Is there a finite limit to PC speed?

[juretrn]

For size, Moore's law won't go on forever, because we're limited to the size of an atom.

 

Yes, that is very small, but it means that Moore's law does end. We have a size limit.

Silicon transistors will once be replaced with something else. Graphene is the next evolution in microelectronics. Higher clocks, smaller manufacturing processes...

So for the same performance, you'll need less components. 

But remember, Lockheed Martin is already using a quantum computer for some top secret stuff. 

 

Basically, the limit is not the atom, but the effect of quantum tunneling. When we hit that limit, the gate will be 5 nm wide. We are currently at 22 nm. 

[Lord Pantaloons]

I believe that in 100 years be will see intel CPU's at over 9000Ghz

[Glenwing]

Yes, there is.  Nothing is infinite.

 

How far away the limit is and what the maximum is?  No one can know.

[Godlygamer23]

Yes. There is a limit to PC speed. Transistors will eventually become about the size of an atom. At that point, you cannot do anything except make computer chips bigger to accomodate more transistors.

[Lil Chillbil]

There were people who thought that pentium 4's and 2 gigs of ram could be used forever. Then the enthusiasts like us wanted better multitasking and now we have more and more cores. People thought battlefield 1942 looked supreme and no one could really max it out in 2002. The enthusiasts wanted higher res and 1080p and so it was. 

 

 

 

 

the pros/enthusiasts make it better for the students/casual's and then they call us nerds... 

[Vitalius]

Nothing is infinite.

When you realize that Nothing is technically something, you will see how profound a statement that is. And how paradoxical. 

... But I kid. 

I personally think we can't know if there is a true absolute limit to anything. Who knows, maybe we will discover a way to change the rules of the universe we exist in, rather than purely exploit them as we have been, some day. But that would be crazy right? 

[WoodenMarker]

I think they will be a point where we can directly affect our brains/thoughts with technology so yes. Eventually, we could probably affect how we perceive things in ways we would not notice.

[AraSarkisyan]

For size, Moore's law won't go on forever, because we're limited to the size of an atom.

 

Yes, that is very small, but it means that Moore's law does end. We have a size limit.

not unless we make processors bigger

[Vitalius]

I think they will be a point where we can directly affect our brains/thoughts with technology so yes. Eventually, we could probably affect how we perceive things in ways we would not notice.

Oh man. Imagine playing Super Hot, but in Augmented Reality with technology screwing with your perception of time. ... that would be so disorientatingly awesome. 

Ugh. My head hurts thinking about it. Maybe that is just the Black Coffee and fact I have not eaten anything all day.

[WoodenMarker]

Oh man. Imagine playing Super Hot, but in Augmented Reality with technology screwing with your perception of time. ... that would be so disorientatingly awesome. 

Ugh. My head hurts thinking about it. Maybe that is just the Black Coffee and fact I have not eaten anything all day.

The problem is not enough coffee. The solution is always more coffee.

[delet my acc pls]

I believe that in 100 years be will see intel CPU's at over 9000Ghz

what if intel goes under in 100 years and lenovo have bought AMD and is therefore the biggest CPU maker in the galaxy. What i mean is that lots can happen in 100 years

[Godlygamer23]

Basically, the limit is not the atom, but the effect of quantum tunneling. When we hit that limit, the gate will be 5 nm wide. We are currently at 22 nm. 

Atoms are smaller than 5nm. Way smaller. Atoms are eventually going to be the limit. We cannot construct any element smaller than an atom.

[Vitalius]

Atoms are smaller than 5nm. Way smaller. Atoms are eventually going to be the limit. We cannot construct any element smaller than an atom.

Who said it has to be an element? What if 1's are represented by sent electrons over a period of time and 0's are represented by the lack of electrons sent over a period of time (the same period). Assuming we can calculate that on the fly. 

... That would be like... 10 bajillion GHz fast.

[Godlygamer23]

Who said it has to be an element? What if 1's are represented by sent electrons over a period of time and 0's are represented by the lack of electrons sent over a period of time (the same period). Assuming we can calculate that on the fly. 

