip is a priority for us, and we'll make sure to cover the process on the channel along the way.

[@toonnut1]

I've delidded a few cpu's im using direct die on my 13900kf right now. If the block is copper then I do 2 layers of liquid metal with a 30 minutes gap between the 2 layers. You really should use nickel plate instead of copper for better use with liquid metal

[@ahnilatedahnilated7703]

liquid metal should NOT be used on a copper block. It will be absorbed into it and cause issues. Always use nickle plated copper.

[@toonnut1]

@ahnilatedahnilated7703 that's WHY I SAID use 2 applications the 1st absorbs the second works the same as anything else. YOU CAN DEFINITELY USE IT ON COPPER ITS JUST NOT AS GOOD AS NICKEL PLATE! and that's why I ADVISED to use nickel plate in the 1st place. Your simply wrong to say copper CAN'T be used.

[@emcoolinc]

Hey guys! Thanks for commenting. There are a few articles debunking this topic, which were not discussed in the video for the sake of length. But your opinion is important to us and also gives us a good idea of what the customer perceptions are here. If the demand is there in the future, we are open to offering a nickel-plated solution as well. That being said, here are some of the underlying engineering principles behind bare copper’s effectiveness: 1) Bare copper with gallium (liquid metal): Copper reacts and alloys with Gallium, creating a really thin layer or stain, but this is NOT corrosion and the thermal effect is minimal due to the negligible thickness of such layer. 2) Nickel-plated copper with gallium: There is also a reaction but it is slower and therefore the staining is reduced compared to bare copper. In both of these cases, the thin liquid metal layer remains between silicon and bare copper or nickel-plated copper, regardless of the staining (thin film alloy) since the liquid metal, at least in the proportions we use in the video, is more than enough to both create a thin alloy and a liquid layer. Our first prototypes date to 2018 and have two systems daily-driven like this with no significant effect on thermal performance. Your comments gave us an idea to do a future video showcasing both units, which in 6 years of operation have never been serviced since they don’t need to. Finally, here is some more detail by Steve from Gamers Nexus, who did an amazing job testing both of these cases, and in his words and own conclusions: “Gallium has a negative potential and copper has a positive potential, which will cause the gallium to migrate and plate the copper. In terms of performance, the copper RETAINS all of its original performance characteristics (in our testing), and so looks a lot worse than it is in reality. This is a stain, NOT heavy pitting or corrosion.” Around min 13:00 on his YT video: kzbin.info/www/bejne/mYLEooivgLt8e6c Again, appreciate the comment and we will keep this in mind moving forward.

[@der8auer]

@@emcoolinc Hey! I can't agree with you there. Corrosion of Gallium on copper is a well documented phenomenon (unless you use something without Gallium which doesn't seem to be the case here). It has nothing to do with the negative or positive potential. There is a nice publication about this topic in the International Journal of Thermophysics. You can find it in Volume 39, article number 113 from 2018. Grain boundary embrittlement is especially a problem on aluminium because the grains are so much bigger than on copper but it also happens here. Amalgamation is the much bigger problem weakening the copper and forming brittle areas on the surface which much reduced thermal conductivity.

[@jamegumb7298]

@@der8auer What other liquid metals are there for use in cooling?

[@hangemhi001]

Outstanding

[@shadowarez1337]

I got the non wc version of this 2 infact one for the 14900K and my back up 14900KF rig. Amazing that it took Der8auer no time to create something like this and seriously tame the hottest CPUs.

[@avantijoshi4380]

Excited to keep learning about this!

[@emcoolinc]

Hey! Thank you! Appreciate the enthusiasm.

[@JDD_Tech_MODS]

Would love to see you test your design against ICEMAN Direct die waterblock and DerBauer's direct die block.

[@emcoolinc]

Hey! This is something we are very open to, but in an effort to remain impartial we would prefer another independent channel to benchmark our product. Along those lines, we are actively looking for third-parties to do this and look forward to independent reviewers covering this on their channels. However, we will definitely still have videos showcasing extensive testing of our own EMCOOL systems. Thanks for watching!

[@Sectorz]

​@@emcoolinc send it to GamerNexus not LTT !! But do your own test can help to know of it really better, same perf ou worst at 1 or 2 degree. Don't forget optimus V3 WB it the best at this moment for Direct Die. For the price/perf it the Supercoolcomputer one.

[@tulasisarge]

Very intresting topic posted as video......great work man !

[@emcoolinc]

Hey! Glad you liked it! Thanks for watching!

[@amirthry]

Very Nice /.

[@emcoolinc]

Hey! Thanks for watching!

[@Peterlan15]

Why Intel and mobo manufacturers didn't catch these issues before release is beyond me. I've been reading the new arrow lakes will use less power but offer lower gaming performance (fps) than this 14900K, dissapointing if that is truly the case. But we'll see in a few days...

