Computers Faster Only for 75 More Years?

G

Geriatrix

Executive Member
Joined
Nov 22, 2005
Messages
6,554
Reaction score
3
Location
Right here
http://www.physorg.com/news174750105.html
A pair of physicists has shown that computers have a speed limit as unbreakable as the speed of light. If processors continue to accelerate as they have in the past, we'll hit the wall of faster processing in less than a century.

Intel co-founder Gordon Moore predicted 40 years ago that manufacturers could double computing speed every two years or so by cramming ever-tinier transistors on a chip. His prediction became known as Moore's Law, and it has held true throughout the evolution of computers -- the fastest processor today beats out a ten-year-old competitor by a factor of about 30.

If components are to continue shrinking, physicists must eventually code bits of information onto ever smaller particles. Smaller means faster in the microelectronic world, but physicists Lev Levitin and Tommaso Toffoli at Boston University in Massachusetts, have slapped a speed limit on computing, no matter how small the components get.

"If we believe in Moore's laW ... then it would take about 75 to 80 years to achieve this quantum limit," Levitin said.

"No system can overcome that limit. It doesn't depend on the physical nature of the system or how it's implemented, what algorithm you use for computation … any choice of hardware and software," Levitin said. "This bound poses an absolute law of nature, just like the speed of light."

Scott Aaronson, an assistant professor of electrical engineering and computer science at the Massachusetts Institute of Technology in Cambridge, thought Levitin's estimate of 75 years extremely optimistic.

Moore's Law, he said, probably won't hold for more than 20 years.

In the early 1980s, Levitin singled out a quantum elementary operation, the most basic task a quantum computer could carry out. In a paper published today in the journal Physical Review Letters, Levitin and Toffoli present an equation for the minimum sliver of time it takes for this elementary operation to occur. This establishes the speed limit for all possible computers.

Using their equation, Levitin and Toffoli calculated that, for every unit of energy, a perfect quantum computer spits out ten quadrillion more operations each second than today's fastest processors.

"It's very important to try to establish a fundamental limit -- how far we can go using these resources," Levitin explained.

The physicists pointed out that technological barriers might slow down Moore's law as we approach this limit. Quantum computers, unlike electrical ones, can't handle "noise" -- a kink in a wire or a change in temperature can cause havoc. Overcoming this weakness to make quantum computing a reality will take time and more research.

As computer components are packed tighter and tighter together, companies are finding that the newer processors are getting hotter sooner than they are getting faster. Hence the recent trend in duo and quad-core processing; rather than build faster processors, manufacturers place them in tandem to keep the heat levels tolerable while computing
speeds shoot up. Scientists who need to churn through vast numbers of calculations might one day turn to superconducting computers cooled to drastically frigid temperatures. But even with these clever tactics, Levitin and Toffoli said, there's no getting past the fundamental speed limit.

Aaronson called it beautiful that such a limit exists.

"From a theorist's perspective, it's good to know that fundamental limits are there, sort of an absolute ceiling," he said. "You may say it's disappointing that we can't build infinitely fast computers, but as a picture of the world, if you have a theory of physics allows for
infinitely fast computation, there could be a problem with that theory."
Click to expand...
 
I am sure in 60 years cpu's will not be as they are now. They will find something new to power pc's that will consume very little power and be far more powerful.
 
i dunno...

so what if they hit a barrier and can't go smaller - you can always add more cores (as long as programmers can code multi-core apps)
 
I thought one day they would get a time machine in a CPU that would give it unlimited time to compute, and therefore a Digital Watch CPU can process Adobe Photoshop C20
 
yes, but one's thinking must look beyond these physical limitations; as 'speed' is just a means to an end.

E.g. why does one want to travel faster than the speed of light...surely its not for the thrill of it...its because you want to get somewhere faster...hence theoretically we start thinking of folding space, wormholes and the like; and so bypassing the theoretical limitations of c

In the same way, we will in the future think differently about computing power since lightning fast processing power is just to get to an answer quicker...so maybe we will start thinking "hey, why do we have to calculate the answer, why cant we just 'see' the answer?"...like a savant :cool:

Maybe we will then have 'Savant' processors, for specific disciplines, that just knows stuff, without any calcs

OK, I'm probably talking BS :D
 
Geriatrix said:
http://www.physorg.com/news174750105.html
Click to expand...

Scott Aaronson, an assistant professor of electrical engineering and computer science at the Massachusetts Institute of Technology in Cambridge, thought Levitin's estimate of 75 years extremely optimistic.

Moore's Law, he said, probably won't hold for more than 20 years.
Click to expand...

I agree! Fifteen years even.

The way information is processed is predominantly serial (and binary). Parallel processing is limited to the problem (serial processing is more versatile). I reckon a new processing paradigm will be developed to accomplish more processing within the limitations. Time is running-out. As usual, there will be a last minute panic at midnight on the eve of the limitations being reached.
 
Kompete said:
yes, but one's thinking must look beyond these physical limitations; as 'speed' is just a means to an end.

E.g. why does one want to travel faster than the speed of light...surely its not for the thrill of it...its because you want to get somewhere faster...hence theoretically we start thinking of folding space, wormholes and the like; and so bypassing the theoretical limitations of c

In the same way, we will in the future think differently about computing power since lightning fast processing power is just to get to an answer quicker...so maybe we will start thinking "hey, why do we have to calculate the answer, why cant we just 'see' the answer?"...like a savant :cool:

Maybe we will then have 'Savant' processors, for specific disciplines, that just knows stuff, without any calcs

OK, I'm probably talking BS :D
Click to expand...

Nice try
 
Heat dissipation aside (a limit that must not be underestimated) there are other factors. Development cycle to the next generation of processes (with satisfactory yield), fabrication plants, architecture (some trade-offs may be two seps forward and three steps back), tools, and software (bug free!) will also became longer. Pushing back the boundries will became harder and harder at each frontier as we approach the limit.

