Improve your career using my code “Anastasi” for 30% off on all their programs! Sign up for a FREE TripleTen career consultation with my link: get.tripleten.com/AnastasiInTech
Other thing. Why there is no an hybrid solution with HBM and GDDR memories in commercial Graphics cards?. Imagine, 12GB of GDDR and 4 of HBM, both with a parallel access to GPU with a priority of LOD?, it would be awesome, more performace and stability a reduced power consumption.
I would like to know if you thing that this new discovery at MIT might be a game changer in your tech interest... ua-cam.com/video/17Y82tJDk2o/v-deo.html
A thought, inall my years I've seen many wonder technologies or materials that were supposed to supplant semiconductors as we use them today, and composed today, and never did. What seems to be missing from most of these is the relative gain capability, therefore allowing for a clear signal to pass through the transistors one after the other. Or in the materials that do manage to have the same or better gain, they are so very expensive, that they were uneconomical even 30 years ago. Is this new material both potentially economical compared to current technologies, and does it provide an equivalent amount of gain? Or get around that by providing an equivalent mechanism?
any time some tech announces they "mimic the human brain" you have to keep this in mind as a baseline: 1. neurons fire using their own power, completely independently of each other (i.e. true parallelism); 2. neurons are highly interconnected -- up to 10k dendritic connections each, all working in parallel; 3. activation has inherent decay, which means there is no external mechanism for creating these spikes; 4. lastly, and imho most important: memory is NOT some localized thing that is analogous to a bit that is on or off. memorization equals recall! in other words, in networks of neurons there is ZERO distinction between storing information and retrieving it. maybe we should be a little reluctant to compare things like back-prop nets or neuromorphic computing to the brain.
I've been in the semiconductor industry since 2001... I currently work in a lithography machine company and always wonder what comes next? Meaning, we must be extremely close to the absolute limits of optically-based circuit production. I've looked into Canon, EV Group (and similar) NIH Technology, but even that will have physical limits (the sheer fragility of molecular structures). I wonder when either new materials, methods, or completely new tech will "overtake" the current industry production methodologies. No doubt, we've come a LONG way since the early days. I'm going to believe, soon we will see what some will consider "alien" tech. Thanks for all the great content! ...and, thank you for doing your research and development in chip design. You are what some consider, "high-value" gray matter! Keep up the GREAT videos!!!
I love your channel so much! I am an electronics repair tech and I have spent the last 14 years working on everything from bare boards to thermal imaging cameras. Thank you so much for showing people this process and the details behind it all.
best and favorite tech channel on youtube! Love hearing all about new tech and research and i can always tell you put such effort and passion into these videos! much apreciated //guy from sweden!
I love the start of the video mentioning Intel with a breakthrough of 90% and then shows an AMD processor. I know it's proberbly just b-roll footage but I found it funny. Great video as always, Anastasi.
Because I like to invest in companies delving into futuristic discoveries and development, I find all your videos on this topic fascinating. You widen my range and understanding in this realm of technology. Thanks. Like you, I look into the future of things. We can't do anything about the past but we can try to prepare for the future.
@@realryder2626 Before you called my attention to it, I did not know that Netlist existed. Thanks for making me aware of it. I only just looked at it briefly. Based on what I quickly read, I think it is an interesting stock because of the copyright infringement by Samsung. If Samsung used Netlist's property, then there must be something to it. Because the high bandwidth memory or HBM and DDR5 memory of the patent are foundational to generative artificial intelligence ("AI") computing, the company is worth further investigation. At its current closing price of $1.24 , I would gamble with a bid of $1.20 for somewhere between 10 - 100 shares. The more I learn, the more I would change my bet. I look at investing akin to playing Blackjack at a casino and card counting is akin to learning more about a stock. My philosophy is that you should never spend more than you are willing to lose totally. At this point, I am gambling only with the "house's money"(= dividends). So its no big deal if I lose playing this hand(buying stock in Netlist). I play to have fun(= trying to "outsmart" the house) not to try to win big. If you lose Big, then gambling is NO Fun. I gamble on a stock's potential if it is a startup, especially if it has some type of "moat" to protect it. From what I can tell by briefly looking at it, there seems to be a "moat" that has stopped Samsung at least for now. Samsung may just buy Netlist if it really wants the patent or it might try to buy a license right to use the product.
@@realryder2626 Looks interesting if you are willing to gamble. I have to look into it more but from what I read, I would place a bid for 10 to 100 shares @ $1.20/share. As I learn more, I would most likely change the bid. If you don't gamble, you never have a chance to win. But don't gamble if you can't afford to lose it.
This direction of inquiry is fascinating. I appreciate the long haul of tech as when I learned the basics of electronics transistors were individual parts to sub for vacuum tubes.
I do not want to be sarcasm killer in this current situation, but Intel can handle it. They just got the latest Tech from ASM/ASML/TSMC and fired their old staff ;-)
No transistors = no nukes + no ICBMs + no drones = no chance to wipe ourselves out. Also no twitter and no musk and trump bullying the entire world with their bullcrap 24/7. No CNN, no fox news, no social medias, no onlyfans, no crypto, no techbro fascists, no neonazis spreading hate worldwide, no russian interfering with elections everywhere... What a peaceful world.
