The World After Silicon - From Vacuum Tubes to QUANTUM

Free 10 pcs PCB for New Members: pcbway.com
Support me on Patreon: / coreteks
Visit: coreteks.tech/
Buy a mug or a t-shirt: teespring.com/...
Follow me on Twitter: / coreteks
Follow me on Instagram: / hellocoreteks
Disclaimer: The technology timeline in this video is based on IEEE.org's predictions and roadmap, on information from TSMC and other fabs, and from other sources. It's by no means "gospel".
Hyperdimensional Computing - Neuromorphic:
www.rctn.org/vs...
Videos from the AMD official KZbin channel were used for illustrative purpose, all copyrights belong to the respective owners, used here under Fair Use.
Videos from the Intel official KZbin channel were used for illustrative purpose, all copyrights belong to the respective owners, used here under Fair Use.
Videos from the Samsung official KZbin channel were used for illustrative purpose, all copyrights belong to the respective owners, used here under Fair Use.
A few seconds from several other sources on youtube (*including other youtubers*) are used with a transformative nature, for educational and illustrative purposes. If you haven't been credited please CONTACT ME directly and I will credit your work. Thanks!!
#Transistors #Moore'sLaw #Quantum

[@hyperpug2898]

I hope this video will be still up until 2050. It'll be fun to see how much you predicted right

[@skel2333]

It'll be funnier to see "outdated" tech, that he is predicting today for the coming years.

[@satellite964]

I don't think that is very likely, some natural or man made disaster will hit Google servers and all will be lost.

[@XxWillsFirexX]

You could also download the video

[@HarbingerOfTears]

satellite964 You do realize that Google has multiple servers around the globe, each of which have copies of their entire video library, right?

[@satellite964]

@@HarbingerOfTears yes, am a computer scientist. Govt/military/PMCs could destroy those if they really wanted to.

[@MarceloTezza]

The Transistoorrr...

[@phillgizmo8934]

Thank you for reminding me. I forgot that I was going to try that game. Oh, you've meant something else, haven't you?

[@itsJoshW]

Heard this. Then immediately wanted to click 'close'.

[@gameresearch9535]

Have you seen the Australian company named Archer Exploration and how they plan to use Graphene to bring Quantum Computers to room temperature? The first scale chip has been completed already, progress is accelerating so fast for emerging technology and beyond. This means that we could see Quantum Computers in our homes much sooner than you think! With Graphene being used. Some companies are using Photonic - Quantum chips. Some universities are using Spintronics with Graphene to make our devices like cell phones, classical computers (that we use now), and cars.. more energy efficient while making them faster, in the next 4 years. Spintronics is basically electronics 2.0, when it tilts the Electron = electricity. Also Spintronics is being used to tilt the Photons (light), which creates Quantum Teleportation, and Australia plans to make their Quantum Internet from this emerging technology by 2030, and will share it with Europe, no word on if America will have it yet. We need to get rid of Silicon, and go with Graphene transistors, they can actually do it now, I think AMD and Intel are just being stubborn, it's been proven that Graphene could make our classical computers 1,000 times faster, there are videos on how this works, we just need a new OS, and just have to wait for Microsoft to make one for such a computer. The Matrix 4 movie is on the way.. I know what you're thinking.. but imagine instead of Bits / Binary Code of 1's or 0's, both 1's and 0's with Qubits "Quantum Bits" with Quantum Computers.. scrolling down the screen in the first Matrix movie. Because Bits are how our computers think to see images and things on our screen like an apple as an example, but instead.. imagine it being Qubits scrolling down the screen in the movie. Imagine Photonic - Quantum chips using also Neuromorphic Computing emerging technology combined, just for those machines in the Matrix that attacked them outside the Matrix, and imagine them using all 3 of those techs with Spintronics and Graphene material mixed. With a Quantum Internet for the Matrix connectivity of each mind, at Photonic Computing speed. Imagine the medical field, financial field, predicting weather (supposedly).. before it happens, dinosaur DNA acceleration progress, space exploration and path finding for warp drives on a star map done very quickly and during warp to avoid objects at warp speed, and also imagine fusion energy progress "private companies are using A.i. and some plan to use Quantum Computers", imagine all this with Quantum Computers. And new emerging and beyond technologies that we don't know yet, that can be made from Quantum Computers. And when we find new materials in asteroids with Quantum Computers, well.. imagine how Graphene is a wonder material of the future, and it will help with new emerging technologies and will be in all our devices, it will also change our world, so imagine new materials found in space could change it even more after Graphene, or at the same time, there's already asteroid mining on the way, it's becoming a reality very soon and there's a space race for it, for value, how it would change the infrastructure with new materials and for our devices, even machines / cars, and it might even lower the cost of our computer parts, and many other devices / machines, structures and so on, as materials become less rare, from space asteroid mining. Btw when you mentioned Graphene and lasers like in that video, I watched that video a few months ago and thought the same thing! I hope that is used with Photonic Computing! And our internet! They also mentioned that it could be for space exploration! Want to find out more information on this? Check out all the playlists I made. ----------------------------------------------- I was giving out helpful links, but it won't allow me to do that now, so I made playlists. 1. Check my channel, find a subscribed channel called Technology Research, go to the playlists there, and click "created playlists", that should show them all. 2. After that, click on the title / text of each playlist, not on the pictures. 3. Don't forget to click the "more" button in each playlist description for more articles and playlists.

