• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

What Speed Won't We Use Transistors Anymore?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Foxie3a

Normal Member
Joined
Sep 7, 2003
This is kinda like a Poll.

What speed do you guys think that we will stop using transistors in our desktop processors.

An example would be Quantom Computing.

I think that around 20ghz we'll find some alternative, I doubt Quantom computers will be it though.
 
20 GHz is easily achievable with near-future transistor designs. Intel has already demonstrated 1 THz transistors, which could eventually be used to build an IC running at roughly 40 GHz (depending largely on the design). And IBM has demonstrated a working IC running at 110 GHz (though of course not a full, general purpose processor).

Ultimately, the barrier for transistor speeds would be if and when single-atomic-scale transistors can be easily fabricated. It's difficult to predict how fast such transistors could run, but 500 THz to 1 petahertz could be feasible.

Even if quantum computing and optical processing become widely used, they cannot displace transistors completely. Both are limited in what kinds of processing they can do, and transistors would most likely still be needed for control logic and digital/analog conversion.

I would hazard a guess that transistors will be widely used for most of the 21st century. But transistors 50 years from now will look very different from those of today.
 
Some thoughts on the question and the idea of Quantum computing.

We are tending to see the theoretical limits of traditional semicondutor transistor switching IC's come to light. I have heard many top flight researchers discuss the next major evolution/inovation will be quantam level computing. Whether or not this will actually be the advent of our next step beyond our current speed barrier, we just dont know.

Quantum computing brings in many new and foreign questions and challenges to the computing world. Quantum physics itself is the only real modern science. There are very few world experts who actually understand and can manipulate quantum concepts with ease. Einstein himself discounted the theories he himself had worked on in quantum, because they went against all known computational and theoretical practices.

One thing people dont understand is that quantum is a science of unpredictability. What actual happens in a quantum based algorithm is that an answer can be predicted within a certain acuracy. Though the method to the answer is rather complex. Quantum algorithms involve theories of granulating order to dissorder and then back to order again with a formulatiing factor 'K'. Basic entropy. So what challenges our scientists is how to harness an algorithm that gets its answer every time, but uses a different method to get that answer every cycle. This goes angainst all current bread and butter 1*0=0 logic.

Its going to be an interesting next 50 years. Even the US government knows this. Which is why they have publicly released all the unfinished Enstein diaries hoping some will continue his unfinished research that he was too apprehensive to finish.

On the question of when we will abandon traditional FET gates as the staple of our CPU designs? Probably not anytime soon. The FET switching archetecture will most likely stay around for at least our lifetimes. Change takes time, and no one is willing to change any time soon.

Silicon based chips will most likely stay around for a very long time also. Optical interconects as well as high-K interfaces are looking like the new way. Though these are just modulations of current transistor designs.

So to answer your question in different way. I assume we may never see a transition from FET archetecture, at least as long as we live. Though dont quote me on that... :D
 
but that's not a processor, it more like an entire system.
 
Uhhh, the brain actually operates at something around a few hertz... its just a different type of system that is very powerful. The brain isnt supposed to be fast, but I do believe the way nerves work is sorta like transistors being on and off (I slept through my high school anatomy/psyiology (sp?) class).
 
itizme said:
The fastest processor known to man is ..............................
THE BRAIN

Not true at all. The Brain is relatively slow since it is a bio-computer designed over millions of years to analyze situational problems and control body functions. None of those things require speed as much as computational ability.

The Brain is an awesome multitasking, input processing and creative piece of equipment...but lets face facts, any 10 dollar calculator will smoke it in a speed contest. Just don't ask the calculator to make a decision on its own...;)
 
This is just my theory/oppinion, i don't know what really goes on or what respectable scientists believe.

