Thanks Anastasi. Good innovation in chipboard engineering. *Current interests in my brood normalization with optimal AI only and no super intelligent AI mainframes of device to ppl controls due to ant to aphid asset control phenomena. Interested in re-spacecraft instrument control panel concepts. Daniel J Blatecky USA
@@dchdch8290ASM were the original founders of ASML (ASM Lithography), which was a joint venture between themselves and Phillips, before it was spun out as a separate company.
I worked in semiconductor plants all over the world starting in 1993 through 2012. I worked as a consultant in CIM automation. I saw a revolution starting with etching 2 and 3 inch wafers in beakers doing all this by hand through the widespread adoption of 300mm manufacturing. The pace of innovation during this time was wild. Since I got off the road I got all my technology news, albeit in sparse clumps, from the internet. Watching your channel takes me back to when I was in the industry and I love it!
@stephenmiller4948 which countries did you work in while working in the semiconductor industry? I'm in the UK & would love to work in a semiconductor factory over here or work in fabless design.
Sir, I am recent BTech graduate in ECE. Interested in working tsmc like semiconductor industry. Please guide or provide refferal for any of these companies, will be highly thankful sir. If you allow, I'll share my contact details sir.
And now you can buy a full PC in a small 6x6x2 inch box that has 4.2 billion transistors in its cpu alone, and puts out to four high definition displays and you could do you chip design CAD on it! Wow. I remember doing HAND drafted PCB layout and on AutoCad 2 on a 286 to print on mylar on a D size plotter for an in house camera shot at ten feet for the PCB house to skip a step. I never worked on chips but even PCB technology has come a long way from the days when VIA count mattered because they were not yet good at making those with high integrity. Nobody counts vias any more and they make huge layer counts and get soldered at much higher temperatures and are still high integrity. And the world advances so that someone anywhere else in the world can see this comment within seconds on a rectangular device they hold in their hands which ALSO has billions of transistors in it. I remember my 8 transistor AM radio that was smaller than a pack of cigarettes... now I feel old... again.
@@AndrewMellor-darkphoton ... They are half way to a thought. There isn't a bottleneck of Neon Gas for etching lazers if they aren't being used for this generation of computing. (Neon Gas is concentrated in processes related to the mining and steel industries.) Funny how the "advancement" of using both sides of a product matures after resources from the other side of the planet are no longer available. This lady makes the simplest principles in circuit design sound like miracles, and new thinking, when it is just more advertising.
I have BS in electrical engineering, and a I took a few classes in semiconductor physics from a great professor. But I hated those courses. You obviously loved it. Kudos to you!
Intel and AMD will definitely have their share of the market. TSMC is at max capacity and investing in other semiconductor companies will be an absolute power move, I keep increasing my shares manageably. Different chips are good at different things and Nvidia has been very specialised, which leaves other aspects of Al open.
certainly, i had bought NVDA shares at $300, $475 cheap b4 the 10 for 1 split and with huge interest I keep adding, i’m currently doings the same for PLTR and AMD constructively. Best possible way to get ahead, is participating behind top experienced performers.
You are buying a company to own it and not a piece of paper, The market is a zero-sum game (2 sides), Know what you are buying not just out of trend interest.
Good day. Fantastic presentation, thank you very much. I am a french electronic engineer and I got a master in electronics in 1969, so I appreciate your videos a lot. When I started to study electronics, it was with vacuum tubes and then later on it was with circuits containing a few transistors. It is for me very interesting to see today apparatuses containing hundred of billions of tiny transistors like e.g. in my mobile phone or in my laptop. You are a huge source of inspiration for younger generations. Thank you again.
My electrical engineering degree is the same vintage as yours. I'm wondering how they can learn so much in so little time. Back when I studied, most of what we learned was only theoretical. They couldn't actually implement most of the concepts they were teaching... for example digital signal processing. Our lab had one computer and it was a DIGITAL PDP-8. I didn't get my hands on an op amp until HP gave me a couple for my senior class project. We wasted lots of time learning stuff that was of no use. Maybe they have discarded that now to make room for what they must know. I programmed my first computer in 1964, an IBM 1620. It took me 3 weeks to multiple two numbers and output the results... in machine language. And it was a no-credit class. Calculators hadn't been invented yet. Our most valuable tool at my first job was the PACE analog computer. At that time I knew computers were no threat to my job security. I have been programming them ever since.
@@toddmarshall7573 Actually I think all the "useless" stuff imparted understanding which many young engineers do not understand,... I just finished a tube amp design and build,... fun-ish.
When this is said and done, AI is fully deployed I dont think we are going to look back and say, wow all our hard work really made the world better. Rather we are going to be faced with the evils that unbridled capitalism can unleash on our natural world. Life on earth is not getting better for most people, only for a few who are lucky enough to have been born in the right place at the right time.
That what you address is not capitalism, it's human. Look at the old communist's Industry plants ... How are they better than capitalist's Industry plants ?
@@ghostmourn Except that applies to all good things that have helped humanity everything comes with a cost. Captalism itself is a great example. So many pros, so many cons, but ultimately life is better with it than without it. The goal is to find an equilibrium to limit the cons and maximize its pros. It will be the same with AI but to a greater degree.
@@artysanmobile I can understand where you're getting at for focusing on the positives... respectfully, but I still think it is important to know where we can improve moving forward (by looking at what has been wasted in the past), perhaps in how we can continue to plan for efficiency and reduce waste, as I am sure in the future, we will have more cases of which we find ourselves wasting resources and/or space. It is truly a testament to today's innovations but a reminder of the complexity the universe holds. Also a great video overall, very informative!
This. The only company executing at the OS level on software is Apple. I don't know why anyone would buy anything else right now. Linux is a clusterF and Windows is laughable.
@@stoler7980 It's an endless parade of problems. No matter what bullshit the linux community tries to brainwash people into believing, linux is a laughing stock .
Researching and producing these video presentations must be time-consuming, but I watch and rewatch them and am amazed by the gigantic advances of modern-day technology. Again, thank you.
Great informative layout for this next step in semiconducotor world, I’m from the Netherlands and of course am familiar with ASM ASML and the simple fact that without ASML litho the tech would be not small and efficient like todays chips are. Real proud to know that the Dutch are great in more than water management, food production, flowers and trading…. This actually is ground breaking and unique tech used all over the world.
those machines look insane. thanks, i worked at a computer company in huntington beach for 4 years as their lead tech and supervisor. going from z80s to intel's 486, now we're lightyear's beyond all that, weird stuff.