[@emcoolinc]

Hello! Thanks for watching our video! Intel and their partners really missed the mark here. While the i9-14900K is a powerful chip, its high power consumption means it requires more than conventional liquid cooling to perform reliably. We're excited to get our hands on one of these new CPUs and put our cooling system to the test in some benchmarks!

[@wilkuuu4016]

Arrow Lake is new base for future performance improvements without heating your whole home at this same time.

[@reynoldoramas3138]

The liquid metal does not corrote the bare copper in the cooler plate?

[@glubrix]

No , but it joins his atoms with copper. it is normal

[@C0smicr3aper]

It fuses to the atoms in the copper, why a nickel plated block is the best one for liquid metal.

[@jamesp206]

You can "season" the bare copper with liquid metal/quicksilver before installing, like RockitCool recommends. This helps a lot.

[@emcoolinc]

Hey guys! toonnut1 started a thread below with a similar discussion. Would advise taking a look at that as there's some good commentary on this topic! Thanks!

[@thomaswilson2079]

great product and video but dude presenting, try talk clearer and stop saying Ugh after ever few words. Im not to insulting you, just trying to let you know the video would be more enjoyable.

[@kingbradley6602]

Will this be available in europe at some point ?

[@emcoolinc]

Hey! Thanks for your interest. While we're primarily shipping in the US, we're working with our shipping partners to figure out international options on a larger scale. In the meantime, if interested in one of our products, please reach through the form/chat on our website and we'll be happy to work with you on shipping. Thanks! www.em-cool.com/

[@PhillipJermakian]

My black cat loves sunbathing even when it's 90ish outside. His oily thicker hairs disapate a lot of heat. My point is it's great that you are innovating but patents are such a sad system that stifles progress and it is insane that something like "more surface area" can be patented when it's at least as old as black cats=p

[@glubrix]

Good job

[@emcoolinc]

Hey! Thank you very much!

[@TeamStevers]

Rog and asutek’s 3d printed water cooling plates are better.

[@LeBurkaTron]

Great vid. maybe cut the . Uh , Uh .

[@emcoolinc]

Hey! Appreciate the feedback - I agree and I'm working on it :) - Daniel

[@jiichamaa]

"Ugh..."

[@ddzang]

What’s the benefit of direct die?

[@DIMKAZDS]

You say you have a patent on some technology. What technology? it is a basic pin waterblock.

[@emcoolinc]

Hey, We posted details on this on another comment, but summarizing it here for visibility: Happy to provide some details on the granted patent - feel free to read more here. Patent name: System and method for providing direct silicon footprint microfluidic cooling for electronics. USPTO Patent number: US12080626B1. Here is the link: patents.google.com/patent/US12080626B1/en For avoidance of doubt, we’ve been working on this research for years during our time at Georgia Tech and feel extremely comfortable with the engineering backing of our product. If you have any technical questions, feel free to ask them here and we’ll respond. Regarding your other point citing basic pin fins, a pin fin array or geometry can not be patented as it is common sense for anyone skilled in the art. However, increasing the density of such pin fins is a challenge in order to provide more surface area for cooling when confined to the size of the chip. Our pin fin density is at least double that of other commercial blocks, which enhances both the heat transfer coefficient and surface area, two key parameters here. Granted that the flow restriction is higher since smaller fins are used, our fin array has significantly shorter lengths than any other technologies in the market, hence the pressure drop is balanced. Our block focuses on the chip footprint. We would like to see a "basic pin waterblock" remove 500 W of heat from the Core i9 14900K while keeping the chip’s temperature below 95 °C. In fact, we have published research in scientific journals proving we can remove 1000 W of heat across 1 cm^2. To put this in perspective, the i9 14900k has a chip area of 2.57 cm^2, which means that we could theoretically remove 5 times more heat than what this chip actually produces in extreme conditions. Happy to share those articles if interested.

[@snu666]

Unbranded, £10 delivered on ebay from China. I got one last week. 😂

[@everope]

Just buy AMD so you won't need such exotic cooling lol

[@0Mindmachine0]

isn't this speedrunning degradation?

[@payback4803]

False temperature readings and failing power limits cause degredation. Effectively exceeding safe temperatures. Gamers nexus did plenty of videos about it.

[@emcoolinc]

Hey guys! Thank you so much for engaging with our content-we really appreciate it! When it comes to overclocking, higher voltages mean higher temperatures. However, if you can manage the heat effectively and keep temperatures low, you can actually extend the lifespan of your CPU. We've been pushing this CPU hard for nearly a year, and it shows no signs of degradation. Our Cinebench scores remain consistent and repeatable, with full system stability. This is thanks to our direct-die approach and efficient cooling system. As heat increases, so does electrical resistance, leading to what's known as parasitic currents. One of Intel's biggest missteps with this generation was releasing a power-hungry CPU with a heat spreader, combined with motherboards that don't limit voltage adequately. This leads to high temperatures and, ultimately, degradation. So, yes, for most users who aren’t using direct-die cooling (which is the vast majority), degradation can happen due to these thermal limitations-especially depending on overclocking levels, usage, and cooling setups. We understand that direct-die cooling is only for enthusiasts because of the effort involved and the fact that it voids your warranty. Ironically, though, it can actually save your CPU from Intel’s design flaws if you're looking to squeeze maximum performance out of your chip. Hope you found this helpful, and thanks again for your comments!

