@Game Over they reached an angreement where AMD licence x86 while intel licence x64 without paying any royalties also RISC-V is the future It has the specifics of arm but with an open standards so companies nutoriously push this intructionset
thats a good question, might seem rude but asking who he is in this kind of atmosphere it a very respectful thing because the questioner is wanting to know who they are
This would come in handy for prosthetic bionic limbs, hands, & feet. It would make it a lot easier to place flexible motherboards to lay within the layers of the structure itself.
You literally can, right now, with an ordinary inkjet printer. www.wired.co.uk/article/inkjet-print-an-electronic-circuit @@menotu000 The ink is really expensive, but not bad compared to normal printer ink - it's about $30 a gram.
You literally can, right now, with an ordinary inkjet printer. www.wired.co.uk/article/inkjet-print-an-electronic-circuit @@menotu000 The ink is really expensive, but not bad compared to normal printer ink - it's about $30 a gram.
@@nathangamble125 There's a difference between printing an electronic circuit and printing a full proccessor i.e. what Jason ^^ was referring to. Did you really think he was really waiting for the day he can print out a blink LED circuit?
I was going to help a friend out to fix his ECU in his French car. It was not a new car by any means. The ECU was made by Siemens or Bosch. The box looked really robust, aluminum, clam shell type. I opened it up and every component was soldered to this, folded plastic PCB. It looked like a 1990's calculator inside. I closed the box again and told him I simply lack the tools to fix it. I think it's cool that the IC's take the leap into platics. But I hope they do it in such away that DIY'ers don't need to buy expensive tools.
oh, i'm fully aware... it's painful since no one around me really is it's one reason why i'm trying to avoid becoming one of their under-valued worker bees
chip never heats itself ...the whole component heats altogether....if they provide it with a good cooling technology ...the estimated heat produced in 0.1 microsecond will be transferred immidiately and whould be completely safe.....everything depends on heat discharge rate
Arm processors use very little electricity and I doubt they'll have a high atdp but if they do have a heat problem in the future they could just use Liquid to transfer the heat away like a non conductive mineral oil
Th is is the most exciting thing in ICs I've seen since the 1970s when yields began scaling up and processing power began to enter the truly useful range for mankind. What they did for $10s of dollars, this company is doing for pennies. There is a whole new frontier here and they are at the horizon of total victory. Thank you for bringing us this interview. If you get the chance in the future, please continue to interview this team, and this company.
@@radomiami No, actually. CPUs are very cheap to manufacture thanks to economy of scale, and this has been true for decades. The AMD 9080 (a clone of Intel's 8080, and AMD's first CPU) only cost 50 cents to make in 1974. Most CPUs today are actually more expensive than they used to be to manufacture - usually tens of dollars - but they still charge hundreds (or thousands, for server CPUs) of dollars for them. It's worth it for consumers, but the CPU industry has very good profit margins. The reason we don't see more companies competing is because they need compatibility with common software in order for their CPUs to be worth anything.
Dang 2 years late but still a cool topic not only that imagine, true flexible phones ultra light weigth peripherals and devices normally more smaller than usual, this stuff is sick
Instead of making it smaller with slight improvements, keep the size, make it more powerful. First gen cell phones had bigger processors than desktop PCs. If we had batteries to keep up with our density of power for the size, the only limitation I see, it would be amazing. But if it's done, parallel research will make up for the deficit.
its based off of a cortex-m0 which is.... drumroll.... 2.33MHz. you could maybe run dos on it and do some assembly code? it's still super insane that you can actually create a CPU just by shaping up plastic though.
I agree with the ARM guy, this probably can't ever beat silicon in performance or possibly even performance per dollar, but silicon won't ever be as cheap or thin as this.
might have realised it's hard to implement these as they said they seem too easy to damage even by a slight mistake to me..... which would suck in the real world
@@fadrium1464 Heat is fair concern and one of the questions not answered in the original video - along with clock speed. But I've done plenty of MCU designs where power is only a few milliwatts. You can still get a lot done with that.
I think it's for stuff that's don't need much power Only some mhz :p I think you could run a bios on it but a os no only a really small one that fit on rom chip
@Murphy deffa yea right... super computers only use 32 cores, right? GPUs also have 32 cores? They have thousands. Investigate about "CUDA" (Nvidia) and "Stream Processors" (AMD). As Leicht Sinn says, it all depends on how you program. Also depends on the task you want to do. Granted, for most users even 32 cores are to many.