@jensonee z80 recently got discontinued after all these years & some people designed an open source version. czcams.com/video/GI1e22A2J3U/video.htmlsi=kgAm2IwvKjxM6Fxy
Anastasi, thank you for another brilliant news program! Technology architectures at the bleeding edge are moving so fast that they're almost impossible to keep up with. New hardware implementations are racing ahead, and with all that horsepower, new software architectures are being explored. You're one of my knowledge concentrators. You pack a lot of information into a very short presentation which alerts me to breakthroughs, and then I can choose where I'm going to deep-dive for the knowledge that will flesh out the Big Picture.
I used to be a computer designer in the 80's when the latest and greatest technology was a 90 mb hard drive the size of a brick .. this is pretty amazing to me.
There is no one here... Inspite of 900 something comments. Just sheer monstrosity of ASML scanners is telling the story of decadence. Look for the specs on the latest chips and simple arithmetic will tell you... She is as terrible as the video, by the way.
I am not a scientist, an engineer, or a technician. I find your sunny enthusiasm so attractive I watch all of your videos, even though my understanding is.....limited. Thanks
I feel like Intel needs to be more risky in order to catch up. They've kind of been lagging behind TSMC when it comes to cutting-edge tech and innovations.
@edwardduda4222 Intel recently bought one of 2 of ASML's newest next generation semiconductor manufacturing machines called high NA EUV , TSMC hasn't bothered to buy one yet & ASML keeps the other for R&D.
lagging behind? I mean, I guess, they are also a customer of TSMC as they go mixed fab/fabless for creating processors. Only relying on your own is not always a good move.
@ 11:59 , this Intel dude is really selling it. He’s smooth and slick. History is against Intel, but favors TSMC. Intel is a like a show horse- looks great but cannot cross the finish line before TSMC. You know where to put down your money.
Congratulations for the ASM sponsorshiop; you truly deserve it for the quality of your knowledge and the educational value of how you share it. Also, I feel that all your viewers would be interested in a video about Hyper NA UV machines...
Yours is one of the few videos that I need to slow to normal speed with occasional rewinds so I can keep up with the great content you provide. May more tech firms find your channel and sponsor your content.
I'm a software engineer with a keen interest in hardware, and I've found your channel to be the best resource for learning about hardware technology. I'm particularly curious about a few topics and would love to see videos on them: Tesla vs. NVIDIA in Full Self-Driving (FSD) Technology: Could you provide a detailed comparison of their current computational capabilities and progress in the field of autonomous driving? Intel vs. AMD vs. NVIDIA: It would be great to see an in-depth analysis of how these companies stack up against each other in terms of CPU and GPU performance, particularly in the context of high-performance computing and gaming. Cloud Vendors Comparison (Microsoft Azure, AWS, Google Cloud): I'm interested in understanding how these cloud providers compare in terms of hardware infrastructure, GPU/CPU offerings, and overall performance for various computing tasks. Thanks in advance for considering these topics!
Imagine if we finally moved on from our current silicon based microchips and started using graphene based transistors in our electronics i know its been in the realm of theory for better part of 2 Decades but it seems MIT and other institutions are paving the way in breakthroughs that yeild actually results but we all know when something is proven to work and its repeatable its up to the industry heavy hitters to implement these new technologies. Thank you for being a amazing channel to learn drom and be amazed at the knowledge of how our tech work's and seeing the advancements coming.
@@thesnare100 the promises of graphene haven't been utilized in anything substantial yet but the conductivity is off the charts the speed of this was measured in Terahertz
@@shadowarez1337 what was measured in Terahertz, the clock speed of a CPU? Electrical conductivity is not measured by that, or at least electrical resistance is measured in Ohm's.
I feel compelled to compliment and thank you. Thank you for being perfectly balanced between knowledge, enthusiasm, femininity, and classy beauty. Keep up the consistantly good work, and you'll be very successful.
I really enjoy your content. Thank you for taking the time and putting in so much effort to create these videos. Your style and delivery are very engaging and the descriptions are easy for an average person like me to follow. Thanks again and I look forward to more of your content.
Dear A ; The fact that you can narrate so well ads to the production value of visuals, with cuts to you having a nice rhythm and cadence . Are you actually an AI driven character? With your easy on the eyes, pleasing to the ears manor, it’s hard not to have a crush on the beauty while admiring the beast which is your intellect. Intel is rolling the dice on dual architecture change added to new production equipment…. This test their coordination of talent/skill pool. It’s a Lamborguini story in the making. Keep up the great coverage!
Ana, I love the fluency with which you present these videos. The technology is advancing so fast it is very difficult to know what is happening let alone trying to keep up to date. You do a very good job of presenting this for us laymen who are out here in the woodwork. It is so nice to see such a beautiful woman with such a beautiful brain to go with it.
@@rogerphelps9939 So what isn't a waste? Making AI chips that optimize targeted advertisement to manipulate masses into buying more garbage? Or making chips for guided missiles to be thrown at children in ga...a by genocidal regiemes? You see, all technology is subservient to some other goals. I would rather have it to power video games, which by the way were the reason for development of GPUs on which first AIs did run, than the former.
I suspect that Intel is willing to buy the overly expensive fab machine and operate at a loss for the reputation and brand boost they would get for being the first out the door, beating TSMC. It would earn them clients that would otherwise have gone to TSMC and show them as the “king” of fabs. Especially with the government subsidies Intel has from the US, they can afford this. Then, they can invest in affordability down the line to return to profitability after they have acquired all the new business. In other words, I don’t think they even need Direct Self-Assembly to work to execute their overarching strategy.
Well Intel was the undisputed semiconductor manufacturing process leader for a few decades until 2018 when they started to have issues pushing 10nm due to not using EUV lithography machines and issues with new interconnect materials, and I believe it's quite normal for them to want to regain their previous status. Intel still has brilliant engineers and TSMC is already on high alert about Intel's new process developments, and if Intel managed to bring Intel 20A products on time they will gain a lot of interests to various design firms.
You are absolutely right about the issues connected with making two changes simultaneously. If a doctor prescribes two medication and the patient suffers a side effect, both medications need to be stopped that sends you right back to the beginning.
Ana, love your channel, your videos are always fascinating to these computer geek... As to if your video is ever too long? No, never. Thank you for your channel
I designed and managed the design of microprocessors at Intel through 2000 after which I retired. I've followed the technology roadmap in other roles and was extremely disappointed to see Intel blow a 2 year lead in silicon technology to now lagging badly. Bad management, loss of focus, and failure to maintain investment has led to this. I still have hopes for Pat to turn the ship around and think it's entirely reasonable for Intel to largely catch up since they don't have to codevelop equipment after TSMC has done it. The intel transition to a foundry is extremely risky, but still not as risky as their failure to keep chip architecture and design on the leading edge. Even if Intel had the equivalent to TSMC now they'd still have missed several market shifts, the latest being AI and neural network on board. It was good to see some research into quantum computing, but that's still a ways off and they desperately need the right products for their hoped for resurgence in silicon. Great job explaining all this Anastasi!