[@nowonmetube]

Kettenunterbrechooorrr

[@Broockle]

FULLBRIDGE RECTIFIER!!!

[@PunishedV]

Can't wait to play Battlefield WW4 on my BrainStation™

[@quecksilber457]

Oh you talk about Battlefield WW4 - The Climate Wars? It will be developed by EA - Extraterrestrial Aliens.

[@quecksilber457]

@@ultraranger ? :)

[@ryankl1984]

As long as that brainstation also gives brain

[@Solid_Snake99]

Can’t wait to play Fortnite 5: War of the universe on my quantum pc

[@jakdaxter6033]

Pft. Try getting on my level you pleb.#HyperdimensionalPCmasterrace

[@CNC-Time-Lapse]

One of the most intelligent channels on KZbin. Always enjoy and watch your videos. Thanks for publishing such high-quality content. Seriously, your videos would be great as a Cable sponsored TV show.

[@poot111111]

10:27 Didn't expected him to say "THICC BOI"

[@3800S1]

XFX GPU anyone?

[@ratmdex]

I subbed after that

[@Falconite]

I thought the same thing hahaha

[@asiftalpur3758]

@@3800S1 it sucked ass tho

[@scarletsletter4466]

This is a great video. Also your use of b-roll from what appears to be advertising & promo of Intel, Samsung & other companies is very well done. In biotech we basically need implantable devices with piezoelectric nanogenerators which will likely operate based on the involuntary muscular contractions in the body, such as a beating heart or blood vessels. For example, this will mean patients don't need surgery to replace a pacemaker battery. Right now we see zinc oxide being researched (since it can be vaporized & made into very thin nanowires) but other materials are being used too. It's truly a fascinating time to be a scientist

[@lordswaggity1213]

YAYAYA NEW CORETEKS VIDEO! Was waiting for the upload! :)

[@Fredjoe5]

This is probably one of the best videos around with regards to tying together the list of various technologies proposed and putting timelines on them.

[@jurgentreep]

Hey, I absolutely love your content. You've got a great voice.

[@Embattled5211]

Personally, I prefer a little more enthusiasm in voices, like Austin McConnell. Great information and visuals, nonetheless.

[@UnrealZii]

I agree! Very soothing. Please continue to make videos man!

[@kalioude]

ZZZZZZZZ

[@l-cornelius-dol]

Way too monotonic and fixed-cadence for me. Content is excellent, but the narration is hard to listen to.

[@turkepic3637]

I don't know about other people , but the narrative is great in my opinion.

[@ichibrosan]

I have been around since the birth of microprocessors. I was involved with the ARPA network when there were only a dozen hosts. As a 15-year old, I was exposed to virtual memory and network sockets. That was around 1970. At that time, I had a burning desire to have my own computer someday, although that was a dream at the time. I didn't foresee that the ARPA network would actually turn into the ubiquitous Internet and that average humans would walk around with mobile IP numbers. I am really enjoying your videos and I signed up on Patreon so you can eat, and continue helping us imagine what is coming. In the past, I have had some success considering what was next and been ready for it, but the future from this point on is potentially unlimited, now that big players are pushing forward the boundaries. This has been an amazing time to live and see us going from tubes to T05 cans to SMT and beyond. Please tell us more Sir.

[@illusjon87]

Absolutely stellar video. You got me excited for the future with this one!

[@uzefulvideos3440]

17:54 The electron drift speed is irrelevant. The velocity of the electric impulse is what matters. And that is extremely close to the speed of light in vacuum as electric fields spread with lightspeed (c). But the thing is, light can be significantly slower than c, depending on the medium. The refractive index of glass is about 1.5, so light will only travel with about 2/3 c through a glass fiber while the copper cable wins with about 50% higher signal speed.