[@successshampoo]

​@@emcoolincnice

[@wilkuuu4016]

Can you provide any information about patent? ALphacool Core 1 LT had cross-section fins, the type of fins that you use is used by chineese companies from LGA 775 times to create the cheapest waterblocks possible, stop with that BS.

[@emcoolinc]

Hey, Sure, happy to provide some details on the granted patent - feel free to read more here. Patent name: System and method for providing direct silicon footprint microfluidic cooling for electronics. USPTO Patent number: US12080626B1. Here is the link: patents.google.com/patent/US12080626B1/en For avoidance of doubt, we’ve been working on this research for years during our time at Georgia Tech and feel extremely comfortable with the scientific backing of our product. If you have any technical questions, feel free to ask them here and we’ll respond. Regarding your other point citing “cross-section” fins, a pin fin array or geometry can not be patented as it is common sense for anyone skilled in the art. However, increasing the density of such pin fins is a challenge in order to provide more surface area for cooling when confined to the size of the chip. Our pin fin density is at least double that of the products you mentioned, which enhances both the heat transfer coefficient and surface area, two key parameters here. Granted that the flow restriction is higher since smaller fins are used, our fin array has significantly shorter lengths than any other technologies in the market, hence it is balanced. We would like to see any of those blocks you mentioned remove 500 W of heat from the Core i9 14900K while keeping the chip’s temperature below 95 °C. In fact, we have published research in scientific journals proving we can remove 1000 W of heat across 1 cm^2. To put this in perspective, the i9 14900k has a chip area of 2.57 cm^2, which means that we could theoretically remove 5 times more heat than what this chip actually produces in extreme conditions. Happy to share those articles if interested.

[@Peterlan15]

Emcool is sassyyy. I love It haha. Keep It Up with these great videos!!

[@GremaxLP]

@@emcoolinc hey, did you compare your block against other blocks? like Thermal Grizzly lga1700 pro variant. there is also much engineering in there and from the looks of it - it should perform better

[@emcoolinc]

@@Peterlan15 appreciate the support! 😉

[@emcoolinc]

@@GremaxLP Hey! Sure, we had some other people asking this, but here is our policy right now: This is something we are very open to, but in an effort to remain impartial we would prefer another independent channel to benchmark our product. Along those lines, we are actively looking for third-parties to do this and look forward to independent reviewers covering this on their channels. However, we will definitely still have videos showcasing extensive testing of our own EMCOOL systems. Thanks for watching!

[@SAVA47-00]

So you stole idea of the block from supercool and branded it to yours...that's terrible

[@Peterlan15]

Did you read when these guys at EMCOOL patented their block? Filed since 2019 and they have videos since 2020 doing direct die blocks, look them up, no way you can change vídeo dates on YT. When did supercool start doing these? The earliest I could find from supercool was 2022, so you tell me now who did it first?

[@SAVA47-00]

@@Peterlan15 supercool computer have been in business sin es 2015

[@Peterlan15]

​@@SAVA47-00sure, where is the YT video of their direct die blocks before 2019?. I'll wait. Have they filed any IP? Cause if that's the case with your baseless comment, then go ask chat gpt: who are the cofounders of EMCOOL INC? Then go Google Yogendra Joshi, who is the founder, and go SEE how this guy was doing watercooling before I was even born AND how many patents he has since the 80s. Stop with that BS if you're not willing to do some basic Google research about a company AND who the founders are before writing such non sense.

[@SAVA47-00]

Go look at supercool computer inventory on when they were doing direct die ...KZbin video mean nothing'

[@Peterlan15]

​@@SAVA47-00Your comment means nothing if you don't have the sources to BACK IT UP. The founder has patents, videos, articles since the 1980s for all sorts of thermal engineering products, so yeah he just felt like copying some dude. Go read before writing non sense here. AND yes, videos on YT count since those dates can NOT be changed, can't find anything from those other dudes you say before 2022. Based on your ignorant rationale, then IBM should own every other wc company since they have been doing direct die watercooling since the 1970s. Smh.

[@Dave20VT]

send review sampels to ppl like jay/GN/derb8uer

[@emcoolinc]

Hey! This is something we are trying to do, but the reality is we are a relatively new business and still working on building traction so we can get on their radars. Definitely looking forward to an independent reviewer covering this on their channel. Thanks for watching!