Breaking the 2nm barrier, the looming holy grail mirage of the impossible . . . until now. As always, love your presentations. And of course, congrats on having ASM as a sponsor! That's a big deal, coming from the extremes of the nanoworld.
2 nm is a commercial name. It's not a CD (critical dimension) of 2 nm. Even less a gate of 2 nm. The best CD reached are close to 8 nm. And It's already insane.
ASM is not ASML! I did a deep dive into this to understand. ASM is ASM International and they make Atomic Layer Deposition machines (for the wafer substrate). They spun off ASML which makes the Photolithography and Etching Machine. ASML is the huge company everyone always talks about. ASM is also a large company but much smaller than ASML (10 times fewer employees and revenue). Not to throw any shade on Anastasi who is great and since ASM (not ASML) is sponsoring the video, of course it probably wouldn't be right for her to point this out.
@SurfCatten why spin them off when they're both making products designed for a similar end goal of being machines to help manufacture semiconductors? 🤷🏾♂️
From a business perspective, I'm glad Andy Grove's "Only The Paranoid Survive" philosophy still has traction at Intel. This new technology must represent what Grove described as a Strategic Inflection Point. Their RibbonFET variant of the innovation represents that inflection point. Please give Intel's engineers and designers a chance to rise to the challenge! Intel needs this innovation in order to give more time for Intel Labs to develop the scalable silcon-based quantum proccesor, the next inflection point. Quantum computing has the potential to completely obsolete Moore's Law.
Intel can take risks in their internal products and use TSMC as a n-1 back up in case they struggle. My concern is that Intel Foundry customers won’t want to take the same risks until Intel proves out the process. I think Intel may struggle as much with ramping multiple technologies, as with managing multiple chip types from many customers.
Good to mention is that the backside power delivery will make the power supply a lot more stable. This is because power lines and signal lines interfere. More stable power supply will allow for higher clocks.
Samsung has been shipping its 3 nm gate all around (GAA) process, called "3GAA", since mid-2022. Samsung announced in June 2022 that it had entered mass production of its 3 nm GAA manufacturing process. However, yield improvements have been slow, and only a few chip design companies have used Samsung's 3 nm GAA process to produce simple chips like mining chips.
I kind of love that it seems we're all but bending the laws of physics to keep on adhering to Moore's Law. And yes I would very much love a video on Hyper NA Lithography.
I am layman when it comes to transistor technology, but you made this video interesting and engaging even for someone like me who doesn't know the vocabulary of chipmaking.
The fact that the human hair is approx 1,000,000 atoms across is insane, building these chips with material layers being only---- 1 atom thick ---- is mind bogglingly jaw dropping!
Thank goodness for a real human voice and not some synthesized voice over. I love your accent and your charisma. I find it so amazing that we can make parts on a chip literally atom for atom.
i love your videos. i got out of computer technology, as an engineer tech, in 1986/7. i went into networking. it's nice to watch it all keep changing and becoming more and more integrated into everything, scary.
I follow you closely. I'm not dumb. Almost every story I read, not just yours, is some new development that is the next big thing, a game changer, etc. can you start working on a story, ranking all these breakthroughs by importance, and then by development, for the past 6 months?
This is a big one. This is top 2-3, if not 1st. We're looking at at least a 100% improvement in processing power, but could be up to a 1000% improvement in processing power. Firstly, double-siding the wafer is going to free up huge amounts of physical space for more transistors - not to mention give further ability in higher clock speeds (by distancing the T's from the power lines - and better efficiency and power usage, equating to less heat and the ability to drive things a bit harder and consume less power. Another thought, for me is, barring physical design restraints, there's nothing stopping anyone from putting transistors on both sides of the wafer.
I must admit I do not know much about chip design but the way things are explained in this channel is really peaking my interest in deepening my knowledge. Thank you and great work
I’m not sure if I can listen to anyone else explain this to me. Either that or I fell in love. No, it’s probably because I can truly feel the passion you have in this topic. Thanks for providing information in an understandable way!
In 1977, I was QC testing high-purity silicon tetrachloride using an Applied Materials mini-epitaxy furnace and probing the resulting growth layer on the test wafer for resistivity. We were buying impure material from Texas Instruments (a by-product of their wafer manufacturing) for 50-cents a pound, distilling it to remove N-type and P-type impurities, then selling it for anywhere from $50 to $200 a pound, depending on the quality. Seeing so much economic value being generated, I jumped out of the lab and bought a retail ComputerLand franchise and sold a lot of Apple IIs. Nice to see so much has changed so quickly.
Thanks for highlighting these forward thinking and engineering techniques. Unless a person is part of the fabrication environment you don't hear about these things. Most fascinating. Your doing a great 😂.
Having taken an EE, hydraulics/pneumatics and chip fabrication course for 2 years (Sponsored by Infineon) - then becoming a union electrician that helped build the 200 and 300mm fabs in Richmond, VA - You did a fantastic job at explaining this. Good job! Intel might be a stock buy - nobody's talking about this yet.
Hi Anastasia. My name is Roberto and I am from Italy. I'm not a specialist in the topics you cover but, for some years now, i read about spintronics, skirmions and antiferroelectric materials as the electronics of the future. Could you explain us these technologies and why they are non yet been implemented? Thank you very much for your work of popularisation❤
brilliant video! I am a bit frustrated that I found your channel just a couple of days ago. I remember my classmates in the ´70s coming along with punched cards explaining a lot of nonsens what the machines in the IBM lab doing with that. The milestones in the processor technology during the last 50 years is not comparable to any other industry. IBM in the early days, ASML, TSMC, AMD and INTEL have changed life for the entire humankind (good or bad...I leave it open 🙂).
i thought 14th gen intel used power via which is a form of back side power delivery. maybe i heard wrong or is this different from what tsmc is doing? but yeah would love to hear about ultra na
11:00 I agree. This is super risky and everyone holding Intel stock should be careful. Intel is leveraging its chip production bet yet another time. I'm not sure this bet will work.
Intel has had all kinds of problems with trying to introduce too much at the same time. That's part of the reason they had so many problems with their 10nm system, if I remember correctly. Great video, I think I've found a new tech fiber to follow.