GPUs have 32-64 cores. Stream processors are execution units, not cores. Super computers and server halls do not generally accelerate one task, because that’s hard; but that’s what end users tend to care about. You’re not interested in running an ensemble of games, you want to run a single instance of the game faster. You don’t generally want to do more each frame (e.g. simulating the NPCs you don’t see) you want to make each frame faster. It’s a different use case. Also, Bill Gates/Microsoft never said 640k was enough for everybody. The 640k limit came from IBM hardware combined with the need for adresses that map to other things than physical memory.
@@soylentgreenb Yes, they have "few Compute Units". As I understand it, those compute units will send instructions to multiple small processors running in parallel. Nvidia calls those smalls processors "CUDA cores". One could argue they are cores inside a core. Scientists use GPUs as "low end cheap super computers" to execute simulations and other stuff, normal users for gaming and mining bitcoins. But ultimately they are processors running in parallel, and that is what we want. Games tend to use "few cores" most won't use more than 4, 4~8 cores in the newest ones, even though some are programmed to take advantage of more cores. I am sure we will soon have games that will require easily 8+ as a minimum. Applications that benefit from more and more cores are for work (graphic, science, engeneering, video editing, movie making, etc). Servers for virtualization. It seems you are thinking only about games. Outside of games and simple office use, the higher the core/processor count the better. However, I was wondering what is the life span for the processors in the video. Don't think a processor printed on a plastic will last long for a heavy task. Maybe they can improve (and I bet they will, if they haven't already) their technology to print on other substrates that will compete with or be better than a silicon wafer.
Wouldnt the processor be ruined if it generated enough heat to warp the plastic thou? It would seem to me like heat could really effect the lifespan of the cpus and with plastic being the base it would not take very much to break them. Maybe for low-end applications it might make sense from a cost-effective standpoint but anything more complicated than a stopwatch you may have problems with the cpus getting too hot and damaging themselves. Need something that can resist heat and/or cooling I guess.
These arm socs could be easily stacked together. And these are literally paper thin. Imagine stacking these socs up to the thickness of a phones battery? Probably be a good 60 layers. That would equal an immensely powerful smartphone.
I'm wondering about heat dissipation. Like if you were to attempt to scale this up to say... Risc-V with a certain number of cores running Linux, would you begin to run into problems with keeping it cooled? Or does the thinness allow heat to dissipate more easily? In terms of manufacture, with its more robust process, is this something you could then manufacture anywhere?
if it will be reduced to 2mm, why it needs to be flexible? and if it acceptable to make it rigit at the size of 2mm, do not we have much better density already?
My fear is when you crinkle the plastic, you make microscopic tears in the interconnects or the layers, and you IC dies! He says it's robust, but that's possibly only robust to gentle rolls of the plastic!!
Does Scott White ever blink? He looks like a crazy man, but in reality he's extremely bright and surrounds himself with the brightest in the industry. He's totally focused on his objectives.
Sounds great but how sturdy is it? How long will this work if put into clothes for example? Plastic connectors like this in laptops can break within a year with no movement at all and those processors look even thinner than those...
so basically this is just like making 2 way mirror material like space blankets, I love how he danced all around saying they use vacuum vapor deposition, and then acids to etch since the plastic is impervious to the majority of those acids, while say the aluminum coating on those thermal or "space" survival blankets is exposed on one side,
Not actually a problem. You must be thinking of plastic as thermal insulator, but at sufficiently low power, it isn't, it doesn't have a lot of thermal conductivity but it can have plenty. If you think of most integrated circuits, the IC die is sat a millimetre deep inside an epoxy enclosure (typical black chips) and some heat goes into the PCB where the copper fill dissipates the heat through yet another layer of epoxy namely the solder stop lacquer and the silkscreen, basically all thermal dissipation occurs in most typical low-performance circuits through copious layers of plastic. The materials they're dicing around here have similar thermal performance, i think something like polyimide even has a slightly better performance than epoxy. Another thing to consider is the Cortex M0 electronics itself. The typical ones i've seen have an internal 1.8V regulator and a bit of the power consumption comes from the regulator, a tiny bit from the computational core itself, but most really from I/O circuits which is a 3.3V signal converter, static protection circuits, pullups, draw induced by connected devices, so basically all of it is due to having to interface with integrated circuits which have existed now for decades, standard electronics. Now if you design this as a disposable circuit from the get go, you can dispense with all of that complexity, and with further integration you don't need to interface with a different I/O voltage and can also reduce the pullup current. So at the end of the day it depends on the switching performance and voltage thresholds of the underlying flexible MOSFET, the applicable supply voltage, whether they're able to make it work, they can work with their library of fully flexible/printable circuit elements to achieve a good leeway in power consumption. I'm almost more worried about how they want to go about power delivery for most use-cases rather than power dissipation. For RFID tag it's trivial enough, and it's a fundamentally low power circuit, orders of magnitude lower power than a processor, but other things? You can dissipate a lot more power under a fairly thermally unfortunate chair cushion than you can store energy chemically in a button cell. Given the power delivery constraint, i'm not sure the world necessarily needs a flexible ARM Cortex M0 processor for now, though in due time it may be promising as a cost cutting measure.