Your explanation of electron quantum tunnelling at 2:40 reminded me of the old joke: "Q. Why did the horse run over the hill? A. Because it couldn't run under it!" 😆
Your videos about scientists doing research on new transistor technologies are always fascinating❤️❤️❤️ About new techs, i would like to see you do a video, sometime in the future, on wafer-to-wafer / die-to-wafer hybrid bonding and trench 3d nand and how they going to help keep Moore's law alive for some years to come❤️
I have to mention, that at 3:00, the wave after the barrier should be much shorter because the amplitude shows probability, and it shrinks exponentially inside the barrier.
This is mind blowing. Thank you for the interesting explanations on developments in nanotech. I've grasped as much as a whole chapter of a book in a few minutes. Your work is very much appreciated.
@UCXE2nJ11YoJ2YzqrsQJOOdQ Yes and we would not have a 48k Spectrum at home if it wasn't for micro processors and chip's. I was 10 when i first got mine for Christmas. Because of that. I passed my I.T Course in 1991 when I was 17.
Super interesting & super well presented. The low power & relatively simple architecture is so attractive, bringing great opportunities for mobile systems. Tunneling is such an amazing property of the quantum world, allowing applications that are impossible in classical physics. Great how quantum physics & new materials are expanding our technical capabilities & there are so many potential applications for this approach. Thank you for sharing!
We've been told about these new "break through materials" for an insanely long time, but no matter how good, and how easily adopted, no Fabs are going to take it on unless they've really shown the technology is ready which is a goooood few years away.
SOunds to me like we are still gathering data on the Electron and its magical hop across the transistor . Its amazing break through ? Or are we still hatching data from this find? I love this channel Anastasi Thank you for being HUMAN !
It's so wonderful to find a channel where it is obvious that extensive research was done prior to production. Now add a gorgeous and intelligent presenter, and you have this channel. Fantastic!!!
Apparently there are nano-scale vacuum transistors (en.wikipedia.org/wiki/Nanoscale_vacuum-channel_transistor), which are made using lithography and whatnot, and due to the tiny size, don't actually need to be vacuums because the gap is so small that collisions with molecules of gas become a negligible problem.
Reminds me of a poem called "ode to the microwave oven" The last line of the poem went as follows.. "and now her cooking is just as bad but faster..." have we reached a tipping point where there is nothing more to gain in our quality of life and where surveillance and intrusions are going to make life much more difficult and complex...🤔🤨
Material like graphene coming into their own is pretty exciting, though this kind of application wasn't what I'd expected. Thanks for sharing your research in a way an old fellow can begin to understand. 👍☺👍
Could you cover silicon photonics, that will replace copper in gpus by photonic communications between cores. Companies into this are POET technologies, Sivers Semiconductors from sweden, and Navitas Semiconductors, all working with big companies. Poet is working with Foxconn to develop this.
Super interesting and promising, I think development of chips just becomes more interesting because of the limits we face. Thanks for your great vids always ^^
Leave it to IBM to look at spiking neural networks and assume that must be how human brains work. Something tells me that human brains work with different cells that do different things... Eistien didn't have tons of neurons, but he did have lots if extra glual cells, for intance. Glia cells seem to be a support cell for neurons, but who knows what other functions they might have
I like to think of it like water filling a basin sink. You can plug it to keep the volume and open to release the flow. We basically created the overflow system
Wow. The whole of Semi is the quantum band states so it doesn't surprise me that quantum tunneling is now the target of the Semi industry. I mean 0.1v is amazing for an operational point of a transistor. What may be concerning is the switching frequency, but I would bet that is a materials issue and how quickly a quantum tunnel is created and destroyed to pass the electron or hole. Very cool tech. Thank you for the video. Very informative.
I'm sure that there will be a very large number of applications for mobile devices due to the low voltages and low losses. Many of these applications simply don't need blistering performance in the GHz range. Certainly IOT could make good use of this. Maybe this will be paired with those high efficiency indoor Solar Cells to produce products without batteries. This is a fantastic channel, keep up the good work!
Great video, Anastasia! Your explanation of quantum tunneling and its implications for new semiconductor technologies was clear and fascinating. The way you broke down complex concepts like the duality of electrons and the principles behind tunnel transistors made it accessible even for those without a deep physics background. I especially appreciated your enthusiasm when discussing the potential efficiency gains in new microchip designs. Building on your insights, I would like to know if you've heard about a speculative mathematical framework called Tribernachi Math. While highly theoretical, it offers some intriguing ideas relevant to quantum tunneling in new semiconductors, especially with materials like graphene. Here are a few ways Tribernachi Math could potentially contribute to this field: 1. Discrete energy states: Tribernachi Math proposes that quantum states correspond to prime numbers. This could offer a new way to model the discrete energy levels in semiconductor bands, potentially leading to more accurate predictions of tunneling behavior. 2. Three-dimensional modeling: The framework emphasizes volume-based measurements, which could be particularly useful for modeling tunneling in 2D materials like graphene, where out-of-plane effects can be significant. 3. Bounded infinity concept: This idea may help address some computational challenges of modeling quantum effects at tiny scales, allowing for more efficient simulations of tunneling phenomena. 4. Novel mathematical operations: The unique operators in Tribernachi Math could provide new tools for calculating tunneling probabilities and current flow in advanced semiconductor devices. While this is all highly speculative, it's exciting to think about how new mathematical approaches might help us better understand and harness quantum tunneling in next-generation semiconductors. What are your thoughts on exploring unconventional mathematical frameworks in semiconductor physics? Do you think there's value in these kinds of speculative approaches? Thanks again for the great content, Anastasia. I am looking forward to your next video!