To do a back side power on a wafer would require pure silicon and microvias in the wafer through the die. Won't be easy to do this, but you could put the bigger stuff on the bottom and the computing stuff on the top.
I guess I saw this video a little late, but in a timeless industry such as chip manufacturing, where time never stands still, this video will be usable for a couple of years I would guess. This topic really allows my inner nerd to fully geek out, and also understand how far in the future these chipmakers really are. Considering that one of those ASML-machines takes about 2 years to build, the ordering process of them must start well ahead of time. It just makes my brain boil when I try to understand the whole chain. Thank you very much for all of these wonderful videos.
To follow up on the economics of High-NA - the reason the wafers-per-hour throughout is so slow is because the field size that is exposed on the wafer is half the size of standard EUV (429 mm² for high-NA vs 858mm²), thereby doubling the amount of exposures required per wafer and more or less doubling the process time per wafer (at least for the layers that require Hugh-NA EUV, which is only a small number compared to the total number of layers on the chips). However, the smaller field size also has another major impact that directly affects their customers and chip designers. The field size determines the largest possible die size. And in recent years, as transistor counts have exploded with AI chips and all that, and with transistor shrink rates slowing, die sizes have been growing, with many commercial chips in the 500-600 mm² range. Monolithic (single die) chips of that size will be impossible to make on high NA, which may severely hamstring chip designers. So with the crazy upfront costs, high manufacturing cost due to the lower throughput, maz die size that is only half of previous generations, I imagine high NA is going to be a very hard sell for a lot of customers for a long time.
This power-under-logic system is really interesting. One wonders what unexpected gremlins will be revealed in thermal performance of these designs. A fairly high amount of the waste heat per clock cycle is in power delivery itself, simply from resistance, so if you stick the power delivery on the other side of a silicon wafer- a lousy heat conductor- then what happens? Then there's the problem of the pinouts, needing to coat both sides in protective layers, etc., etc. I think this is going to prove very difficult!
The technology is mind blowing. I respect those in the industry that keep on pushing the tech forward. I wonder what is the limit? What is the actual vision all these scientists have in their mind of what is possible next 100 years. Born in 1991, I saw a rapid change in economy, tech,lifestyle. Maybe no need for 100 years to see massive changes this time around.
This sort of thing was pioneered for Solar Cells maybe 20 years ago, by a company called Sunpower. Their motivation for back-side power connectors was to maximize the amount of silicon exposed to sunlight. It is interesting that this tech also helps with the new generation of VLSI chips.
thank you, this was an awesome video and it was NOT to long. the knowledge and breaking down the structures of the nextgen chips is helping my information gathering as we go into the future at blazing rates of time and speed. YES, please complete the CFET video as I will be wait to consume the information. understanding the chips and new memory and coding is imperative for any of use in Tech or stem. there is not enough hours in a day to stay up on the massive amounts of new data on new methods and tech. again thank you for you work and channel
I really hope Intel's gamble pays off here, they have really been struggling with fab issues over the past couple of nodes and it is starting to impact their competitiveness which isn't a good thing in an industry where there are now so few players. Looking at More's law from an economic perspective is actually something I have thought of but it actually makes more sense than from a feature size perspective. Gate all round is going to be a game changer especially for mobile devices because of the power reduction achieved, IC design really is entering another era of rapid progress.
Learn more about ASM and their products: lnk.bio/ASMOfficial
Thank you, never heard of ASM, but now it is clear that their tech is fundamental. How about beyond silicon tech ?
As a supporter, I enjoy your video every week. What impresses me most about this video is ASM sponsorship!!! The future is bright
Thanks Anastasi. Good innovation in chipboard engineering. *Current interests in my brood normalization with optimal AI only and no super intelligent AI mainframes of device to ppl controls due to ant to aphid asset control phenomena. Interested in re-spacecraft instrument control panel concepts. Daniel J Blatecky USA
I though L missed from ASM
@@dchdch8290ASM were the original founders of ASML (ASM Lithography), which was a joint venture between themselves and Phillips, before it was spun out as a separate company.
I worked in semiconductor plants all over the world starting in 1993 through 2012. I worked as a consultant in CIM automation. I saw a revolution starting with etching 2 and 3 inch wafers in beakers doing all this by hand through the widespread adoption of 300mm manufacturing. The pace of innovation during this time was wild. Since I got off the road I got all my technology news, albeit in sparse clumps, from the internet. Watching your channel takes me back to when I was in the industry and I love it!
@stephenmiller4948 which countries did you work in while working in the semiconductor industry? I'm in the UK & would love to work in a semiconductor factory over here or work in fabless design.
what was your major in college?
They were great times !! I have a similar story to yours I worked for VG Scientific.
Sir, I am recent BTech graduate in ECE.
Interested in working tsmc like semiconductor industry.
Please guide or provide refferal for any of these companies, will be highly thankful sir.
If you allow, I'll share my contact details sir.
And now you can buy a full PC in a small 6x6x2 inch box that has 4.2 billion transistors in its cpu alone, and puts out to four high definition displays and you could do you chip design CAD on it! Wow. I remember doing HAND drafted PCB layout and on AutoCad 2 on a 286 to print on mylar on a D size plotter for an in house camera shot at ten feet for the PCB house to skip a step. I never worked on chips but even PCB technology has come a long way from the days when VIA count mattered because they were not yet good at making those with high integrity. Nobody counts vias any more and they make huge layer counts and get soldered at much higher temperatures and are still high integrity. And the world advances so that someone anywhere else in the world can see this comment within seconds on a rectangular device they hold in their hands which ALSO has billions of transistors in it. I remember my 8 transistor AM radio that was smaller than a pack of cigarettes... now I feel old... again.
Ana, it is truly a great honor and recognition of your great achievement to have your video sponsored by ASM! Congratulations!
Has the Chinese stolen this tech?
@@popcorn6931 Has the West been murderers and bandits for centuries?
@@popcorn6931 How in the hell is having things to do with the comment
Her videos are quite valuable now
@@AndrewMellor-darkphoton ... They are half way to a thought.
There isn't a bottleneck of Neon Gas for etching lazers if they aren't being used for this generation of computing.
(Neon Gas is concentrated in processes related to the mining and steel industries.)
Funny how the "advancement" of using both sides of a product matures after resources from the other side of the planet are no longer available.
This lady makes the simplest principles in circuit design sound like miracles, and new thinking, when it is just more advertising.
I have BS in electrical engineering, and a I took a few classes in semiconductor physics from a great professor. But I hated those courses. You obviously loved it. Kudos to you!