I only have 2 questions ,, is that sheet hes holding a working unit ? and secondly are these room temperature conducters since what I understand that's not a thing yet so how would they be cooled ???
I have no knowledge on whether it's a working processor, there hasn't been a powered-on demonstration after all. But what do you even mean by second question? It's not like you need to dunk your phone into liquid nitrogen for it to work, it's as if you're confusing semiconductors and superconductors. How would they be cooled? You can dissipate several dozens to hundreds of mW through the plastic sheet easily, no problem whatsoever, but a circuit that consumes that much would be impractical due to power delivery issues, so it stands to reason they're likely aiming for a sub-mW power circuits. In fact most simpler electronic circuits you handle in your daily life dissipate heat through a fairly unfortunate amount of plastic and stale air and other material arrangements with bad thermal conductivity and they're perfectly fine.
@@SianaGearz so then its basically just a shiny idea on a piece of plastic paper coming from 3 pocket dwellers out of some think tank at the risk corp im assuming !?! and also yes if you've ever had your phone apart and looked at the processors some have a metal shield glued on for heat spreading so they don't crack which splits the circuit and destroys it and others either expel heat straight to the board and through the faraday casing and also super cheap ones like kyoceras old pocket phones the whole board array heats up due to lack of proper adaptation ,,its no the battery you feeling when youre texting on a hot day and and you think holy toast my phone is hot its the processor cooking the whole phone inside out ,, as to my supposed confusion no im not confusing semiconductors , transistors heat to expelled traped energy it cant expel and if a super conducter is used all energy is either used or diverted thus being a "super" since it has no resistance ,, no resistance ,no heat.
@@jessemazo4902 They're not some basement dwellers, they have research, they have a production line and they're marketing a product - a relatively low-tech RF tag that is super cheap to make and has enough of an unencrypted digital storage to uniquely identify it. It's suitable for product tracking and not suitable for high-security use, but it's super cheap and causes less handling and production line automation issues than traditional tags that are made of paper and have a silicon IC glued onto them, and allegedly higher reliability. The processor shown is not a product and might or might not eventually be, but that's what they're working towards - they say they have a long way to go towards it being practical enough for anyone to actually buy one. It would be prudent to assume that they have validated that these structures that they're showing are actually being built correctly enough to work. There are some things they aren't saying though, they aren't disclosing the risks that may lead to it never becoming a product, and i have some ideas in that regard. On phones, depends, the processor shown won't run a smartphone! It's an ARM Cortex M0, the kind of processor you'll find in a wireless controller chip or in a high-end gaming mouse or keyboard doing just fairly easy interface duty. Existing Cortex M0 implementations have a sub-1mW core consumption and most of their consumption is in the IO circuit which can be eliminated if you don't need to interface with traditional existing ICs, but only with other flex-printed circuits. Smartphone's power envelope is around 300-900mW, so it does need some cooling considerations, but the metal shielding is an "RF can" to suppress interference, it's not for cooling. On more powerful phones you'll find a hole in the RF can and the processor heatsunk into the back of the screen or alloy middle enclosure directly. On less powerful ones, you'll often find atrocious "thermal compound" that doesn't deserve the name or worse, a bubble of stale air, one of the most vicious thermal insulators short of vacuum - but of course they still have cooling via the PCB, through copious layers of plastic but at least with some area to go with it. Which is to say these guys here also can hit the cooling targets of vaguely similar order of magnitude, but it just wouldn't make for a good product if they do. I think ultimately power delivery and power efficiency are going to be the issues that limit their products or make them unmarketable, not heat dissipation. Heat dissipation is an issue they'd LOVE to have, because if they were in position to have that, it would mean they solved the power delivery issue at least.
i would hate having this dude put a camera 2mm from my face and cut me off while im trying to explain my product.
irritating voice as well
exactly what i tought, seems like a very egocentric and arrogant person :/
I like the camera movements, it makes the video more dynamic.
i like your positivity 😆 *+mcdoublemaster2*
yeah this guy SUCKS
can't wait for my rug to understand its environment ...