[@Basement-Science]

You´re mostly right, but electric and magnetic fields are also slowed down in any material. All electromagnetic waves are. Basically copper is just not conductive enough at these frequencies and small sizes. If room temperature superconductors were discovered, they may be a great solution to this problem.

[@uzefulvideos3440]

​@@Basement-Science Hm, on Wikipedia it says "In electrical cables, the velocity factor mainly depends on the insulating material": en.wikipedia.org/wiki/Velocity_factor , which seems rather odd to me.

[@uzefulvideos3440]

en.wikipedia.org/wiki/Relative_permittivity#Metals .

[@freyja5800]

The relevant speed for the information transfer is the group velocity, and it is highly dependant on the (base) frequency of the signal. for copper wires, the group velocity is on the order of metres per milisecond, 5 orders of magnitude below the speed of light (and decreasing with higher frequencies). Thus, a significant improvement by using optical wires is possible, despite them also being significantly below the speed of light in a vacuum.

[@willdavis677]

en.m.wikipedia.org/wiki/Velocity_factor The speed of electric field propagation through a copper wire is similar to that of a fibre optic cable. The electric field would propagate faster in a hollow core fibre due to the lack of a medium in the core. But the main increase in data transfer speed for an optical fiber over a copper cable is the increased bandwidth, not the electric field moving faster. Edit: If I remember correctly the reason you can fit more data down the optical fiber is because the pulses can be made narrower (due to the higher frequency range, wider frequency range = narrower pulse, from Fourier transforms) and because the dispersion is low in a monomodal fibre at the telecommunications wavelength in the IR (something like 1550nm if I remember correctly), low dispersion means the pulses spread out slowly as they propagate, allowing you to put the pulses close together without them overlapping. So narrower pulses and closer together pulses = more pulses that you can send down the fibre in a given time interval compared to a copper wire.

[@liju40]

The biggest innovation you failed to mention in the video is the move from binary to ternary logic. Photonics is that we will be able to use to switch from binary to ternary cpu's by controlling the polarization of the light. For those who don't know; Ternary is base 3 instead of base 2, and is either balance {-1,0,1} or unbalanced {0,1,2} and is much more efficient than binary. The Setun computer built in the U.S.S.R. used balanced ternary being cheaper and more power efficient than computationally equivalent binary computers. This change would give a massive increase to computing power and converting old 64 bit x86 binary instructions to ternary would not be hard. Building new ternary assemblers would be worth the effort as well. I think this is where the biggest improvement of computing in the near future will come from.

[@user-zu1ix3yq2w]

Tell us more; I've been very interested in photonics for awhile now.

[@willdavis677]

Well 'photonics' is quite broad and includes non-linear optics like continuum generators, as well as linear and non-linear fibre optics, aswell as using light in computing and quantum computing and I'm sure many other areas.

[@KyBr2000]

Mmm. 64 bits = ~18,4 million millions. 64 tits (?) = ~3400 million million millions. Quite a bit more space to breathe using ternary logic.

[@simple-cut-progress2778]

The big question would be how long it would take to get the density low enough so that ternary chip could compete with the few nanometer transistor chips we have right now. If the technology exists, it would be very hard to get the cost of production low enough so that it would make any sense.

[@yashaswikulshreshtha1588]

@@willdavis677 I wanna know how would the code look like? How would we assign variables and stuff, what's gon happen to pointers and syntax?

[@horstmeier2664]

About the quantum computer: There are actually very few algorithms profiting from quantum computing. Think about the required increase in logic expertise to go from single threaded to multi threaded. The jump from binary logic to quantum logic is orders of magnitude more difficult to do. Quantum computers will be made and they will be amazing at specialized tasks. But it will take many decades (probably more then 5) until it will replace binary logic machines in consumer applications.

[@MrMattumbo]

@KRYMauL Give them acid before the lesson maybe

[@MrGarkin]

Quantum is a pure milittary technology ATM, since only known practical aplications are counter-cryptographic. Even Nuclear Tech at least had a power plants.

[@bluellownogreen3331]

You are grossly overlooking A.I.'s contribution to developing quantum computers.

[@bluellownogreen3331]

Also , "ok boomer" me now, but you're making a bad utopian assumption that the Capitalists and Government regulators EVER!EVER!EVER! want the populace to have ownership of such a powerful device as a Quantum Computer.

[@deathhog]

@@bluellownogreen3331 That implies the government has any real choice in the matter. The only thing that's going to keep QPU out of consumer hands is the sheer impracticality of owning one. Unless we have a major breakthrough and make QPUs that work without needing to be cooled to bear absolute zero, it's just never going to be practical for even small companies, let alone individuals, to own one.