Prof chenming hu
I read BS as BullShit💀
Solid state physics is hard. Pun? I guess I'll take it.
@@tusff07Youre here now learning
Hate/Love? You seem conflicted.
ps GED classes are available for free at most community colleges
Intel and AMD will definitely have their share of the market. TSMC is at max capacity and investing in other semiconductor companies will be an absolute power move, I keep increasing my shares manageably. Different chips are good at different things and Nvidia has been very specialised, which leaves other aspects of Al open.
This is the type of in-depth detail on the semiconductor market that investors need, also the right moment to focus on the rewarding AI manifesto.
certainly, i had bought NVDA shares at $300, $475 cheap b4 the 10 for 1 split and with huge interest I keep adding, i’m currently doings the same for PLTR and AMD constructively. Best possible way to get ahead, is participating behind top experienced performers.
How much of their stock do you own? Seems like a lot of your investment is riding on this
You are buying a company to own it and not a piece of paper, The market is a zero-sum game (2 sides), Know what you are buying not just out of trend interest.
Amazingly, people are starting to get the uniqueness of Palantir.
Good day. Fantastic presentation, thank you very much. I am a french electronic engineer and I got a master in electronics in 1969, so I appreciate your videos a lot. When I started to study electronics, it was with vacuum tubes and then later on it was with circuits containing a few transistors. It is for me very interesting to see today apparatuses containing hundred of billions of tiny transistors like e.g. in my mobile phone or in my laptop. You are a huge source of inspiration for younger generations. Thank you again.
Amen brother.
Vacuum tube technology, as well as analog devices, are still potentially worth their weight in gold.
My electrical engineering degree is the same vintage as yours. I'm wondering how they can learn so much in so little time. Back when I studied, most of what we learned was only theoretical. They couldn't actually implement most of the concepts they were teaching... for example digital signal processing. Our lab had one computer and it was a DIGITAL PDP-8. I didn't get my hands on an op amp until HP gave me a couple for my senior class project. We wasted lots of time learning stuff that was of no use. Maybe they have discarded that now to make room for what they must know. I programmed my first computer in 1964, an IBM 1620. It took me 3 weeks to multiple two numbers and output the results... in machine language. And it was a no-credit class. Calculators hadn't been invented yet. Our most valuable tool at my first job was the PACE analog computer. At that time I knew computers were no threat to my job security. I have been programming them ever since.
@@toddmarshall7573 Actually I think all the "useless" stuff imparted understanding which many young engineers do not understand,... I just finished a tube amp design and build,... fun-ish.
What did they call that back in the day? A Masters is a pretty new concept right? Just graduates? :D
I am Taiwanese 🇹🇼🇹🇼🇹🇼
and I am very happy that TSMC has contributed to improving the quality of human life.😊
When this is said and done, AI is fully deployed I dont think we are going to look back and say, wow all our hard work really made the world better. Rather we are going to be faced with the evils that unbridled capitalism can unleash on our natural world. Life on earth is not getting better for most people, only for a few who are lucky enough to have been born in the right place at the right time.
@@ghostmournmaybe your could be right bro
That what you address is not capitalism, it's human.
Look at the old communist's Industry plants ... How are they better than capitalist's Industry plants ?
Get ready to become Chinese.
@@ghostmourn Except that applies to all good things that have helped humanity everything comes with a cost. Captalism itself is a great example. So many pros, so many cons, but ultimately life is better with it than without it. The goal is to find an equilibrium to limit the cons and maximize its pros. It will be the same with AI but to a greater degree.
What's astonishing is how clunky software has wasted all this enormous computing power.
@@artysanmobile I can understand where you're getting at for focusing on the positives... respectfully, but I still think it is important to know where we can improve moving forward (by looking at what has been wasted in the past), perhaps in how we can continue to plan for efficiency and reduce waste, as I am sure in the future, we will have more cases of which we find ourselves wasting resources and/or space. It is truly a testament to today's innovations but a reminder of the complexity the universe holds. Also a great video overall, very informative!
This. The only company executing at the OS level on software is Apple. I don't know why anyone would buy anything else right now. Linux is a clusterF and Windows is laughable.
@@EverSpaceTimewhat's your issue with linux?
@@stoler7980 It's an endless parade of problems. No matter what bullshit the linux community tries to brainwash people into believing, linux is a laughing stock .
@stoler7980 Perhaps his "Tuxophobia" is caused by a lack of walls around the garden?
1st awesome sponsor I see on CZcams since... ever! Go ASM!!
Edit: Also love the video and your excitement as always! Thank you very much!!
Thank you!
I think you are a brilliant educator and scientist. Thank you for your work.
Researching and producing these video presentations must be time-consuming, but I watch and rewatch them and am amazed by the gigantic advances of modern-day technology. Again, thank you.
Great informative layout for this next step in semiconducotor world, I’m from the Netherlands and of course am familiar with ASM ASML and the simple fact that without ASML litho the tech would be not small and efficient like todays chips are. Real proud to know that the Dutch are great in more than water management, food production, flowers and trading…. This actually is ground breaking and unique tech used all over the world.
those machines look insane. thanks, i worked at a computer company in huntington beach for 4 years as their lead tech and supervisor. going from z80s to intel's 486, now we're lightyear's beyond all that, weird stuff.
@jensonee z80 recently got discontinued after all these years & some people designed an open source version.
czcams.com/video/GI1e22A2J3U/video.htmlsi=kgAm2IwvKjxM6Fxy
Anastasi, thank you for another brilliant news program!
Technology architectures at the bleeding edge are moving so fast that they're almost impossible to keep up with. New hardware implementations are racing ahead, and with all that horsepower, new software architectures are being explored.
You're one of my knowledge concentrators. You pack a lot of information into a very short presentation which alerts me to breakthroughs, and then I can choose where I'm going to deep-dive for the knowledge that will flesh out the Big Picture.
I used to be a computer designer in the 80's when the latest and greatest technology was a 90 mb hard drive the size of a brick .. this is pretty amazing to me.
There is no one here... Inspite of 900 something comments.
Just sheer monstrosity of ASML scanners is telling the story of decadence. Look for the specs on the latest chips and simple arithmetic will tell you... She is as terrible as the video, by the way.
YES! In fact at Miniscribe the drives in inventory really were bricks wrapped to look like finished drives. Thanks for reminding me of that heady era.
I am not a scientist, an engineer, or a technician. I find your sunny enthusiasm so attractive I watch all of your videos, even though my understanding is.....limited. Thanks
I feel like Intel needs to be more risky in order to catch up. They've kind of been lagging behind TSMC when it comes to cutting-edge tech and innovations.