Quantum computers do have the potential to take the place of transistors; however, human technology has a long way to go before quantum computers can completely take over, and believe me... quantum computers are just about the final step as far as computer hardware evolution is concerned... theoretically a quantum computers power depends on how clever a programmer is, with potentially infinite amounts of processing power. In a way a brain is like a kind of quantum computer with the way that its biological properties have influence over the quantum vaccume fluctuations. On the quantum level, there is a thing called hilbert space, which is a kind of energy which passes through different outcomes of the same universe. These outcomes and possibilities are travelling in waves which go off in opposite directions of eachother and then reunite, theres a whole bunca these things. Its like a quantum does multiple outcomes due to effects from hilbert space (a place outside of all of these universes that appears to be composed of an infinite amount of energy taking up an infinite amount of space. Things happen on a sort of cycle dictated by the laws of physics of a particular universe. An example is like, if u were a quantum particle, and u lived a normal human life, u might stop reading this now and do something else, and also keep reading. My point is that even if u did either, there are certain fundamental schedules that you still follow. You still will probably wake up from bed at roughly the same time the next day most likely because not only may you use an alarm clock, your biology needs a certain amount of sleep. In the quantum world there are simular schedules that are dictated by a giant equation that is only a kind of estimate and is very very long that describes an "alpha" of a particular universe... the laws of physics can also change from universe to universe and within a universe over time they also can change; however, the parrellel universes that overlap with this one on the quantum level are wrapped tightly around this one... and in a sense are part of this one- its what causes such a massive amount of stability quantum physics, in terms of a universe- alot of people think quantum physics is really insane and hard to understand why it works the way it does when really it works the way it does in order that "possible" things happen, while impossible things don't. Impossible things are like impossible shapes, for example try drawing a picture without any form of instrament to mark the paper. That is sort of what its like, trying to draw a universe without a special universe drawing marker and artist. The main idea or thesis of this paragraph, is that while different outcomes occur simultaniously for one particle, the end result of those two different outcomes can be the same. You stop reading this sooner, but you still wake up tommarow morning at the exact same time... and over time the waves reunite. This is just my theory and opinion, i don't really know what is going on or what scientists think.

Quantum computers becomming practical is a long way off for humans. Before they even get there they have to get to the atomic level... sure they can proove they work so far and utilize them to an extent, but not without a good deal of $$$ + they arn't optimized for much anything now other than proof that they work and experimentation to further learn more about quantum physics. Its definitely not like blowing up an atomic bomb... you can't have a theory like e=mc^2 and then smash a few atoms and you have a quantum computer, a true quantum computer that utilizes its amazing potential, would be alot like configuring an atomic bomb to explode and then output to some viewer a very accurate answer to pi or some extremely ridiculously large prime number (unimaginable amounts of digits! Trying to take all that energy and organize its displacement is far more difficult than just having alot of POWER, it requires alot of knowledge and sitting down and figuring out solutions to all of the problems we must overcome in order to facilitate such powerful potentials.
 
Last edited:
:-/ my friend i know added-

Technically, I could design something that makes an atom bomb produce a very large cheeseburger

Well, see, the detonation releases gamma rays. Two intersecting gamma rays can be made to form a subatomic particle. So you can use a device that produces a complex magnetic field structure, to force the released gamma rays to line up and interact in such a way to materialized through gamma-gamma collisions, all the subatomic particles required to comprise a cheeseburger with everything on it.
 
Well, look at it this way. We don't NEED more MHz to make things faster. Just look at AMD vs Intel. It's possible to just make the processors more powerful per MHz rather than requiring extra MHz just for power. How about a processor that can actually process instead? We can't follow Intel into the future because it's become painfully clear all they care about is how many MHz they can pump through a sheitty, weak core.
 
I wasn't intending to get into that, it was just supposed to be a nice little thread. Not debate.
 
just a random thing, quantum physics and quantum computing are 2 different things, quantum physics is not yet even known, those who say they know what quanum physics is, and the rules to it are just plain mis informed, for the truth is that stephan hawking is still figuring out what rules quantum physics would follow, and what the basis is. quantum physics at this point is just a mere idea, and theory, but has no rules.

quantum computing on the other hand is just the introduction of 2, so u have yes, no, maybe. this however makes it quite scary since at that point a computer can guess....(think about it........)

quantum computers however use light, not parrellel universes to do computing. just thought I'd share
 
Back