It massages your feet when you walk over it by using a large pressure sensor
@@INeedAttentionEXE It also cleans itself by using it's threads to push garbage under the couch.
@@legacyoftheancientsC64c Can imagine instead of skeletons coming out of your closet they come out from underneath your sentient rug
Eric Hung damn, that’s almost as stale as my joke
CPU is free, you pay for licence :)
dzhiurgis for the first Billion, the license should be free..
After "the first billion", I guess they go to RISC-V. :- )
AMD still pays a fee to Intel for the x86 platform.
@Game Over Maybe because X64 is supposedly by AMD, not that I care...
@Game Over they reached an angreement where AMD licence x86 while intel licence x64 without paying any royalties also RISC-V is the future
It has the specifics of arm but with an open standards so companies nutoriously push this intructionset
Coming to you soon: “Oh no! I ripped my processor in half!”
@@lonewretch bs
anthro ponym it’s a joke, shut it..
flex tape it back together.
What about plastic pcb yellowing and destroying itself just by virtue of time?
@@MisterHunterWolf THAT'S A LOT OF DAMAGE
Best way to introduce somebody: "So who are you?"
thats a good question, might seem rude but asking who he is in this kind of atmosphere it a very respectful thing because the questioner is wanting to know who they are
@@Hash-Slinging-Slasher just say "would you introduce yourself" or something like that
@@user-bj3pq2si2l thats a normal conversation, usually things like this are more professional and quick
Lol
This would come in handy for prosthetic bionic limbs, hands, & feet. It would make it a lot easier to place flexible motherboards to lay within the layers of the structure itself.
Wonder what kind of wattage and thermal limits this has the potential for and how many mips you can expect
I don't imagine the thermal limits would be to high considering melting points
Probably 10-100mV
animated wallpaper here we come.
It's called "the wall" I believe
That already exists on Windows 10 Pro lol (but thats not worth the premium ;))
@@Zenheizer He's meaning physical wallpaper, not computer background
@@danholli123 oh nvm 😅
@@Zenheizer It happens, BTW, did you mean the slide show background or something else?
I'm still waiting for the day I can print out a new computer
can you imagine the cost of the ink though?
You literally can, right now, with an ordinary inkjet printer.
www.wired.co.uk/article/inkjet-print-an-electronic-circuit
@@menotu000 The ink is really expensive, but not bad compared to normal printer ink - it's about $30 a gram.
You literally can, right now, with an ordinary inkjet printer.
www.wired.co.uk/article/inkjet-print-an-electronic-circuit
@@menotu000 The ink is really expensive, but not bad compared to normal printer ink - it's about $30 a gram.
@@nathangamble125 There's a difference between printing an electronic circuit and printing a full proccessor i.e. what Jason ^^ was referring to. Did you really think he was really waiting for the day he can print out a blink LED circuit?
@@spacedesigner849 i mean you could do that with the right inks and layering
I love the way this guy's interviews humanizes CEOs
Cue Styx music mr roboto and listen close
he is like who the fuck are you but yes I see what you mean
@mPky1 Its literally just a job description, what is "completely without humanity" about that?
I was hoping that there was some zoom set on the camera, since he was otherwise getting very close to their faces. Amusing in itself.
@mPky1 subjective but true to many
I was going to help a friend out to fix his ECU in his French car.
It was not a new car by any means.
The ECU was made by Siemens or Bosch.
The box looked really robust, aluminum, clam shell type.
I opened it up and every component was soldered to this, folded plastic PCB.
It looked like a 1990's calculator inside.
I closed the box again and told him I simply lack the tools to fix it.
I think it's cool that the IC's take the leap into platics. But I hope they do it in such away that DIY'ers don't need to buy expensive tools.
who even thought that was a great idea?!
@@ETXAlienRobot201 When you look at a product and wonder why something nonsensical is the way it is, the answer is always "marketing".
oh, i'm fully aware...
it's painful since no one around me really is
it's one reason why i'm trying to avoid becoming one of their under-valued worker bees
Thank you. Welldone gentelmen. It is really amazing to discover people like you still exist. You made me happy in several ways.
Any heat issue?
mino dino it will melt the plastic lol
chip never heats itself ...the whole component heats altogether....if they provide it with a good cooling technology ...the estimated heat produced in 0.1 microsecond will be transferred immidiately and whould be completely safe.....everything depends on heat discharge rate
Arm processors use very little electricity and I doubt they'll have a high atdp but if they do have a heat problem in the future they could just use Liquid to transfer the heat away like a non conductive mineral oil
that's arm 1, from 22years ago so probably not much heat is produced
@Richard Farrey there goes your flexibility!!! :)
You need to stop moving the camera and get it out of their face my man.