Tunnelling can be viewed as virtualization. Like the object virtually existing within the barrier and maintaining it's properties or information, id even go as far as saying it's natural computation, if i model and simulate a particle inside a computer program it's similar to the computer and i could print it out as different structures 2d, 3d, etc but unlike modelling and simulation Tunnelling doesn't leave the original object out of the natural computer, the physical object or garbage is input to become information that is part of the barrier being the computer which is outputted on the other side of the barrier, the advantage of this computation is you can ise one object to make multiple similar objects or clones or copies, but i believe copies are objects completely different from the original so that is the risk for quantum tech for me, like a trojan
I have never before bought a stock, but this news makes me consider it. I would invest in the manufacturing process of this new material. This TFET technology will revolutionise our digital world completely.
wow , I am extremely happy to have stumbled upon your video. This is really educative. I am currently undergoing my undergrad in computer engineering and I am really keen on developing myself to become a very good hardware engineer. where I am from there is no company that deals in chip production. Please can you give me advice on what I can do to get hands on experience before I finish the program? I am currently in my first year.
❤❤❤ Anastasi, danke für eine weitere sehr lehrreiche Episode. Ich liebe sie und ich liebe dich auch. Wenn ich groß bin, hoffe ich, ein Mikrochip-Experte wie du zu werden.🌹🌹🌹🌹😘😘
The SSDs already use quantum tunneling effect. I was wandering when big techs companies would start to use it on processors too. I'm happy seeing this when I'm alive.
you can use this "bad" effect for free energy by using a low work function material, a nanogap , and a high work functioning material. it's called an evercell (Clark Boyd old dominion university , virginia) and the spontaneous electron tunneling creates a small but eternal current. you make them in arrays to get more power
I’m curious to see how this manifests in hardware 5 or 6 for full self driving and other complex robotic tasks. I’m also interested in the pace of development as we clearly have been on a plateau for a while right now. Maybe discussing where and how we are likely to see progress in real world applications 1 year from now, 3 years and also 5 years into the future as we may be entering the most interesting era in human history.
It would be nice to make transistors for the low power, high frequency uses. However, there are certainly many use cases for low power transistors that work in the GHz range as well, specifically battery powered devices.
I have watched switching voltages over the years. 5V TTL, then dropping, dropping, dropping... The i486 and PCI bus enjoyed much lower switching voltage requisites. We have come a long way. But 0.1 V That's just WOW. Way less juice needed. Way less heat generated, and much faster transitions. That's just WOW... considering how far we have already come since even the i486. Do you have a link to you thesis paper?
"my many years in" honey.... you are young. you have lived for many years, sure. but i got shirts older than you.. many years. gave me a good chuckle that :) pretty one too this
Wouldn't it be amazing if it turned out that quantum computing and conventional Moore's Law manufacturing could meet in the middle? That could potentially solve many of the very hard problems in both fields. It is a very exciting thought for me.
Tunneling was initially described by physicist George Gamov, his theory did not obey the quantum mechanical “essential” (Max) Born rule, so Born published his own version of qm tunneling that satisfies his “essential” rule. However, in de context of Pilot Wave theory, the Born rule is not essential (not always true), which means that qm processes like tunneling electrons can benefit from Pilot wave energy to gain in potential energy, and this meand that qm tunneling transistors can achieve efficiencies of over 100%, by converting Pilot wave energy into potential electric energy. This is the dawn of powerful ‘free energy’ devices that are independent of solar energy, since the Pilot wave energy exchange with any particle in the universe is huge.
Was spending a lot more time outdoors would be one good thing about it, right? Yes, they are going through a substance such as an oxide, but electrons pass through solids all the time, metals, and it's not that the atoms of the insulator are closer together than the atoms of the metal and therefore block the electrons, it has to do with energy levels inside the substance and not with physical barriers.