@edwardduda4222 Intel recently bought one of 2 of ASML's newest next generation semiconductor manufacturing machines called high NA EUV , TSMC hasn't bothered to buy one yet & ASML keeps the other for R&D.
Intel and TSMC are just the end-point integrators of an immense component-technology ecology.
@obimk1-ms1jw Since when does Apple run a fab?
@obimk1-ms1jw: HOW DID YOU HAVE ACTIVE LINKS IN YOUR COMMENTS!??? 🙄🙄🤔
lagging behind? I mean, I guess, they are also a customer of TSMC as they go mixed fab/fabless for creating processors. Only relying on your own is not always a good move.
I am retired now, but I love to keep up with the latest and technology particular with computers and chip manufacturing. Thank you for sharing.
@ 11:59 , this Intel dude is really selling it. He’s smooth and slick. History is against Intel, but favors TSMC. Intel is a like a show horse- looks great but cannot cross the finish line before TSMC. You know where to put down your money.
Congratulations for the ASM sponsorshiop; you truly deserve it for the quality of your knowledge and the educational value of how you share it.
Also, I feel that all your viewers would be interested in a video about Hyper NA UV machines...
Yours is one of the few videos that I need to slow to normal speed with occasional rewinds so I can keep up with the great content you provide. May more tech firms find your channel and sponsor your content.
Thank you for this update! Great video !
Looking forward how a14 will impact intel’s stock
I'm a software engineer with a keen interest in hardware, and I've found your channel to be the best resource for learning about hardware technology. I'm particularly curious about a few topics and would love to see videos on them:
Tesla vs. NVIDIA in Full Self-Driving (FSD) Technology: Could you provide a detailed comparison of their current computational capabilities and progress in the field of autonomous driving?
Intel vs. AMD vs. NVIDIA: It would be great to see an in-depth analysis of how these companies stack up against each other in terms of CPU and GPU performance, particularly in the context of high-performance computing and gaming.
Cloud Vendors Comparison (Microsoft Azure, AWS, Google Cloud): I'm interested in understanding how these cloud providers compare in terms of hardware infrastructure, GPU/CPU offerings, and overall performance for various computing tasks.
Thanks in advance for considering these topics!
Wow, I learn more from you than I do from the quarterly meetings with the Factory head in Aloha.
Imagine if we finally moved on from our current silicon based microchips and started using graphene based transistors in our electronics i know its been in the realm of theory for better part of 2 Decades but it seems MIT and other institutions are paving the way in breakthroughs that yeild actually results but we all know when something is proven to work and its repeatable its up to the industry heavy hitters to implement these new technologies.
Thank you for being a amazing channel to learn drom and be amazed at the knowledge of how our tech work's and seeing the advancements coming.
Thank you 😊
I'd like know how graphene transistors would compare to optical ones, I think it will be another "format war"
@@thesnare100 the promises of graphene haven't been utilized in anything substantial yet but the conductivity is off the charts the speed of this was measured in Terahertz
@@shadowarez1337 what was measured in Terahertz, the clock speed of a CPU? Electrical conductivity is not measured by that, or at least electrical resistance is measured in Ohm's.
I feel compelled to compliment and thank you. Thank you for being perfectly balanced between knowledge, enthusiasm, femininity, and classy beauty. Keep up the consistantly good work, and you'll be very successful.
This wafer looked like a plastic thing but when you lifted it up in a flat motion it sounded like glass metal blade 😂🤯
I really enjoy your content. Thank you for taking the time and putting in so much effort to create these videos. Your style and delivery are very engaging and the descriptions are easy for an average person like me to follow. Thanks again and I look forward to more of your content.
Dear A ;
The fact that you can narrate so well ads to the production value of visuals, with cuts to you having a nice rhythm and cadence . Are you actually an AI driven character?
With your easy on the eyes, pleasing to the ears manor, it’s hard not to have a crush on the beauty while admiring the beast which is your intellect.
Intel is rolling the dice on dual architecture change added to new production equipment…. This test their coordination of talent/skill pool. It’s a Lamborguini story in the making.
Keep up the great coverage!
Thank you Anastasi- a video on the Hyper NA EUV machine would be awesome!
Ana, I love the fluency with which you present these videos. The technology is advancing so fast it is very difficult to know what is happening let alone trying to keep up to date. You do a very good job of presenting this for us laymen who are out here in the woodwork. It is so nice to see such a beautiful woman with such a beautiful brain to go with it.
Man gaming pc's will be insane in the 2030's...my Rtx 3060 does not even feel dated to me yet.
3060 is epic.
Of course I remember when floating point co-processors were cool.
It's got crazy good price to performance.
Gaming is a complete waste of the ingenuity and perseverence tthat has gone into these chips.
@@rogerphelps9939 So what isn't a waste? Making AI chips that optimize targeted advertisement to manipulate masses into buying more garbage? Or making chips for guided missiles to be thrown at children in ga...a by genocidal regiemes? You see, all technology is subservient to some other goals. I would rather have it to power video games, which by the way were the reason for development of GPUs on which first AIs did run, than the former.
@@rogerphelps9939 Speak for yourself, it's not a waste if people have fun with it.
@@rogerphelps9939 go tell that to the companies innovating and making them. don't be a joker
Here's hoping ASM gives you a tour soon!
I thought it was ASML. Where did the L go? XD
Great to see you getting those important companies as sponsors. Much more impressive than some cash cow start ups with no value to society.
I suspect that Intel is willing to buy the overly expensive fab machine and operate at a loss for the reputation and brand boost they would get for being the first out the door, beating TSMC. It would earn them clients that would otherwise have gone to TSMC and show them as the “king” of fabs. Especially with the government subsidies Intel has from the US, they can afford this. Then, they can invest in affordability down the line to return to profitability after they have acquired all the new business. In other words, I don’t think they even need Direct Self-Assembly to work to execute their overarching strategy.
Too naive, Intel is losing more money on the fab business than the subsidies from the US government.
Well Intel was the undisputed semiconductor manufacturing process leader for a few decades until 2018 when they started to have issues pushing 10nm due to not using EUV lithography machines and issues with new interconnect materials, and I believe it's quite normal for them to want to regain their previous status. Intel still has brilliant engineers and TSMC is already on high alert about Intel's new process developments, and if Intel managed to bring Intel 20A products on time they will gain a lot of interests to various design firms.
I was waiting for you to discuss this technology, thanks!
You are absolutely right about the issues connected with making two changes simultaneously. If a doctor prescribes two medication and the patient suffers a side effect, both medications need to be stopped that sends you right back to the beginning.