Th is is the most exciting thing in ICs I've seen since the 1970s when yields began scaling up and processing power began to enter the truly useful range for mankind. What they did for $10s of dollars, this company is doing for pennies. There is a whole new frontier here and they are at the horizon of total victory.
Thank you for bringing us this interview. If you get the chance in the future, please continue to interview this team, and this company.
I think you meant 10s of thousands, but point still stands.
@@radomiami No, actually. CPUs are very cheap to manufacture thanks to economy of scale, and this has been true for decades.
The AMD 9080 (a clone of Intel's 8080, and AMD's first CPU) only cost 50 cents to make in 1974.
Most CPUs today are actually more expensive than they used to be to manufacture - usually tens of dollars - but they still charge hundreds (or thousands, for server CPUs) of dollars for them. It's worth it for consumers, but the CPU industry has very good profit margins. The reason we don't see more companies competing is because they need compatibility with common software in order for their CPUs to be worth anything.
This is exciting? We already use PCB's in electronics (plastic)
Very cool!
NSA and China will particularly love this.
“I dunno if I asked too much but,”
Dude, I like the style the way you interview.
But can it run games without melting?
Also how would you connect this to a motherboard?
ARM CPU's aren't for gaming.
@@AsttoScott you can play pubgm on phones with ARM CPUs
@@AsttoScott As long as computer games require CPU calculations, any CPU can be used for gaming.
Only a couple of minutes in, and I'm wondering how it stacks up against a 8008, 8088, and 386.
I am thinking when they can use Graphene insted of plastic .
It's gonna be a long long time.
2030 for sure! Not just graphene but carbon nanotubes and photonics!
When graphene chip can be made for 1 cent and not 1k
My jacket will be packed with ARM CPUs that will keep me warm by farming bit coin.
Dang 2 years late but still a cool topic not only that imagine, true flexible phones ultra light weigth peripherals and devices normally more smaller than usual, this stuff is sick
i dont understand why they did circles if this tech is produced by printing or something like that?
Imagine having multiple layers of these
This company is going places , very interesting Indeed.
This is ARM, if you have an android phone you are using one of their chips.
And even if you don't have an Android phone, you are using their designs. (remember, Arm does not manufacture anything)
REGALPALADIN they did go somewhere, they pretty much own the RISC chip scene now
@@firstdayversion1015 It's not plastic and fully biodegradable
Soooo sometime in the future my polyester t-shirt is actually my gaming PC that is powered by the static electricity created by friction.
A book with different logic device on each page that actually works astonishing indeed . 👌
Instead of making it smaller with slight improvements, keep the size, make it more powerful.
First gen cell phones had bigger processors than desktop PCs.
If we had batteries to keep up with our density of power for the size, the only limitation I see, it would be amazing.
But if it's done, parallel research will make up for the deficit.
Haha this video is from 2017 but youtube has just started to recommend it. Good stuff as always charbax. Love your dedication
How much processing power does this actually have? Can it run a simple operating system?
It's probably still a prototype, so that would be unknown unless there's estimates/expected result
They mentioned an equivalent of a Cortex M8 and I remembered it used in gaming mouses. It's just good enough as a co-processor.
@William Baric it's a processor.
its based off of a cortex-m0 which is.... drumroll.... 2.33MHz. you could maybe run dos on it and do some assembly code?
it's still super insane that you can actually create a CPU just by shaping up plastic though.
Much more enjoyable interview. I can not wait to see where this technology goes.
I agree with the ARM guy, this probably can't ever beat silicon in performance or possibly even performance per dollar, but silicon won't ever be as cheap or thin as this.
Interesting but substantially fact-free. Now, late 2019: where are they, boys?
might have realised it's hard to implement these as they said
they seem too easy to damage even by a slight mistake to me..... which would suck in the real world
Heat from CPU will melting the plastic, so not sure how to make it durable, but for capitalist view this is great for "planning obsolescence"
@@fadrium1464 I can see these arm processors possibly being used in cheap calculators or low powered microcontrollers.
@@fadrium1464 Heat is fair concern and one of the questions not answered in the original video - along with clock speed. But I've done plenty of MCU designs where power is only a few milliwatts. You can still get a lot done with that.
ruclips.net/video/MrqTmKF_nDA/видео.html
They are actual scientists
@@JamecBond They have invented something
Can it play Crysis without melting the core ?