About halfway through the video at the moment.... There are two things I would like to mention as items of academic discussion. First, I have actually wondered if the future will involve a form of data storage and flow which is a bit paradoxical. Most losses in a CPU will come from ohmic losses and switching losses which accompany depletions of gate charges and capacitive biases. The movement of static charges is lossy. By contrast, if I have an LC resonant tank circuit passing that charge back and forth, the properties of the capacitor and inductor drive voltage and current out of phase, which results in power dissipation being significantly reduced. While it would certainly take up more area and be a hassle to get it to work correctly, a CPU architecture which worked closer to a zvs tank circuit at the logic gate level could be less lossy than traditional architectures, at least where heavy processing loads are expected. Second, I often wonder if vacuum tubes will make a return, almost specifically because of the challenges of using modern CPUs outside the shielding of Earth's ionosphere and atmosphere. Cosmic rays corrupt RAM and cache memory like no one's business. Without mosfet and related technologies, I have wondered if we still would not have seen development of microscopic arrays of vacuum tubes and analog computers similar to today - obviously different in specs and capabilities, but if we were to miniaturize vacuum tubes or related devices to produce miniaturized devices, how far could we go? Power electronics would be the biggest area where we'd be at a loss. I am not sure something like a variable frequency drive could be realized without some weird motor-to-generator setup. Though at that point, exotic power vacuum tubes would be just as reasonable to invest in. Though, to be fair, part of me is just biased. A friend of mine owns an arcade and I fully intend to redevelop CRT manufacture simply because nothing quite matches the phosphor glow of a real CRT. I figure I may as well invest in vacuum tube technologies while I am at it, and see what can be done with modern materials and micro/nano fabrication. As for materials... Silicon carbide has been known about for some time and it seems more useful for power components than processors... Gallium arsenic has been used in microwave fpgas for some time and gallium nitride seems to be generally superior substitute... But there must be some reason it hasn't been instantly swept into CPU manufacture, yet, aside from just cost. Maybe germanium, but again, if that isn't already in use then there's a reason for it. Titanium nitride had some interesting band gap properties and has already been in use for certain layers... Diamond substrate has been known about and played with in laboratories for some time, and it would be odd if all the various studies into deposition processes hadn't resulted in reliable diamond layers on the commercial scale by now. Graphene has been a candidate and has some fun properties that make it more ideal than diamond, but doping of a "1 dimensional" lattice is a bit of an adventure in fundamental forces... Some kind of magic insulator that increases its band gap in response to a difference in voltage would be amusing... I was horribly disappointed in elementary school when the science book had a picture of a robot and we did build, study, or learn anything about said robot in the class that year. Someone says "new thing" and my mind runs away with it... I'll be quiet and listen, now.
Electrons aren't the only particles that exhibit wave-particle duality behavior. The whole underlying idea behind wave-particle duality is that all subatomic entities exhibit properties of both models while not fitting exclusively neatly into either, therefore we need a way of reconciling the two. At its core, the issue is that subatomic entities aren't really "waves" or "particles" but instead some sort of variably quantized areas of space that we do not fully understand.
Improve your career using my code “Anastasi” for 30% off on all their programs! Sign up for a FREE TripleTen career consultation with my link: get.tripleten.com/AnastasiInTech
Hello anastasi. Which ranking is it reliable to know where can i study Computer Hardware Engenieering?
Other thing. Why there is no an hybrid solution with HBM and GDDR memories in commercial Graphics cards?. Imagine, 12GB of GDDR and 4 of HBM, both with a parallel access to GPU with a priority of LOD?, it would be awesome, more performace and stability a reduced power consumption.
You talk like you have decades of experience but look like 20 😀
I would like to know if you thing that this new discovery at MIT might be a game changer in your tech interest... ua-cam.com/video/17Y82tJDk2o/v-deo.html
A thought, inall my years I've seen many wonder technologies or materials that were supposed to supplant semiconductors as we use them today, and composed today, and never did. What seems to be missing from most of these is the relative gain capability, therefore allowing for a clear signal to pass through the transistors one after the other. Or in the materials that do manage to have the same or better gain, they are so very expensive, that they were uneconomical even 30 years ago. Is this new material both potentially economical compared to current technologies, and does it provide an equivalent amount of gain? Or get around that by providing an equivalent mechanism?
I'm a physicist and applied mathematician, and I appreciate your video. Good job.
any time some tech announces they "mimic the human brain" you have to keep this in mind as a baseline: 1. neurons fire using their own power, completely independently of each other (i.e. true parallelism); 2. neurons are highly interconnected -- up to 10k dendritic connections each, all working in parallel; 3. activation has inherent decay, which means there is no external mechanism for creating these spikes; 4. lastly, and imho most important: memory is NOT some localized thing that is analogous to a bit that is on or off. memorization equals recall! in other words, in networks of neurons there is ZERO distinction between storing information and retrieving it. maybe we should be a little reluctant to compare things like back-prop nets or neuromorphic computing to the brain.
I've been in the semiconductor industry since 2001... I currently work in a lithography machine company and always wonder what comes next? Meaning, we must be extremely close to the absolute limits of optically-based circuit production. I've looked into Canon, EV Group (and similar) NIH Technology, but even that will have physical limits (the sheer fragility of molecular structures). I wonder when either new materials, methods, or completely new tech will "overtake" the current industry production methodologies. No doubt, we've come a LONG way since the early days. I'm going to believe, soon we will see what some will consider "alien" tech. Thanks for all the great content! ...and, thank you for doing your research and development in chip design. You are what some consider, "high-value" gray matter! Keep up the GREAT videos!!!
ua-cam.com/video/QFYyduL9WZA/v-deo.html&pp=ygUaTm9iZWwgTGVjdHVyZSAyMDI0IFBoeXNpY3M%3D
A.I is overestimated in the short term and underestimated in the long term
I love your channel so much! I am an electronics repair tech and I have spent the last 14 years working on everything from bare boards to thermal imaging cameras. Thank you so much for showing people this process and the details behind it all.
best and favorite tech channel on youtube! Love hearing all about new tech and research and i can always tell you put such effort and passion into these videos! much apreciated //guy from sweden!