[@Siuolplex]

Its a Coreteks. This is going to be great. Man, you make great content. Keep on keeping on for the content. Your content is unique. I dont have anyone else to compare you to. Cause your that good. Well, I better get watching the video. Edit: I watched the Video yesterday. It was really good. Like always of course. I hope these predictions will be right or close. Then again Coreteks has very close predictions. Oh by the way. To quote a certain Tech News host. "CARBON NANOTUBES!"

[@siamsurf]

what's this "that good" that you say he has? Or did you mean to write you're by any chance?

[@Siuolplex]

It's a rare hardware decoder that allows good content to be made easily.

[@thenextweek2416]

Oh man, that music in the beginning plus your "I'm about to blow your freaking mind." somber voice made this video. Thank you!

[@IamCrusaderRUS]

"...the only real limit is the size of atom..." Doesn't quantum tunneling become a major problem much earlier than that?

[@mohhaf9384]

i think photonics would solve that

[@skyhop]

It's already a problem with current processors. I'm not sure how exactly they deal with it.

[@MaxVandenbussche]

Yeah sure. But it's a good example to set a absolute minimum to make a point. Even if that minimum is not viable.

[@joesterling4299]

It's a problem with current designs. Radically new designs are needed to keep shrinking further. He covered that, with the example of a new transistor design replacing FinFET.

[@darshan5726]

@@joesterling4299 currently finfet process is the reason of tunneling effect. If intel brings something new lothography tech then we can solve tunneling effect.

[@martir.7653]

At 17:55, Coreteks says that "light travels much faster than electrons". That's true technically, but the speed of individual *electrons* is irrelevant. What matters (for latency) is signal velocity, which is more about the speed of the electromagnetic wave. The signal velocity in both eletric cables and fiber-optic cables varies depending on the exact material, but they are quite close, within tens of percents at the most. Nowhere near the "hundreds of times" claimed here. Of course this only affects latency, not throughput. But then, the speed of photons/electrons has nothing to do with throughput.

[@user-zu1ix3yq2w]

i don't think computers compute anywhere close to as fast as their "latency" though

[@Fredjoe5]

Right, the speed of the carrier isn't really the big advantage. The bigger gain from photonic interconnects will be in the data density that can be sent at any given time.

[@AntonioNoack]

@@user-zu1ix3yq2w 30cm / ns is the light speed, 6 cm at 5 GHz. We are not far off.

[@stachowi]

HOLY SHIT this is awesome! Your content is TOP NOTCH!

[@TekExplorer]

*cough* minecraft *cough*

[@ayaanharyani4516]

I hope I see a transition phase in my lifetime just like we did from vacuum tubes to transistors. My dad used to told me how they had vacuum tubes back then and tvs would use to heat up and would take a minute to turn on and then solid state tvs came to market.

[@glenfoxh]

I would think these Quantum processors would be very useful for Physics processing in games. It would be very useful to have something that could calculate a mass of collisions over a mass of objects in parallel.

[@user-bj3pq2si2l]

This channel is probably the best tech channel i have ever seen.

[@rickandelon9374]

Your voice is soothing!! Didn’t mind watching in late 12 o’clock hours!

[@selenascott7278]

That opening had me thinking it was documentary straight out of the 80's but glad I stuck around because I just learned a lot.

[@agdgdgwngo]

Your voice is very... grand? I feel more like I'm listening to some fantasy audiobook or Warhammer lore video or something. Love it though, suits the video and of course good insights

[@turkepic3637]

*Voice of God revealed by KZbin channel "Coreteks'*

[@jonashallbook6312]

Warhammer lore videos

[@dougsteel7414]

If processes froze today, in 2 years we'd still see a 200% speed increase due to a shift in programming paradigms founded in the idea that this is all we've got to work with. Memory and execution resources used by developers partly expands independent of function and relative to what's available.

[@SianaGearz]

What the hell with this "photons travel faster than electrons" nonsense. It's technically true... but it's also largely irrelevant, since to receive electrical signal at the end of the interconnect, you're not waiting for the electron that you sent to come out, no. You're waiting for a wave of electron movement to come back out, and it does so very nearly at light speed.

[@ne1cup]

I think photons travel a lot faster than electrons. both systems use modulation of the current flow to send "data". we can not look at single electrons or photons , unless you are looking for entanglement of 2 atoms?( which is faster than light). The amount of data that can be sent in an optical fiber is 100 times more than can be sent over a copper wire..

