What Happens After When We Get To The Lowest Manufacturing Process?

[Veltrix]

I just was curious what happens after Intel or AMD reach like the 1nm process, I know it won't reach like 1nm I think because I know the increments doesn't really work that way but I am trying to say is there going to be a new standard of the manufacturing of CPU when Intel and AMD are done with the nano meter process? Just wondering now because we are getting close and technology I see if rapidly in fast development.  

[fishymamba]

Well there are smaller measurements than nanometers...maybe they will jump down to picometers. But that will take a pretty long time and there will be many problems to overcome, like heat. 

[XiaoPigs]

I think we'll be stuck with nanometers for quite some time. I'm not an engineer but I'm pretty sure a dye that's in Picometers will be really really hard to manufacture.

[Veltrix]

I think we'll be stuck with nanometers for quite some time. I'm not an engineer but I'm pretty sure a dye that's in Picometers will be really really hard to manufacture.

Oh thanks for clearing that up and btw I like your profile pic :D 

[XiaoPigs]

Oh thanks for clearing that up and btw I like your profile pic :D

No problem ^^ Thanks for the compliment as well.

[amintha1992]

my prediction (not a educated guess or something that i know for sure) - by the time we run out of nano meters, the standers that is used in computing will be absolute. who knows we might move on to quantum computing.  

[Dodge]

There is a limit and we'll hit it in a few years. You can't change the size of a Si-atom and you can't change the distance between these atoms. Then you have to use different materials.

[bradscoolio]

The "nanometer" manufacturing process you speak of will almost certainly never reach 1nm, it simply would take far to long to get to that point.

What I see happening is the adoption of molecular based transistors on chip and then that being refined over a long period of time until it's not deemed profitable to refine it any longer.

Then at that point we'll start to see quantum based CPU's popping into consumers hands. 

 

Both quantum computing and molecular computing have proven successful in lab, with scientists being able to use only 3 quantum atoms to represent the number 15 (in binary that would use many transistors at thousands of atoms per transistor) and store an entire song on 1 strand of DNA.

 

 

PS: Google & Lockheed Martin bought quantum computers from a company called D-Wave quite some time ago and apparently Google has been using it to experiment with image recognition. No clue what Lockheed is doing with theirs though :P

[TechFan@ic]

There is a physical limit to how small we can use Silicone as a semiconductor & so the industry will have to move from Silicone to a new standard once we hit this nm wall predicted sometime between 2022 & 2030.
Graphene has been successfully shown to be superior to Silicone in many ways & many predict that it will compete with Silicone or replace it as the industry's standard in a few years time.

[Nerdsauce]

Silicone is already nearing its limit in regards how small they can make processors. During tests they've discovered that long before we reach the 1nm scale Silicone actually starts leaking information, until the point it leaks faster than the input rate (meaning it literally cannot store information full stop). In the next ten years or so it will hit that limit, and so they're already looking into replacements for it. Graphene is one solution, although they are also looking at plastic composites and there has even been experiments in using algae cells as processors (the benefit with those being they can actually produce the power they need to work internally, and are near-infinitely scalable in that if you need more processing power you just allow the algae to reproduce, and if the "processor" starts to get old and die you can simply replace it with new algae cells meaning a brand-new processor on-demand).

[XiaoPigs]

Silicone is already nearing its limit in regards how small they can make processors. During tests they've discovered that long before we reach the 1nm scale Silicone actually starts leaking information, until the point it leaks faster than the input rate (meaning it literally cannot store information full stop). In the next ten years or so it will hit that limit, and so they're already looking into replacements for it. Graphene is one solution, although they are also looking at plastic composites and there has even been experiments in using algae cells as processors (the benefit with those being they can actually produce the power they need to work internally, and are near-infinitely scalable in that if you need more processing power you just allow the algae to reproduce, and if the "processor" starts to get old and die you can simply replace it with new algae cells meaning a brand-new processor on-demand).

Very interesting information, thanks for that. In the end it'll be all up to the Intel/AMD engineers to bring it to fruition. Let's see where they take us in the future.

[Nerdsauce]

Very interesting information, thanks for that. In the end it'll be all up to the Intel/AMD engineers to bring it to fruition. Let's see where they take us in the future.

 

Exactly! In actual reality, the death of Silicone won't be a bad thing. It's akin to the death of the CRT monitor making way for flat-screen technology. It might actually free up engineers to really go for broke and push the boundaries of our technology to new heights.

[Veltrix]

The "nanometer" manufacturing process you speak of will almost certainly never reach 1nm, it simply would take far to long to get to that point.

What I see happening is the adoption of molecular based transistors on chip and then that being refined over a long period of time until it's not deemed profitable to refine it any longer.

Then at that point we'll start to see quantum based CPU's popping into consumers hands.

Both quantum computing and molecular computing have proven successful in lab, with scientists being able to use only 3 quantum atoms to represent the number 15 (in binary that would use many transistors at thousands of atoms per transistor) and store an entire song on 1 strand of DNA.

PS: Google & Lockheed Martin bought quantum computers from a company called D-Wave quite some time ago and apparently Google has been using it to experiment with image recognition. No clue what Lockheed is doing with theirs though :P

Oh I really tried to emphasize saying that 1nm I should have really said when we reach the maximum smallest manufacturing in nanometer.

[bradscoolio]

Oh I really tried to emphasize saying that 1nm I should have really said when we reach the maximum smallest manufacturing in nanometer.

 

We won't, we'll move on before then.

[Located]

We won't, we'll move on before then.

 

The silicon atom itself is +/- 0.1nm so you can imagine we cant go much smaller with the conventional transistors! :-) http://www.pcworld.com/article/250317/atom_sized_transistor_foretells_quantum_computer_scientists_say.html

 

 

though quantum computers are then a solution. 

[James_AJ]

We will hit limit but not soon.. then we transfer to new technology - something like CPUs operated by light (IBM already has this, releasing in 2017) or quantum computers.

[Glenwing]

Graphene seems very close to being a viable replacement for silicon.  IBM has already demonstrated a 100GHz (not a typo) broadband radio-frequency mixer, however of course this is an analog device.  The inherent problem is that although it's one of the most conductive materials, it doesn't have a band gap meaning it's not a natural semi-conductor like silicon, and that ability to switch on and off is required to handle digital information like computer microprocessors.  However artificial methods of introducing a band gap seem to be bearing fruit now.

 

http://www.extremetech.com/extreme/149333-researchers-create-cmos-compatible-30nm-programmable-graphene-transistor

[Jogi6]

I don't think molecular based transistors or  quantum based CPU's will become a thing in near future, probably we will get to some "nm" limit of how small the can go. But then they will only increase phisical CPU size. More space means more transistors.

[Glenwing]

Chips can only get so large; otherwise the time/signalling voltage required to get a signal from one side of the CPU to another becomes to great.

[Rhotis]

 

 No clue what Lockheed is doing with theirs though :P

 

[bradscoolio]

 

I'm not sure it works like that :P

[Art Vandelay]

Well there are smaller measurements than nanometers...maybe they will jump down to picometers. But that will take a pretty long time and there will be many problems to overcome, like heat. 

 

Quantum tunneling becomes a problem at some point. No idea how they could possibly overcome that.

[crystal6tak]

I saw somewhere that when we are done with silicon, we would move on to using graphene. I don't know the details as I didn't look into it much.

[Glenwing]

It's in the works.  Graphene isn't a natural semiconductor which is necessary to handle digital information like computers do.  There has been a lot of progress designing graphene transistors that overcome this problem though.