Ana, love your channel, your videos are always fascinating to these computer geek...
As to if your video is ever too long? No, never.
Thank you for your channel
I designed and managed the design of microprocessors at Intel through 2000 after which I retired. I've followed the technology roadmap in other roles and was extremely disappointed to see Intel blow a 2 year lead in silicon technology to now lagging badly. Bad management, loss of focus, and failure to maintain investment has led to this. I still have hopes for Pat to turn the ship around and think it's entirely reasonable for Intel to largely catch up since they don't have to codevelop equipment after TSMC has done it. The intel transition to a foundry is extremely risky, but still not as risky as their failure to keep chip architecture and design on the leading edge. Even if Intel had the equivalent to TSMC now they'd still have missed several market shifts, the latest being AI and neural network on board. It was good to see some research into quantum computing, but that's still a ways off and they desperately need the right products for their hoped for resurgence in silicon. Great job explaining all this Anastasi!
Breaking the 2nm barrier, the looming holy grail mirage of the impossible . . . until now. As always, love your presentations.
And of course, congrats on having ASM as a sponsor! That's a big deal, coming from the extremes of the nanoworld.
2 nm is a commercial name.
It's not a CD (critical dimension) of 2 nm. Even less a gate of 2 nm.
The best CD reached are close to 8 nm. And It's already insane.
Hyper NA video please
ASM is not ASML! I did a deep dive into this to understand. ASM is ASM International and they make Atomic Layer Deposition machines (for the wafer substrate). They spun off ASML which makes the Photolithography and Etching Machine. ASML is the huge company everyone always talks about. ASM is also a large company but much smaller than ASML (10 times fewer employees and revenue). Not to throw any shade on Anastasi who is great and since ASM (not ASML) is sponsoring the video, of course it probably wouldn't be right for her to point this out.
@SurfCatten why spin them off when they're both making products designed for a similar end goal of being machines to help manufacture semiconductors? 🤷🏾♂️
From a business perspective, I'm glad Andy Grove's "Only The Paranoid Survive" philosophy still has traction at Intel. This new technology must represent what Grove described as a Strategic Inflection Point. Their RibbonFET variant of the innovation represents that inflection point. Please give Intel's engineers and designers a chance to rise to the challenge!
Intel needs this innovation in order to give more time for Intel Labs to develop the scalable silcon-based quantum proccesor, the next inflection point. Quantum computing has the potential to completely obsolete Moore's Law.
Wow! You got sponsored by ASM nice. Its not like they need any promotion to get known but its amazing they did
Intel can take risks in their internal products and use TSMC as a n-1 back up in case they struggle. My concern is that Intel Foundry customers won’t want to take the same risks until Intel proves out the process. I think Intel may struggle as much with ramping multiple technologies, as with managing multiple chip types from many customers.
Yes, Intel dropped the ball the last 20 years.
Thank you for sharing your knowledge. Your videos are most excellent.
Good to mention is that the backside power delivery will make the power supply a lot more stable. This is because power lines and signal lines interfere. More stable power supply will allow for higher clocks.
Global Foundries was holding AMD back but became a powerhouse when they moved production to TSMC
Samsung has been shipping its 3 nm gate all around (GAA) process, called "3GAA", since mid-2022. Samsung announced in June 2022 that it had entered mass production of its 3 nm GAA manufacturing process. However, yield improvements have been slow, and only a few chip design companies have used Samsung's 3 nm GAA process to produce simple chips like mining chips.
I kind of love that it seems we're all but bending the laws of physics to keep on adhering to Moore's Law.
And yes I would very much love a video on Hyper NA Lithography.
I am layman when it comes to transistor technology, but you made this video interesting and engaging even for someone like me who doesn't know the vocabulary of chipmaking.
I watched the video three times and each time was more enjoyable
The fact that the human hair is approx 1,000,000 atoms across is insane, building these chips with material layers being only---- 1 atom thick ---- is mind bogglingly jaw dropping!
Another great update, Anastasi! Well done! I've shared this on my channel.
Thank goodness for a real human voice and not some synthesized voice over. I love your accent and your charisma. I find it so amazing that we can make parts on a chip literally atom for atom.
Always love ur updates... I think in CZcams has similar video from Intel called RibbonFet and powerVia technology
i love your videos. i got out of computer technology, as an engineer tech, in 1986/7. i went into networking. it's nice to watch it all keep changing and becoming more and more integrated into everything, scary.
I follow you closely. I'm not dumb. Almost every story I read, not just yours, is some new development that is the next big thing, a game changer, etc. can you start working on a story, ranking all these breakthroughs by importance, and then by development, for the past 6 months?
This is a big one. This is top 2-3, if not 1st. We're looking at at least a 100% improvement in processing power, but could be up to a 1000% improvement in processing power.
Firstly, double-siding the wafer is going to free up huge amounts of physical space for more transistors - not to mention give further ability in higher clock speeds (by distancing the T's from the power lines - and better efficiency and power usage, equating to less heat and the ability to drive things a bit harder and consume less power.
Another thought, for me is, barring physical design restraints, there's nothing stopping anyone from putting transistors on both sides of the wafer.
I must admit I do not know much about chip design but the way things are explained in this channel is really peaking my interest in deepening my knowledge. Thank you and great work
Thank you for this super interesting video !! continue post please !!
I’m not sure if I can listen to anyone else explain this to me. Either that or I fell in love. No, it’s probably because I can truly feel the passion you have in this topic. Thanks for providing information in an understandable way!
Thanks for the informative video ... Yes to the hyper NA Euv Machines.
In 1977, I was QC testing high-purity silicon tetrachloride using an Applied Materials mini-epitaxy furnace and probing the resulting growth layer on the test wafer for resistivity. We were buying impure material from Texas Instruments (a by-product of their wafer manufacturing) for 50-cents a pound, distilling it to remove N-type and P-type impurities, then selling it for anywhere from $50 to $200 a pound, depending on the quality.
Seeing so much economic value being generated, I jumped out of the lab and bought a retail ComputerLand franchise and sold a lot of Apple IIs.
Nice to see so much has changed so quickly.
Thank you!
You in the video as well?
Thanks for highlighting these forward thinking and engineering techniques. Unless a person is part of the fabrication environment you don't hear about these things. Most fascinating. Your doing a great 😂.
Your doing amazing stuff, hurray! Keep it up!
Having taken an EE, hydraulics/pneumatics and chip fabrication course for 2 years (Sponsored by Infineon) - then becoming a union electrician that helped build the 200 and 300mm fabs in Richmond, VA - You did a fantastic job at explaining this. Good job! Intel might be a stock buy - nobody's talking about this yet.
what does the wafer taste like?