[@SianaGearz]

@@ne1cup If you had a kilometer of interconnect, you get higher throughput out of fibre optic because it's low-loss and not susceptible to electromagnetic interference - optical interference is easy to shield. On the length of 3 or 20cm it really doesn't matter, you don't get more performance out of fibre optic than out of copper. Once you start to introduce optical semiconductors, i predict you'll have the same issue of leakiness that we had with NMOS circuits back in the day that limit signal fidelity. The reason we don't have them right now is i believe that the problem hasn't been solved, that there is no CMOS equivalent. Automotive industry also uses a lot of fibre optic interconnects for secondary systems, but not because of throughput, but because a car is an electromagnetically hostile environment due to alternator, and because plastic fibre that they use weighs much less than STP Ethernet wiring that would otherwise be required.

[@DoominikD]

its the difference between phase and group speed of wavepackeges, u can go over lightspeed with group-velocety but not with phase -vel...

[@luissosa4u]

In fiber optic conductor is lest loss, less interference, and allow to carry maximum frequency spectrum; more data signal per conductor. Once a teacher told me that the hole city communications can fit in a quarter coin diameter.

[@xxxlastmigxxx]

dude i think what he means is the heat ...photon don't escape and became heat ...i think is about energy efficiency..

[@UnrealZii]

I really love your videos. You're one of the few tech KZbinrs who actually KNOWS exactly what they're talking about. No Dunning-Kruger effect here.

[@lordswaggity1213]

Whats that smarty pants?

[@blue_bear_music]

7:10 it's not linear, the Y axis is log scale, so it's exponential. Same thing for 8:14 Great video, but the last 10 minutes are super speculative :)

[@Coreteks]

Yes, sorry, long hours recording and editing. The point stands though.

[@blue_bear_music]

@@Coreteks Yeah I realize it's a lot easier to critique than to make the video. You're doing great work, I learned a lot from you.

[@feynstein1004]

What a wonderful comment thread 😊

[@dmarksvr]

@@feynstein1004 Guy politely points out mistake, creator kindly acknowledges mistake, Guy acknowledges the difficulty of content creation and complements creator.... This is how the internet should work. A simple/kind exchange of thought, info and opinion... We all win.

[@feynstein1004]

@GuyWithFox Seconded :)

[@Tanakas7]

This video deserves millions of views. Great job on the video!

[@NorwayNico]

Your voice is deeper then the Mariana trench

[@ThomasLee123]

Simply amazing. Since I am one of those who actually saw us move from tubes to semiconductor you can imagine how shocked at this path of growth.

[@spacet1me]

This dude is like the Kurzgesagt of technology.

[@Legion849]

That's true because he talks about things in a manner that is different but exciting.

[@thepartsrunner5758]

.....good ole, Kurzgesagt. A dandy chap indeed.

[@KarlGrabowski]

Exciting stuff! I believe you're on track. Then when you take into consideration the upcoming advances in power storage, Super-conductivity, and development of A.I.; I'm imagining huge leaps in technologies across the board all within my lifetime, and I'm in my early 50s.

[@marcosiedentopf8203]

awesome, thanks for the upload

[@eddiegerwer01]

Now I have to make nano-sized vacuum tubes a must to mess up his predictions!!!

[@scunnerdarkly4929]

Yay my monthly fix of immaculately presented tech analysis is finally here! 😀

[@Anenome5]

New Coretek video... DROP EVERYTHING to watch.

[@theknave4415]

I'm so old... I learned everything about vacuum tubes. They switched to diodes. I learned everything about diodes. They switched to printed circuits. I learned about printed circuits. They switched to silicon chips. That's when I said to hell with it. ;)

[@omnianti0]

maybe its time to make aplication of all your knowelge

[@garmarrod]

Bainsworth it’s all vanity. That’s the lesson.

[@pierrewillful]

Well hey you're still way ahead of the curve

[@deathhog]

The life lesson is that you shouldn't feel bad that everything you know is going to be obsolete. The kids after you will go through the same thing. And so will their kids.

[@tek5670]

@Bainsworth don’t get too complacent. Things change. Change is inevitable.

[@theodor12]

Holy shit! The future is so exciting! Can't wait to see the evolution of technology. Just the idea about carbon nanotube CPUs got me excited, let alone the quantum computer stuff :D

[@payman.amini1]

Live the moment, live now and the future will come as now. So I mean you can see the future when it's "now". Enjoy the life, enjoy the "now"

[@vladimirmoushkov6137]

Coreteks - the place at KZbin where listening for geeky technologies is like watching latest James Bond movie. Thanks for existing!

[@dizzymetrics]

>2050 >Intel finally has good 10nm yields

[@pergunnarvikmjlhus3597]

I won't get to see the day when Intel finally start producing 10 nm chips, but my grandchildren will!