I love the start of the video mentioning Intel with a breakthrough of 90% and then shows an AMD processor. I know it's proberbly just b-roll footage but I found it funny. Great video as always, Anastasi.
I absolutely love how clear and entertaining you are making these videos on these pretty complex and heavy topics
Because I like to invest in companies delving into futuristic discoveries and development, I find all your videos on this topic fascinating. You widen my range and understanding in this realm of technology. Thanks. Like you, I look into the future of things. We can't do anything about the past but we can try to prepare for the future.
so you invest in Intel?
@@bimuhu Not INTEL. Too expensive. I prefer to gamble(invest) in start-ups such EVTOLs and GaN semiconductors and not mature technologies
@DrinkingStar what's your take on Netlist?
@@realryder2626 Before you called my attention to it, I did not know that Netlist existed. Thanks for making me aware of it. I only just looked at it briefly. Based on what I quickly read, I think it is an interesting stock because of the copyright infringement by Samsung. If Samsung used Netlist's property, then there must be something to it. Because the high bandwidth memory or HBM and DDR5 memory of the patent are foundational to generative artificial intelligence ("AI") computing, the company is worth further investigation. At its current closing price of $1.24 , I would gamble with a bid of $1.20 for somewhere between 10 - 100 shares. The more I learn, the more I would change my bet. I look at investing akin to playing Blackjack at a casino and card counting is akin to learning more about a stock. My philosophy is that you should never spend more than you are willing to lose totally. At this point, I am gambling only with the "house's money"(= dividends). So its no big deal if I lose playing this hand(buying stock in Netlist). I play to have fun(= trying to "outsmart" the house) not to try to win big. If you lose Big, then gambling is NO Fun. I gamble on a stock's potential if it is a startup, especially if it has some type of "moat" to protect it. From what I can tell by briefly looking at it, there seems to be a "moat" that has stopped Samsung at least for now. Samsung may just buy Netlist if it really wants the patent or it might try to buy a license right to use the product.
@@realryder2626 Looks interesting if you are willing to gamble. I have to look into it more but from what I read, I would place a bid for 10 to 100 shares @ $1.20/share. As I learn more, I would most likely change the bid. If you don't gamble, you never have a chance to win. But don't gamble if you can't afford to lose it.
This direction of inquiry is fascinating. I appreciate the long haul of tech as when I learned the basics of electronics transistors were individual parts to sub for vacuum tubes.
Your videos are always very interesting and well-explained. Please continue keeping us informed about the latest developments in this area.
I can’t tell if the 8:40 comment about intel was sarcasm or not 👀 😂
definitely sarcasm
Check out Paul's Hardware "Intel just might be F***ed - Tech News Aug 4".
I do not want to be sarcasm killer in this current situation, but Intel can handle it. They just got the latest Tech from ASM/ASML/TSMC and fired their old staff ;-)
@@zbaktube lol, do they still have their own fabs ? can I use their fabs to make my own 80486DX66 I always wanted to make a clone since I was 6
That smile they cut off. I'm thinking she couldn't keep the laugh in :D
More time outdoors? Fewer gadgets? That actually sounds kind of appealing.
Yes, but having a choice is still always better.
No transistors = no nukes + no ICBMs + no drones = no chance to wipe ourselves out. Also no twitter and no musk and trump bullying the entire world with their bullcrap 24/7. No CNN, no fox news, no social medias, no onlyfans, no crypto, no techbro fascists, no neonazis spreading hate worldwide, no russian interfering with elections everywhere... What a peaceful world.
Go get a dog, you'll be more outdoors and the dog will chew all your gadgets 😂
@@randomnumbers84269 agreed.
Yes you'll be outdoors but more dumber, since you have less access to knowledge online.
"Current situation" great pun! :D
Your explanation of electron quantum tunnelling at 2:40 reminded me of the old joke: "Q. Why did the horse run over the hill? A. Because it couldn't run under it!" 😆
That benefits of transistors speech backfired...
Your videos about scientists doing research on new transistor technologies are always fascinating❤️❤️❤️
About new techs, i would like to see you do a video, sometime in the future, on wafer-to-wafer / die-to-wafer hybrid bonding and trench 3d nand and how they going to help keep Moore's law alive for some years to come❤️
This woman is a knock out, but it's her professionalism that does it. Very well done videos
😂 concentrate sir, she is saying very important things
@@dennisalbert6115 oh I do. Transistors something something.
Another excellent video Anastasi! 👍 This could revolutionize the industry.
I have to mention, that at 3:00, the wave after the barrier should be much shorter because the amplitude shows probability, and it shrinks exponentially inside the barrier.
This is mind blowing. Thank you for the interesting explanations on developments in nanotech. I've grasped as much as a whole chapter of a book in a few minutes. Your work is very much appreciated.
It's amazing how far we have come.
@UCXE2nJ11YoJ2YzqrsQJOOdQ Yes and we would not have a 48k Spectrum at home if it wasn't for micro processors and chip's. I was 10 when i first got mine for Christmas. Because of that. I passed my I.T Course in 1991 when I was 17.