I think it's for stuff that's don't need much power
Only some mhz :p
I think you could run a bios on it but a os no only a really small one that fit on rom chip
He talks about NFC applications & product tracking
We could have thousands of these running in parallel (if they design them to support multiple processors) for a cheap price.
@Murphy deffa someone also said you don't need more than 640k...
@Murphy deffa well it depens heavily how you programm the task you want to process
@Murphy deffa yea right... super computers only use 32 cores, right?
GPUs also have 32 cores? They have thousands. Investigate about "CUDA" (Nvidia) and "Stream Processors" (AMD).
As Leicht Sinn says, it all depends on how you program. Also depends on the task you want to do.
Granted, for most users even 32 cores are to many.
GPUs have 32-64 cores. Stream processors are execution units, not cores.
Super computers and server halls do not generally accelerate one task, because that’s hard; but that’s what end users tend to care about. You’re not interested in running an ensemble of games, you want to run a single instance of the game faster. You don’t generally want to do more each frame (e.g. simulating the NPCs you don’t see) you want to make each frame faster. It’s a different use case.
Also, Bill Gates/Microsoft never said 640k was enough for everybody. The 640k limit came from IBM hardware combined with the need for adresses that map to other things than physical memory.
@@soylentgreenb Yes, they have "few Compute Units". As I understand it, those compute units will send instructions to multiple small processors running in parallel. Nvidia calls those smalls processors "CUDA cores". One could argue they are cores inside a core. Scientists use GPUs as "low end cheap super computers" to execute simulations and other stuff, normal users for gaming and mining bitcoins. But ultimately they are processors running in parallel, and that is what we want.
Games tend to use "few cores" most won't use more than 4, 4~8 cores in the newest ones, even though some are programmed to take advantage of more cores. I am sure we will soon have games that will require easily 8+ as a minimum.
Applications that benefit from more and more cores are for work (graphic, science, engeneering, video editing, movie making, etc). Servers for virtualization.
It seems you are thinking only about games. Outside of games and simple office use, the higher the core/processor count the better.
However, I was wondering what is the life span for the processors in the video. Don't think a processor printed on a plastic will last long for a heavy task. Maybe they can improve (and I bet they will, if they haven't already) their technology to print on other substrates that will compete with or be better than a silicon wafer.
Thanks so much for your videos. You're a great interviewer! I'm surprised with each question.
Can that plastic handle lots of heat? If not then this suggests that those arm processors don't get too hot.
Actually there is plastic can stand heat , like the one you use in mircowave oven
This is incredibly impressive. For the first few moments I was thinking this was just a concept. Very cool.
Wouldnt the processor be ruined if it generated enough heat to warp the plastic thou? It would seem to me like heat could really effect the lifespan of the cpus and with plastic being the base it would not take very much to break them. Maybe for low-end applications it might make sense from a cost-effective standpoint but anything more complicated than a stopwatch you may have problems with the cpus getting too hot and damaging themselves. Need something that can resist heat and/or cooling I guess.
random youtube recommendations as its finest.. but two years late.. smh.. nice video btw.. would love a follow up..
i imagine they are easily cooled if cranked up in performance. Just emerge them in water.
Immerse
@White Rice Well, they aren't all Rocket Surgeons.
How's the thermals? How does you cool it down?
Just amazing, I see tons of potential.
can it run crysis ?
No it would melt the plastic or just crash 🤣
These arm socs could be easily stacked together. And these are literally paper thin. Imagine stacking these socs up to the thickness of a phones battery? Probably be a good 60 layers. That would equal an immensely powerful smartphone.
The info in this was great, but the camera work just made me uncomfortable.
How will they solder this plastic on PCB?
Does the plastics will shrink if the prosessor heating up?
Yeah, but can it run Crysis?
yeah but it starts melting
Could we see it working?
I'm wondering about heat dissipation. Like if you were to attempt to scale this up to say... Risc-V with a certain number of cores running Linux, would you begin to run into problems with keeping it cooled? Or does the thinness allow heat to dissipate more easily? In terms of manufacture, with its more robust process, is this something you could then manufacture anywhere?
This is crazy! Really appreciate your videos btw. Sounds weird but, when you stroke the golden contacts can you feel the bumps of the IC design?
Yes it feels kind of like some thick ink on paper except it's some metallic ink on plastic.
I wonder if it melts when you actual run it under load.
Would it burn if it overheated
How do I plug it? How it work with heat...