[@badnewsbruner]

You got a new sub buddy! Well deserved! Love your narrating voice, and this subject fascinates me, and excites me for the future!! Thankfully I'm only 33, so (if I live long enough) I'll be able to witness all of this epic tech and enjoy it!! I hope we get tech implants soon, not the scary 1984 ones, but the cool 2077 ones xD Also, I have an EL34 vacuum tube tattoo, thought you might enjoy that ;)

[@westernautocrat]

thank you coreteks very cool

[@ncoalhw]

This was an enjoyable video. Nice voice, very relaxing. Looking forward to more content.

[@fedor3000]

Just read it yesterday. How tf can you be so quick with this video! 😳

[@volfactory]

This is the best fundamental computing technologies channel on youtube.

[@zarkaztick8973]

Moore's Law. Gordon Moore. Co-founder of INTegrated ELectronics. 90 years old. Imagine all he has seen. All he won't see.

[@piotrd.4850]

And when asked in interview is there anything he wished to be remember for, is "whatever it is ... just NOT the Moore's law".

[@Dallen9]

I see Quantum Computers as the key to Absolute control of Tractor beams, ion propulsion, Teleportation, Phaser and Ion blasters, and Holodecks. The ability to be in a room that is no bigger than 5ft. by 10ft. but be able to explore the streets of New York City from Grand Central Station without running into a wall in the room but also allow another person to enter the room and explore different areas, will be paramount.

[@muuubiee]

"what comes after 1nm"... 500 Picometers? Also I think you'd need at least 3 or 4 atoms for a transistor.

[@ATestamentToTech]

By far the most interesting video you have done to date.. I have watched your vids for years and this is just fantastic

[@mandelbro777]

>"no silicon future" ...damn I feel old hearing that. >"back in my day, the silicon revolution seemed as if it could solve all the world's problems" >*blank looks from GenZ3*

[@pri7920]

YOUR VOICE IS GREAT. AT LEAST 4 MILLION SUBSCRIBERS FOR IT PLUS 6 MILLION FOR CONTENT. SO I THINK YOU DESERVE AT LEAST 10 MILLION SUBS SOON IN NEXT 1-2 YEARS.

[@khoakirokun217]

22:22 “quantum’s supremacy can be archived within a few year”. ... may be ... 2 months from the time the video uploaded.

[@markgreene8600]

The transistor size (e.g. 7nm) is not just a marketing term. It's not referring to the overall size of the transistor, but to the gate length of the transistor. The gate length is how far across the channel is underneath the gate, which is where the field effect is occurring. This is very important parameter because peak transistor current and gate capacitance are highly dependent on this. As gate length goes down, capacitance goes down (less area for charge to build up) and current goes up (smaller gap to jump across). Smaller capacitance means less energy consumed per transistor switch (less power consumption), and more current meant you could more quickly charge and discharge the capacitor (higher frequency). Even with thermal limitations, having low switching energy and high switching frequency are huge drivers in performance. That is why gate length has been the holy grail for so long, and why companies tout it. This is obviously just one piece of the puzzle, but it is a huge piece.

[@tkimaginestudio]

Great video, but you are way too optimistic about quantum computing. I also disagree that quantum computing is about massive amounts of data and parallel computing. It is about tackling combinatorial problems with exponential complexity growth. Quantum computers can entertain multiple solution possibilities at the same time, hence they are vastly more suited to such problems. Note that current cryptography technology would become unsafe when quantum computers start working. A working quantum computer would cause a massive disruption.

[@markgresty1633]

agreed ! and a hybrid quantum/silicon passing all data thru a silicon cpu to solve that little problem will bottleneck or nullify the advantages of the quantum componet....... the only solution is to transfer all computing to quantum systems encoding all cryptography in some sort of algorithmic state matter translation matrix LOL idk

[@idesofmarchUNIAEA]

16:00 Gallium arsenide chips were supposedly impervious to EMP blast. I remember that in the early 80s.

[@akramdudin943]

Informative just perfect

[@sam3kperv]

I think this is the most informative and interesting video I've seen on tech, and the advancement on it in a long while, you got my SUB!!