Chicken
Like gold, oil and lithium 😂
Backside power delivery 🤮
Like chips
It is tasteless (literally).
Hi Anastasia. My name is Roberto and I am from Italy.
I'm not a specialist in the topics you cover but, for some years now, i read about spintronics, skirmions and antiferroelectric materials as the electronics of the future.
Could you explain us these technologies and why they are non yet been implemented?
Thank you very much for your work of popularisation❤
Another game changer? What is this, like the 20th game-changer this year? I think we may be entering a game-changing bubble.
Well look at how the chip industry has changed in just a few years. Things really do move that fast.
I think the Chinese Photonic cheap is a better solution for overcoming the energy leakage. It offers a paradigm change.
looks with brains = explosion !!!!!. Weldone im a fan
@@eheroi Pooooooooo 💩💩
brilliant video! I am a bit frustrated that I found your channel just a couple of days ago.
I remember my classmates in the ´70s coming along with punched cards explaining a lot of nonsens what the machines in the IBM lab doing with that. The milestones in the processor technology during the last 50 years is not comparable to any other industry. IBM in the early days, ASML, TSMC, AMD and INTEL have changed life for the entire humankind (good or bad...I leave it open 🙂).
Great info! Looking great too (:
13:24 is not part of the EUV machine, but the vacuum chambers at Zeiss Oberkochen for mirror metrology.
Correct, my colleagues worked on that project.
i thought 14th gen intel used power via which is a form of back side power delivery. maybe i heard wrong or is this different from what tsmc is doing? but yeah would love to hear about ultra na
Outstanding contribution. Ranks among the tops! Very Moore;)
Congrats with ASM as sponsor. Truly a reward for your work.
The future is always so interesting. I love your chair. This manufacturer needs to sponsor your channel. If they haven't already.
The content on microchip technology is educational and truly fascinating. I am eager to discover more about this subject.
When the front side no longer works, you gotta go the back side. 😊
Intel is actually the first using Backside power Delivery not TSMC
11:00 I agree. This is super risky and everyone holding Intel stock should be careful. Intel is leveraging its chip production bet yet another time. I'm not sure this bet will work.
This channel is an excellent source for tech news. I'm sad I only found out today. Will be adding it to my list. Great video! well researched
Intel has had all kinds of problems with trying to introduce too much at the same time. That's part of the reason they had so many problems with their 10nm system, if I remember correctly.
Great video, I think I've found a new tech fiber to follow.
To do a back side power on a wafer would require pure silicon and microvias in the wafer through the die. Won't be easy to do this, but you could put the bigger stuff on the bottom and the computing stuff on the top.
I guess I saw this video a little late, but in a timeless industry such as chip manufacturing, where time never stands still, this video will be usable for a couple of years I would guess.
This topic really allows my inner nerd to fully geek out, and also understand how far in the future these chipmakers really are.
Considering that one of those ASML-machines takes about 2 years to build, the ordering process of them must start well ahead of time. It just makes my brain boil when I try to understand the whole chain.
Thank you very much for all of these wonderful videos.
Will BSP make it easier to cool the chip since the transistors are closer to the surface?
To follow up on the economics of High-NA - the reason the wafers-per-hour throughout is so slow is because the field size that is exposed on the wafer is half the size of standard EUV (429 mm² for high-NA vs 858mm²), thereby doubling the amount of exposures required per wafer and more or less doubling the process time per wafer (at least for the layers that require Hugh-NA EUV, which is only a small number compared to the total number of layers on the chips).
However, the smaller field size also has another major impact that directly affects their customers and chip designers. The field size determines the largest possible die size. And in recent years, as transistor counts have exploded with AI chips and all that, and with transistor shrink rates slowing, die sizes have been growing, with many commercial chips in the 500-600 mm² range. Monolithic (single die) chips of that size will be impossible to make on high NA, which may severely hamstring chip designers. So with the crazy upfront costs, high manufacturing cost due to the lower throughput, maz die size that is only half of previous generations, I imagine high NA is going to be a very hard sell for a lot of customers for a long time.
Anastasi,this is the best talk that I heard all day.I need to watch it again .
Congrats on the career change! Looking forward to seeing where you land or what you start.
This power-under-logic system is really interesting. One wonders what unexpected gremlins will be revealed in thermal performance of these designs. A fairly high amount of the waste heat per clock cycle is in power delivery itself, simply from resistance, so if you stick the power delivery on the other side of a silicon wafer- a lousy heat conductor- then what happens? Then there's the problem of the pinouts, needing to coat both sides in protective layers, etc., etc. I think this is going to prove very difficult!
The technology is mind blowing. I respect those in the industry that keep on pushing the tech forward. I wonder what is the limit? What is the actual vision all these scientists have in their mind of what is possible next 100 years. Born in 1991, I saw a rapid change in economy, tech,lifestyle. Maybe no need for 100 years to see massive changes this time around.
This sort of thing was pioneered for Solar Cells maybe 20 years ago, by a company called Sunpower.
Their motivation for back-side power connectors was to maximize the amount of silicon exposed to sunlight.
It is interesting that this tech also helps with the new generation of VLSI chips.
Finally the gate looks like a real gate.
thank you, this was an awesome video and it was NOT to long. the knowledge and breaking down the structures of the nextgen chips is helping my information gathering as we go into the future at blazing rates of time and speed. YES, please complete the CFET video as I will be wait to consume the information. understanding the chips and new memory and coding is imperative for any of use in Tech or stem. there is not enough hours in a day to stay up on the massive amounts of new data on new methods and tech. again thank you for you work and channel
I really hope Intel's gamble pays off here, they have really been struggling with fab issues over the past couple of nodes and it is starting to impact their competitiveness which isn't a good thing in an industry where there are now so few players. Looking at More's law from an economic perspective is actually something I have thought of but it actually makes more sense than from a feature size perspective. Gate all round is going to be a game changer especially for mobile devices because of the power reduction achieved, IC design really is entering another era of rapid progress.
How did we beat electron quantum tunnelling? It was a huge problem in 2001
Tunneling happens when the gap is easy to jump for an electron. By making the gates taller it intercepts electrons trying to jump the gap.
Cooling. Cooling needs to advance and the cooling will need to be applied to both sides of the chips. Moor's law marches on.
Nothing except a healthy competition can protect us consumers from greedy corporations. For this reason only, I hope Intel doesn't lag behind too far.
This gives me the best understanding of this difficult topic, so far.