It passed almost 2 years. Where are they?
smart socks are coming everyone!
if it will be reduced to 2mm, why it needs to be flexible? and if it acceptable to make it rigit at the size of 2mm, do not we have much better density already?
My fear is when you crinkle the plastic, you make microscopic tears in the interconnects or the layers, and you IC dies! He says it's robust, but that's possibly only robust to gentle rolls of the plastic!!
Can it run windows 10?
My phone gets too hot when i do some high processing stuff, how will plastic withstand that heat?
maybe they use special substrate around
They look like star trek's iso linear chips.
just one of many star trek tech that actually became real
Does Scott White ever blink? He looks like a crazy man, but in reality he's extremely bright and surrounds himself with the brightest in the industry. He's totally focused on his objectives.
Reminds of me 2004 Jeff bezos look
We need ARM processors in our arms :P
They might be able to tattoo the ARM processor on your ARM..
Stephen Sharp: Implement a crypto accelerator and NFC and I'm on board.
Stephen Sharp Brothers in Arms!
hey charbax really enjoying your vids. Is there also a printed electronics company that is doing TFTs to control pixels?
I laughed too hard at this....
So that's what the $5 raspberry pi's are made of
this is just three guys who took the film inlay out of a box of chocolates talking shit
1:34 damn that girl has long legs
@Avis *Slendergirl
Look at the processor bro 😑
Slender
Yup
Chinese fabricators taking notes furiously
Yeah
must be hard to keep the substrate from moving from start to finish
just wondering... doesn't processors gets hot when running... processing... things?
Sounds great but how sturdy is it? How long will this work if put into clothes for example? Plastic connectors like this in laptops can break within a year with no movement at all and those processors look even thinner than those...
If it heat up ?
so basically this is just like making 2 way mirror material like space blankets, I love how he danced all around saying they use vacuum vapor deposition, and then acids to etch since the plastic is impervious to the majority of those acids, while say the aluminum coating on those thermal or "space" survival blankets is exposed on one side,
So I'm curious, how would heat be managed on such a plastic chipset?
Still have solid motherboards though right?
Why Contain it ?
is it available to buy???
I can't wait until I get in my hand the iglass
One queastion:"When"?
3:03 table being cleared, girl wipes table off with hand then wipes hand on apron, not very sanitary.
She was wiping the table off with processors.
2:15 "So who are you ?" is Funniest part. :D
Just one step towards having these chips implanted in our brain :)
8:15 P-E-N-15 micron
I'm losing my shit here. 😂 Nice catch.
Thought you would of moved to diamond storage and processing
It is an crucial for a man to be bald for being a CEO, lol.
It's company policy
the stress will make u bald if ur not already
Is this Aprils day one?
I got a question, how is it with the TDP? Even if its low power the heat has to go anywhere...
Flexible heatsinks?
Not actually a problem. You must be thinking of plastic as thermal insulator, but at sufficiently low power, it isn't, it doesn't have a lot of thermal conductivity but it can have plenty. If you think of most integrated circuits, the IC die is sat a millimetre deep inside an epoxy enclosure (typical black chips) and some heat goes into the PCB where the copper fill dissipates the heat through yet another layer of epoxy namely the solder stop lacquer and the silkscreen, basically all thermal dissipation occurs in most typical low-performance circuits through copious layers of plastic. The materials they're dicing around here have similar thermal performance, i think something like polyimide even has a slightly better performance than epoxy.
Another thing to consider is the Cortex M0 electronics itself. The typical ones i've seen have an internal 1.8V regulator and a bit of the power consumption comes from the regulator, a tiny bit from the computational core itself, but most really from I/O circuits which is a 3.3V signal converter, static protection circuits, pullups, draw induced by connected devices, so basically all of it is due to having to interface with integrated circuits which have existed now for decades, standard electronics. Now if you design this as a disposable circuit from the get go, you can dispense with all of that complexity, and with further integration you don't need to interface with a different I/O voltage and can also reduce the pullup current. So at the end of the day it depends on the switching performance and voltage thresholds of the underlying flexible MOSFET, the applicable supply voltage, whether they're able to make it work, they can work with their library of fully flexible/printable circuit elements to achieve a good leeway in power consumption.
I'm almost more worried about how they want to go about power delivery for most use-cases rather than power dissipation. For RFID tag it's trivial enough, and it's a fundamentally low power circuit, orders of magnitude lower power than a processor, but other things? You can dissipate a lot more power under a fairly thermally unfortunate chair cushion than you can store energy chemically in a button cell. Given the power delivery constraint, i'm not sure the world necessarily needs a flexible ARM Cortex M0 processor for now, though in due time it may be promising as a cost cutting measure.
