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 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.
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: th-cam.com/video/cQaqUyKVIEE/w-d-xo.html Again, appreciate the comment and we will keep this in mind moving forward.
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...
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!
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.
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.
False temperature readings and failing power limits cause degredation. Effectively exceeding safe temperatures. Gamers nexus did plenty of videos about it.
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!
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
liquid metal should NOT be used on a copper block. It will be absorbed into it and cause issues. Always use nickle plated copper.
@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.
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:
th-cam.com/video/cQaqUyKVIEE/w-d-xo.html
Again, appreciate the comment and we will keep this in mind moving forward.
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...
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!
Arrow Lake is new base for future performance improvements without heating your whole home at this same time.
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.
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.
isn't this speedrunning degradation?
False temperature readings and failing power limits cause degredation. Effectively exceeding safe temperatures. Gamers nexus did plenty of videos about it.
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!
@@emcoolincnice