[@donnerschlag41]

Cant wait for coretek to discuss that 1nm processor in 2030. 😀

[@kristiyangerasimov6708]

loved the voice pace of the video and visuals. Subscribed. ++ your sponsorship is different which is a nice chance from square space and Brilliant

[@Etheoma]

photonic based "transistors" will never happen on a large scale, wanna know why? The shortest wavelength of non ionising EM radiation is ~200nm, which means you cannot make any part of your transistor smaller than that, photonic communication on die however is double and almost certainly going to happen, at a bare minimum I would expect at least on package photonic communication. As far as I am aware silicon is supposed to top out at around 10Ghz while photonic transistors could top out at 60Ghz, so there just isn't enough of a performance advantage to moving to photonic "transistors" to offset how big they would have to be, and even at 200nm wavelengths you are talking about ultraviolet which is rather high energy, so the power savings likely wouldn't be worth it either. And yes for communication the speed of light is great and all, but it wouldn't make much difference in something as small as a transistor, for cross die communication it would make a difference though. Also pleas tell me you know why we can't use ionising radiation in a transistor... Also why I said it would never happen on a large scale and didn't say it would never happen is because there likely is some application where it would make sense, like for a computer that needs to operate in strong magnetic fields for example, also if the wave length of light the chip will respond to is very specific it could be used in environments where there is a lot of ionising radiation, like space. Because if the "transistors" are large they will be less damaged by ionising radiation, but you could get similar-ish performance for the same power consumption.

[@robrod7120]

Etheoma So basically if I want a supercomputer chip I have to use extreme ionizing radiation to transfer data? I’m sure government supercomputers could benefit, they have the money for lead lined computer rooms

[@Etheoma]

@@robrod7120 ... You didn't read what I said did you... I said for communication it's fine as you can make the communication as you don't have to worry about the size, at least not on the scales we are talking about. But for a "photonic transistor" size matters as we already have whole transistors smaller than 200nm, where as the smallest part of a "photonic transistor" would have to be 200nm...

[@TheRealLink]

Fantastic overview of what's to come.

[@tempname8263]

Quantum computers only are superior at solving exponential problems, and not anything else.

[@paulwest3905]

Thanks for giving me a shot of tech adrenaline- my hunch is you are dead-on re: our shortsightedness or "can't see the future tech forest for the trees" mentality. Great stuff!

[@kama2106]

6:37 We cant make single atom transistors because of quantum tunelling

[@FireTome]

Actually, it should be: 'We currently have no idea if making a single atom transistor is possible, and we currently do not posses the knowledge to make one, because we don't have a complete enough understanding of quantum tunneling.' If we understood quantum tunneling better, we might actually be able to prevent it from happening, and thus we then might be able to make one. I'm not saying that this will definitely happen, because I don't know if we'll ever posses the knowledge AND have the drive to make single atom processors, but we certainly don't have to rule out the possibilty. Remember: People used to think that atoms were the smallest 'building blocks' in existence, but since then, we have found there to be stuff much smaller (like protons, neutrons, and also: Quarks).

[@filipborch-solem1354]

Thank you so much, Coreteks. Another incredible video.

[@pneumonoultramicroscopicsi4065]

I wish that I was born in the future

[@gamma_ray20xx53]

U are

[@PeteTheL337]

Coreteks you are a champion of the nerd and tech enthusiast. Very interesting video yet again and love that i always learn something new from your videos spite being an enthusiast for 25 years now. Extremely good stuff.

[@jzilla1234]

Wait.... An arab invent mosfets? I also like the sponsor.

[@twiinpk3r]

jzilla1234 arabs have invented a lot of things

[@N0N0111]

And the best is yet to come. I start hearing that more and more.

[@Mastakilla91]

Somehow I imageine that Coreteks is bald. I don't know why, it just fits his voice imho. Tell me that you''re bald

[@adrianene5098]

I love your channel for insight and very well done technical content. I'd love to see much more, I feel that you know much more and I'd love to see much more! Keep up the good technical coverage! I hope to see more about the multithreading of software done at hardware level about which you gave a small tease in a previous video.

[@Coreteks]

@adrian ene I've asked them for an update a few days ago actually, they haven't got back yet. It's definitely something I want to revisit.

[@jacobbailey5498]

“Quantum supremacy may actually be achieved in 2 years”... gets achieved 2 months later

[@noietzpk5169]

Not exactly, we can't prove it achieved quantum supremacy and the problem it solved was designed for it to solve

[@ptrooms]

Fabulous graphics which inspire visions....... one day this video might become the classic of how it all started,.

[@veedrac]

QUANTUM DOESN'T WORK LIKE THAT

[@lordswaggity1213]

Your mom doesnt work like that

[@PlanetFrosty]

We currently working on a hybrid photon/electron architecture for creating “inFlight” processing over fiber on massive wafer sized processor and optical layers.

[@tamarockstar45]

If you use cannibis, definitely do that before watching this video.