Super interesting & super well presented. The low power & relatively simple architecture is so attractive, bringing great opportunities for mobile systems. Tunneling is such an amazing property of the quantum world, allowing applications that are impossible in classical physics. Great how quantum physics & new materials are expanding our technical capabilities & there are so many potential applications for this approach. Thank you for sharing!
The green upward trending arrow in your thumbnail needs to go from left to right.
McPherson junction used this quantum behavior many years ago. Do a show on this historical event?? 😁😁😁
Great video--i'm a time and gravity scientist so you show my where time must has to change at that point to time traval !!!
Omg that Ciao at the end 🇮🇹💙🇨🇭
The algorithm brought me here. I have never seen you before. Great video and very excited to see how it goes ✌️🇬🇧 Earned a sub
Thank you for giving me a glimpse of prehistoric times before unraveling the truth of new chips.........
You are clearly in love with transistors and that's adorable. Love your enthusiasm...
We've been told about these new "break through materials" for an insanely long time, but no matter how good, and how easily adopted, no Fabs are going to take it on unless they've really shown the technology is ready which is a goooood few years away.
SOunds to me like we are still gathering data on the Electron and its magical hop across the transistor . Its amazing break through ? Or are we still hatching data from this find? I love this channel Anastasi Thank you for being HUMAN !
Fascinating! I picture power being applied to stop tunneling, then removed to "open the gate."
Without transistors we would have some AWESOME microtube technology!
Amazing Video, new tech is always exiting !
Thank you for explaining
Love this channel
It's so wonderful to find a channel where it is obvious that extensive research was done prior to production. Now add a gorgeous and intelligent presenter, and you have this channel. Fantastic!!!
@TE-LE-GRAM-AnastasiInTech I don't do scams.
Beauty, brains, excellent clear explanations, and an amazing accent. What a gift.
Very good ! Congrats. I learned a lot with the shows in this channel.
Good morning Anastasi, greetings and salutations from across the pond 👋 ❤😊
I wonder if vac tubes would have been shrunk to miniature and made highly efficient, if given time to mature...
Apparently there are nano-scale vacuum transistors (en.wikipedia.org/wiki/Nanoscale_vacuum-channel_transistor), which are made using lithography and whatnot, and due to the tiny size, don't actually need to be vacuums because the gap is so small that collisions with molecules of gas become a negligible problem.
"Intel has it under control." I saw that sly look! :)
Reminds me of a poem called "ode to the microwave oven" The last line of the poem went as follows.. "and now her cooking is just as bad but faster..." have we reached a tipping point where there is nothing more to gain in our quality of life and where surveillance and intrusions are going to make life much more difficult and complex...🤔🤨
Thank you!
Excellent human mind efficiency
Material like graphene coming into their own is pretty exciting, though this kind of application wasn't what I'd expected. Thanks for sharing your research in a way an old fellow can begin to understand. 👍☺👍
Could you cover silicon photonics, that will replace copper in gpus by photonic communications between cores. Companies into this are POET technologies, Sivers Semiconductors from sweden, and Navitas Semiconductors, all working with big companies. Poet is working with Foxconn to develop this.
Thank you... I am always.
Transformer components are a must in that circuit.
Came for the news - stayed for the tech puns
Like the "current transistor problem"?
Love the way she explains. Awesome lady.
In audio what you call "quantum tunneling" we call crosstalk. However crosstalk has more to do with inductance leakage.
Super interesting and promising, I think development of chips just becomes more interesting because of the limits we face. Thanks for your great vids always ^^
Without transistors I wouldn't be watching your beautiful channel and your beautiful face. 😊
Thanks for another great video!
Leave it to IBM to look at spiking neural networks and assume that must be how human brains work. Something tells me that human brains work with different cells that do different things... Eistien didn't have tons of neurons, but he did have lots if extra glual cells, for intance. Glia cells seem to be a support cell for neurons, but who knows what other functions they might have
I like to think of it like water filling a basin sink.
You can plug it to keep the volume and open to release the flow.
We basically created the overflow system
You make computer science easy to understand,RESPECT to you Pretty GIRL 😻😁🔥🔥🔥❤️🦁
Wow. The whole of Semi is the quantum band states so it doesn't surprise me that quantum tunneling is now the target of the Semi industry. I mean 0.1v is amazing for an operational point of a transistor. What may be concerning is the switching frequency, but I would bet that is a materials issue and how quickly a quantum tunnel is created and destroyed to pass the electron or hole.
Very cool tech. Thank you for the video. Very informative.
I'm sure that there will be a very large number of applications for mobile devices due to the low voltages and low losses. Many of these applications simply don't need blistering performance in the GHz range. Certainly IOT could make good use of this. Maybe this will be paired with those high efficiency indoor Solar Cells to produce products without batteries.
This is a fantastic channel, keep up the good work!
Great video, Anastasia! Your explanation of quantum tunneling and its implications for new semiconductor technologies was clear and fascinating. The way you broke down complex concepts like the duality of electrons and the principles behind tunnel transistors made it accessible even for those without a deep physics background. I especially appreciated your enthusiasm when discussing the potential efficiency gains in new microchip designs.