DIYer - Overclocks
CPU - I'm liquid
PS: Amazing concept tho!
3:23 with the humans that are using them, did he just let slip that he's an alien?
when i socket this in my computer will its temperature rise up to destroy the plastic
Well done , waiting to see it soon available around us.
But don’t they generate a lot of heat
why not risc-v?
He said ARM is one of thier investors
but risc-v has free bsd license . hobbyist not able to full fills that's ideas on arm.
open isa in risc-v.
How this product survive climate..heat?
What's the speed
I only have 2 questions ,, is that sheet hes holding a working unit ? and secondly are these room temperature conducters since what I understand that's not a thing yet so how would they be cooled ???
I have no knowledge on whether it's a working processor, there hasn't been a powered-on demonstration after all.
But what do you even mean by second question? It's not like you need to dunk your phone into liquid nitrogen for it to work, it's as if you're confusing semiconductors and superconductors.
How would they be cooled? You can dissipate several dozens to hundreds of mW through the plastic sheet easily, no problem whatsoever, but a circuit that consumes that much would be impractical due to power delivery issues, so it stands to reason they're likely aiming for a sub-mW power circuits. In fact most simpler electronic circuits you handle in your daily life dissipate heat through a fairly unfortunate amount of plastic and stale air and other material arrangements with bad thermal conductivity and they're perfectly fine.
@@SianaGearz so then its basically just a shiny idea on a piece of plastic paper coming from 3 pocket dwellers out of some think tank at the risk corp im assuming !?! and also yes if you've ever had your phone apart and looked at the processors some have a metal shield glued on for heat spreading so they don't crack which splits the circuit and destroys it and others either expel heat straight to the board and through the faraday casing and also super cheap ones like kyoceras old pocket phones the whole board array heats up due to lack of proper adaptation ,,its no the battery you feeling when youre texting on a hot day and and you think holy toast my phone is hot its the processor cooking the whole phone inside out ,, as to my supposed confusion no im not confusing semiconductors , transistors heat to expelled traped energy it cant expel and if a super conducter is used all energy is either used or diverted thus being a "super" since it has no resistance ,, no resistance ,no heat.
@@jessemazo4902 They're not some basement dwellers, they have research, they have a production line and they're marketing a product - a relatively low-tech RF tag that is super cheap to make and has enough of an unencrypted digital storage to uniquely identify it. It's suitable for product tracking and not suitable for high-security use, but it's super cheap and causes less handling and production line automation issues than traditional tags that are made of paper and have a silicon IC glued onto them, and allegedly higher reliability.
The processor shown is not a product and might or might not eventually be, but that's what they're working towards - they say they have a long way to go towards it being practical enough for anyone to actually buy one. It would be prudent to assume that they have validated that these structures that they're showing are actually being built correctly enough to work. There are some things they aren't saying though, they aren't disclosing the risks that may lead to it never becoming a product, and i have some ideas in that regard.
On phones, depends, the processor shown won't run a smartphone! It's an ARM Cortex M0, the kind of processor you'll find in a wireless controller chip or in a high-end gaming mouse or keyboard doing just fairly easy interface duty. Existing Cortex M0 implementations have a sub-1mW core consumption and most of their consumption is in the IO circuit which can be eliminated if you don't need to interface with traditional existing ICs, but only with other flex-printed circuits. Smartphone's power envelope is around 300-900mW, so it does need some cooling considerations, but the metal shielding is an "RF can" to suppress interference, it's not for cooling. On more powerful phones you'll find a hole in the RF can and the processor heatsunk into the back of the screen or alloy middle enclosure directly. On less powerful ones, you'll often find atrocious "thermal compound" that doesn't deserve the name or worse, a bubble of stale air, one of the most vicious thermal insulators short of vacuum - but of course they still have cooling via the PCB, through copious layers of plastic but at least with some area to go with it. Which is to say these guys here also can hit the cooling targets of vaguely similar order of magnitude, but it just wouldn't make for a good product if they do.
I think ultimately power delivery and power efficiency are going to be the issues that limit their products or make them unmarketable, not heat dissipation. Heat dissipation is an issue they'd LOVE to have, because if they were in position to have that, it would mean they solved the power delivery issue at least.
@@SianaGearz well said , I was assuming these were ARM micro processors
when i can buy flexible screen for my clothes?
Do you want to be a teletubby?
Can it run Doom?
Can it run Crysis?