... That would be like... 10 bajillion GHz fast.

What element is going to allow for that to occur? The element is the medium. How are the electrons supposed to move? You need a conductor, which comes from elements.

[juretrn]

Atoms are smaller than 5nm. Way smaller. Atoms are eventually going to be the limit. We cannot construct any element smaller than an atom.

Gates can be smaller than 5 nm. Why the hell not? However, the transistor will not work correctly when it gets below the limit of 5 nm. You cannot have a gate that is just an atom wide. The electric fields would be all over the place in that case. 

[DirkaFat]

What element is going to allow for that to occur? The element is the medium. How are the electrons supposed to move? You need a conductor, which comes from elements.

Free floating electrons from the human body.

 

 

 

joking

[Bluethulhu]

Theoretically? Yes. Practically? No.

[MJSBLUES]

The day will happen when pixels are a atom wide, we clocked the fastest material and the transistors are also a couple atoms big and produce little heat.

*an

[Vitalius]

What element is going to allow for that to occur? The element is the medium. How are the electrons supposed to move? You need a conductor, which comes from elements.

Gold. The same stuff that is on the CPU now. What CPU's basically do now is look at whether electricity is flowing through the gates. 

However, Electricity is simply Electrons flowing through the elements, basically. 

What if we, somehow, measure the flow of electrons rather than whether or not electricity is flowing? Mind you, measuring the flow of electrons and whether there is electricity or not is basically the same thing, only, in terms of electrons, you can do a lot more in a much shorter amount of time. 

Think of an electron as a packet (like how networking works). Each packet sent is a 1. If a packet is not sent within [insert unit of time], it is considered a 0. 

Not sure how that would work for a CPU, and is probably not feasible. But it is cool to think about.

[Godlygamer23]

Gates can be smaller than 5 nm. Why the hell not? However, the transistor will not work correctly when it gets below the limit of 5 nm. You cannot have a gate that is just an atom wide. The electric fields would be all over the place in that case. 

An optical transistor has been contructed. http://web.mit.edu/newsoffice/2013/computing-with-light-0704.html

[Vitalius]

An optical transistor has been contructed. http://web.mit.edu/newsoffice/2013/computing-with-light-0704.html

Well, if I am right in how I imagine that would work, you could just replace "Electron" with "Photon" in everything I have said and it apply to that as well. That would be cool. And it seems far more feasible. 

Hmm, imagine how enthusiasts would be with CPUs that run on optical transistors. We would make vacuums so that absolutely nothing got in the way of the light, thereby speeding up calculations, if only by 0.0000000000....0001%. lol 

Talk about extreme.

[juretrn]

An optical transistor has been contructed. http://web.mit.edu/newsoffice/2013/computing-with-light-0704.html

If you read what I wrote, you'd see that I was talking about traditional silicon ones. Of course, using different technologies, the boundary for size is moved further. But CMOS, 5 nm is the end of the line.  

[James_AJ]

Once we get into optical or quantum computing things will change as we know it and there won't be a finite limit anymore (almost)

[Mentalguy]

An optical transistor has been contructed. http://web.mit.edu/newsoffice/2013/computing-with-light-0704.html

But you are limited to how fast light can travel through it.

 

 

In terms of clock speed, that is the limit we can achieve. At one gigahertz, light can only move about a foot, per clock(1 light-nanosecond). with 10ghz you are reaching a clock speed so fast light(or electrons) cant cant reach the other end of the chip before the next cycle.

 

Once we get into optical or quantum computing things will change as we know it and there won't be a finite limit anymore (almost)

the speed of light is the limit. its not as fast as you think.

[Vitalius]

But you are limited to how fast light can travel through it.

 

 

In terms of clock speed, that is the limit we can achieve. At one gigahertz, light can only move about a foot, per clock(1 light-nanosecond). with 10ghz you are reaching a clock speed so fast light(or electrons) cant cant reach the other end of the chip before the next cycle.

 

the speed of light is the limit. its not as fast as you think.

Woo for Quantum Tunneling! 

Instantaneous Calculations are the future!