Building on your insights, I would like to know if you've heard about a speculative mathematical framework called Tribernachi Math. While highly theoretical, it offers some intriguing ideas relevant to quantum tunneling in new semiconductors, especially with materials like graphene. Here are a few ways Tribernachi Math could potentially contribute to this field:
1. Discrete energy states: Tribernachi Math proposes that quantum states correspond to prime numbers. This could offer a new way to model the discrete energy levels in semiconductor bands, potentially leading to more accurate predictions of tunneling behavior.
2. Three-dimensional modeling: The framework emphasizes volume-based measurements, which could be particularly useful for modeling tunneling in 2D materials like graphene, where out-of-plane effects can be significant.
3. Bounded infinity concept: This idea may help address some computational challenges of modeling quantum effects at tiny scales, allowing for more efficient simulations of tunneling phenomena.
4. Novel mathematical operations: The unique operators in Tribernachi Math could provide new tools for calculating tunneling probabilities and current flow in advanced semiconductor devices.
While this is all highly speculative, it's exciting to think about how new mathematical approaches might help us better understand and harness quantum tunneling in next-generation semiconductors. What are your thoughts on exploring unconventional mathematical frameworks in semiconductor physics? Do you think there's value in these kinds of speculative approaches?
Thanks again for the great content, Anastasia. I am looking forward to your next video!
Interesting. I wonder if those chip sets will be stackable for applications like quantum computing?
Very interesting will look into it more.
UCSB EE Alumni here 💙💛
Tunnelling can be viewed as virtualization. Like the object virtually existing within the barrier and maintaining it's properties or information, id even go as far as saying it's natural computation, if i model and simulate a particle inside a computer program it's similar to the computer and i could print it out as different structures 2d, 3d, etc but unlike modelling and simulation Tunnelling doesn't leave the original object out of the natural computer, the physical object or garbage is input to become information that is part of the barrier being the computer which is outputted on the other side of the barrier, the advantage of this computation is you can ise one object to make multiple similar objects or clones or copies, but i believe copies are objects completely different from the original so that is the risk for quantum tech for me, like a trojan
@8:35 Sarcasm? I can't tell :)
The new graphics on wave particle duality (plus Heisenberg uncertainty added) is interesting. 😊
I have never before bought a stock, but this news makes me consider it. I would invest in the manufacturing process of this new material. This TFET technology will revolutionise our digital world completely.
@on-TE-LE-GRAM-AnastasiInTech Huh? What did I win? 😀
wow , I am extremely happy to have stumbled upon your video. This is really educative. I am currently undergoing my undergrad in computer engineering and I am really keen on developing myself to become a very good hardware engineer. where I am from there is no company that deals in chip production. Please can you give me advice on what I can do to get hands on experience before I finish the program? I am currently in my first year.
❤❤❤ Anastasi, danke für eine weitere sehr lehrreiche Episode. Ich liebe sie und ich liebe dich auch. Wenn ich groß bin, hoffe ich, ein Mikrochip-Experte wie du zu werden.🌹🌹🌹🌹😘😘
@4:32 LMAO!!! Ahhh! I see what you did there! :D :D :D
The hill just doesn't dampen the electromagnetic field enough to stop the wave packet named electron.
Imagine all the protection necessary for a transistor operating @ 0.1 volts. I sneeze and generate more voltage than that!
The SSDs already use quantum tunneling effect. I was wandering when big techs companies would start to use it on processors too. I'm happy seeing this when I'm alive.
you can use this "bad" effect for free energy by using a low work function material, a nanogap , and a high work functioning material. it's called an evercell (Clark Boyd old dominion university , virginia) and the spontaneous electron tunneling creates a small but eternal current. you make them in arrays to get more power
I’m curious to see how this manifests in hardware 5 or 6 for full self driving and other complex robotic tasks. I’m also interested in the pace of development as we clearly have been on a plateau for a while right now. Maybe discussing where and how we are likely to see progress in real world applications 1 year from now, 3 years and also 5 years into the future as we may be entering the most interesting era in human history.
this topic is kinda alien to me. but i like to learn something new, especially via such beautiful and nice package 🙂
Easy answer - autonomous (self driving) cars. Though leaps have been made, this technology is constantly 10 years away.
It would be nice to make transistors for the low power, high frequency uses. However, there are certainly many use cases for low power transistors that work in the GHz range as well, specifically battery powered devices.
I have watched switching voltages over the years. 5V TTL, then dropping, dropping, dropping... The i486 and PCI bus enjoyed much lower switching voltage requisites. We have come a long way. But 0.1 V That's just WOW. Way less juice needed. Way less heat generated, and much faster transitions. That's just WOW... considering how far we have already come since even the i486.
Do you have a link to you thesis paper?
Thinking... 💡 transistor invention is the thing that perhaps changed human life style the most 💡
Quantum tunneling has been used for years - this is how E(E)PROM works, and this is with a SO2 layer of ~90nm.
8:39 oh you cheeky cheeky woman... That really made me laugh)
"my many years in" honey.... you are young. you have lived for many years, sure. but i got shirts older than you.. many years. gave me a good chuckle that :) pretty one too this
Thanks for the video!