[@mack7207]

you talking about the possibilities of quantum computing gave me goosebumps...

[@altimmons]

Quantum is totally different. It only does one type of thing (hypothetically) well. Your example is not accurate

[@mishpult]

super cool channel, already subbed. One advice - try running deesser on your voice while you record, it'll be a quantum leap in the quality of the audio, since it's a big part of your production. Think about some light background music as well - it will just give another dimension to your production and increase the quality overall. Cheers!

[@JordiFerran]

The quantum section of the video is too speculative and wishful thinking. The quantum is for now just a buzzword, as is being able to control information field [theory] through mechanical means. Does not take into account the consciousnesses that the observer does not perceive. Maybe there is no quantum chaos. It is just a kind of order, that inferior minds are unable to process. The nice idea is the combination with biology. Fiction series Star Trek Voyager show once: the ship gets “sick”. There are bags of neurons for interconnecting ¿the wiring? Inorganic switches are replaced with organic neurons. Maybe because they offer a faster response time, and do not get as hot. That idea is fantastic. Cloning through biology, to place components alive, within an artificial computing system. Because nature is more efficient than any artificial machine. The idea missing in the video is: the programming of crystals like quartz. To encode as in programming, through layers in crystals. To manipulate the energy moving through and irradiate the surrounding area. To manipulate the information field and the “environment”. That sure get us closer to Atlanteans, than just playing with silicon. Just my two cents on this interesting topic.

[@billykotsos4642]

This guy is on another level. Heading over to patreon now ...

[@potatoslayer8139]

Great stuff as always Coreteks!

[@thomasb78]

Great video, quite interesting future to come.

[@HaitiSpaceAgency]

Love this type of video more than just industry analysis. You're awesome.

[@sacamentobob]

this is a well researched roadmap! great work coreteks!

[@pebre79]

Very well produced video. I can tell the hard work you put into it! I subbed. Cheers

[@Mecrom]

You are being very optimistic, I would guess everything will take at least double the amount of time

[@ZAR556]

Been waiting soo long For this kind of video

[@getyerspn]

Great video ...all this talk of wafer fabs and optical chip coupling makes me miss working at an HP research/ Production wafer fab ... worked there for many many years ... they moved production to Asia and I couldn't follow :( .... I loved using the electron microscopes to see what we'd actually made.

[@Zemtex47]

One of the best channels on youtube.

[@karlp8484]

You should change your channel name to Oracle, (as in Oracle of Delphi) I can't believe how much insider knowledge you have and ability to map out the future. Just subbed.

[@verficationaccount]

Coracleteks!

[@WestSeaSpirit]

This video was miraculous. Thanks man! I've learned so much and you've provided so much information.

[@MultiVigarista]

You are undoubtedly the best tech analyst on youtube!

[@dreamyrhodes]

The difference between Tubes -> Transistors and Transistors -> Q-Bits is, that transistors been build to replace tubes as they basically operate on the same purpose: To switch on/off electric circuits and amplify signals. Q-Bits on the other hand have a very specific purpose. Each circuit is designed for a single task that it then performs pretty well and on a scale of multiple times better than normal, general purpose circuits. That's more like ASIC vs general purpose circuits. ASICs are build for a very special purpose, like calculating SHA256 hashes. And they can do only that. An ASIC today calculates a few thousand more SHA256 in the same second than a CPU can do, but could never be programmed for let's say calculating a light ray in a 3D rendering or a mere checksum of a packet which both a CPU could do (although for the light ray slower than a specially designed GPU for that purpose but still technically possible). You'd need to design an own ASIC for that purpose which then can again do only this one thing, let's say calculate a ray in a raytracer. So what Q-Bits would replace are ASICs: Very special applications, but not CPUs that are made for very general applications. You can use a Q-Bit processor for calculating weather models or molecule configurations but not so well for downloading and playing music or processing a database operation as the first are very distinct operations that need a very special formula be calculated with thousand possible input parameters while the latter tasks contain of different steps of different operations be calculated sequentially.

[@perseverance8]

I'm imagining "quantum compute" modules will simply be built as it's own type of functional unit integrated within classical digital compute devices, it will likely be some time before this happens, but I can envision the development of new methods to build quantum compute modules utilizing certain quantum phenomenon currently observed in modern semiconductor materials.

[@jondonnelly3]

I feel asleep, thanks I needed it. Voice is so soothing.

[@ghoulbuster1]

Now this, is exciting! Marvelous video MR.Teks

[@PlanetFrosty]

Electrons, Photons and Radio Waves is where we are developing in hybrid as photonics and radio wave interaction create a new architecture we are now working on at this time.