Small idea - did you think about creating a video about Nvidia's Blackwell delay and TSMC's cowos?
Ah yes ideas like these struck me when i see them or soon before i see them
Wouldn't it be amazing if it turned out that quantum computing and conventional Moore's Law manufacturing could meet in the middle? That could potentially solve many of the very hard problems in both fields. It is a very exciting thought for me.
Love your videos ❤
Tunneling was initially described by physicist George Gamov, his theory did not obey the quantum mechanical “essential” (Max) Born rule, so Born published his own version of qm tunneling that satisfies his “essential” rule. However, in de context of Pilot Wave theory, the Born rule is not essential (not always true), which means that qm processes like tunneling electrons can benefit from Pilot wave energy to gain in potential energy, and this meand that qm tunneling transistors can achieve efficiencies of over 100%, by converting Pilot wave energy into potential electric energy. This is the dawn of powerful ‘free energy’ devices that are independent of solar energy, since the Pilot wave energy exchange with any particle in the universe is huge.
Her videos are so good. Suggest me some good US universities which are affordable for my next year. Im planning to get Master's degree
Was spending a lot more time outdoors would be one good thing about it, right? Yes, they are going through a substance such as an oxide, but electrons pass through solids all the time, metals, and it's not that the atoms of the insulator are closer together than the atoms of the metal and therefore block the electrons, it has to do with energy levels inside the substance and not with physical barriers.
About halfway through the video at the moment....
There are two things I would like to mention as items of academic discussion. First, I have actually wondered if the future will involve a form of data storage and flow which is a bit paradoxical. Most losses in a CPU will come from ohmic losses and switching losses which accompany depletions of gate charges and capacitive biases. The movement of static charges is lossy. By contrast, if I have an LC resonant tank circuit passing that charge back and forth, the properties of the capacitor and inductor drive voltage and current out of phase, which results in power dissipation being significantly reduced. While it would certainly take up more area and be a hassle to get it to work correctly, a CPU architecture which worked closer to a zvs tank circuit at the logic gate level could be less lossy than traditional architectures, at least where heavy processing loads are expected.
Second, I often wonder if vacuum tubes will make a return, almost specifically because of the challenges of using modern CPUs outside the shielding of Earth's ionosphere and atmosphere. Cosmic rays corrupt RAM and cache memory like no one's business. Without mosfet and related technologies, I have wondered if we still would not have seen development of microscopic arrays of vacuum tubes and analog computers similar to today - obviously different in specs and capabilities, but if we were to miniaturize vacuum tubes or related devices to produce miniaturized devices, how far could we go?
Power electronics would be the biggest area where we'd be at a loss. I am not sure something like a variable frequency drive could be realized without some weird motor-to-generator setup. Though at that point, exotic power vacuum tubes would be just as reasonable to invest in.
Though, to be fair, part of me is just biased. A friend of mine owns an arcade and I fully intend to redevelop CRT manufacture simply because nothing quite matches the phosphor glow of a real CRT. I figure I may as well invest in vacuum tube technologies while I am at it, and see what can be done with modern materials and micro/nano fabrication.
As for materials...
Silicon carbide has been known about for some time and it seems more useful for power components than processors...
Gallium arsenic has been used in microwave fpgas for some time and gallium nitride seems to be generally superior substitute... But there must be some reason it hasn't been instantly swept into CPU manufacture, yet, aside from just cost.
Maybe germanium, but again, if that isn't already in use then there's a reason for it.
Titanium nitride had some interesting band gap properties and has already been in use for certain layers...
Diamond substrate has been known about and played with in laboratories for some time, and it would be odd if all the various studies into deposition processes hadn't resulted in reliable diamond layers on the commercial scale by now.
Graphene has been a candidate and has some fun properties that make it more ideal than diamond, but doping of a "1 dimensional" lattice is a bit of an adventure in fundamental forces...
Some kind of magic insulator that increases its band gap in response to a difference in voltage would be amusing...
I was horribly disappointed in elementary school when the science book had a picture of a robot and we did build, study, or learn anything about said robot in the class that year. Someone says "new thing" and my mind runs away with it...
I'll be quiet and listen, now.
Electrons aren't the only particles that exhibit wave-particle duality behavior. The whole underlying idea behind wave-particle duality is that all subatomic entities exhibit properties of both models while not fitting exclusively neatly into either, therefore we need a way of reconciling the two. At its core, the issue is that subatomic entities aren't really "waves" or "particles" but instead some sort of variably quantized areas of space that we do not fully understand.
As we approach the quantum limit, I would expect to see some constructive or destructive Inference between the electrons in these TDMs.
It will be nice if you could please cover a topic on AI and Quantum computing converging
When I started transistors had emitter base and collector. Lol. Replaced individual defective ones.... lol
"Current" situation! Hah! I see what you did there.
All this stuff is good news, however a time line associated with the development, of this tech, would be good. I will be watching for an update video.
Given the potential efficiency of tunneling transistors, what do you see as the biggest challenge in scaling this technology for widespread use?
I subscribed good stuff !!! ❤❤❤❤