Our capacity to use or require the additional power or be tempted by the multi-nationals to always upgrade will also be a factor. Look at Windows XP. They could force the issue by discontinuation but the backlash may be more damaging. It will happen but mass adoption or embrace of next step-up will be slower than before. The huge investment necessary will therefore be progressive, further slowing down the development cycle.

At some point, in order to fully utilise the additional processing power a commensurate development of all the surrounding interfaces or dependencies will need to take place. These will have their own physical limits or dynamics. Again affecting the progress to the Ultimate Machine to pose the ultimate question about Life, the Universe and Everything.

Time aside, can we reach the limit AND use the capability for real world applicatons. Not a chance. How close will we get? I suspect not as close as one would romantically expect (look at space exploration - we seem to have found our limit to be well below our potential).
 
Last edited:
Why does no one remember Moore's second law... That the price of manufacturing will increase in the same manner (or similar). Though I'm not sure on the exact nature of Moore's second law... :)
 
Ok I have read through the article and it doesn't seem to address that multicore processors can stack speed??? It also comes down pipe lines and pathways doesn't it?
 
NoLimit said:
Time aside, can we reach the limit AND use the capability for real world applicatons.
Click to expand...

My personal favourite:

From an article quoted by Geriatrix:
Using their equation, Levitin and Toffoli calculated that, for every unit of energy, a perfect quantum computer spits out ten quadrillion more operations each second than today's fastest processors.
Click to expand...

A working quantum computer implies mature nanotechnology (atomic engineering, so the quantum computer is very small – invisible to the naked eye). A ‘fog’ of quantum computers will be the computing paradigm of the future. A computing task is performed. The future operating system will ‘poll’ the millions of quantum computers who have individually calculated the task. Majority wins. The threshold for the ‘majority’ can be set by the user. A default value comes with the operating system (I hope Microsoft is defunct by then). Pretty fast IMO.
 
Palimino said:
My personal favourite:

The future operating system will ‘poll’ the millions of quantum computers who have individually calculated the task. Majority wins. The threshold for the ‘majority’ can be set by the user. A default value comes with the operating system (I hope Microsoft is defunct by then). Pretty fast IMO.
Click to expand...

I have a problem with this, and that is that computing isn't a democracy. Its foundation is that it is always correct and precise. Besides, your suggestion still requires each computer to perform the calculations in its entirety, so you won't actually have a result quicker.
One could approach it in a manner similar to a super-computer, but then you run into the problems that are the reasons that they don't just add several hundred thousand processors to a system in order to increase its processing power: Overheads.
Communication between cores isn't instantaneous and you have a limited amount of bandwidth between individual cores as well, so although each core may be able to perform vast amounts of calculations per second, getting the processed data from one part of the processor to the other, becomes an issue.
 
Messugga said:
I have a problem with this, and that is that computing isn't a democracy. Its foundation is that it is always correct and precise. Besides, your suggestion still requires each computer to perform the calculations in its entirety, so you won't actually have a result quicker.
One could approach it in a manner similar to a super-computer, but then you run into the problems that are the reasons that they don't just add several hundred thousand processors to a system in order to increase its processing power: Overheads.
Communication between cores isn't instantaneous and you have a limited amount of bandwidth between individual cores as well, so although each core may be able to perform vast amounts of calculations per second, getting the processed data from one part of the processor to the other, becomes an issue.
Click to expand...

Not really. There is a certain amount of butt-covering because I don’t know the future and have made assumptions. Your democracy point; I have assumed that a future operating system will be AI based. With the fuzzy logic dimension, there will be some wrong answers but the ‘majority’ will be right – hence the poll. The remainder of your post makes extrapolations from current technology. I am postulating a totally new computing paradigm.
 
Tpex said:
Ok I have read through the article and it doesn't seem to address that multicore processors can stack speed??? It also comes down pipe lines and pathways doesn't it?
Click to expand...

you can only add so many cores... Basically with multi cores, the more you have, the more limited the data can be transferred between the cores, thus it doesn't scale.

So in other words, a 16 core processor is pretty much the same speed as a 32 core processor. Even with better pipelines, and faster bandwidth.

Also you must remember that multicore processors have an issue with size. it's not hard to slap 8 cores on a chip, is is hard to make those cores fit onto a normal chip, and keep them under the 140watt TDP. This is the major issue that microchips have, they pretty much have to keep under the 140w TDP and also be able to fit on a normal chip! (however, graphics cards don't really have this limit, hence why they are now more powerful then a comparable CPU).

It would be interesting to see where Manufacturers go in the next 10 years...
 
I think Moore's law isn't so much a law as a self-fulfilling prophecy. It gives manufacturers deadlines that they usually meet.

I believe computers will continue to get faster and faster for two reasons.
First, there is so much we don't know about the physical world. I remember a time when people thought the atom was the smallest indivisible piece of matter in the world. Then some guy shot something at it and a whole lot of other crap came tumbling out. We don't know how small matter can get (despite what physicists say, they have been wrong countless times before (think: conventional current)).

And second, there is no limit to how big a computer can be. If the micro-parts can't get any smaller, make the macro-part as big as you want.

I admit there has to be a limit but it is far down the line and we will never know if we've reached it because there is no telling what new discovery could be made.
 
Kaufies said:
I think Moore's law isn't so much a law as a self-fulfilling prophecy. It gives manufacturers deadlines that they usually meet.

I believe computers will continue to get faster and faster for two reasons.
First, there is so much we don't know about the physical world. I remember a time when people thought the atom was the smallest indivisible piece of matter in the world. Then some guy shot something at it and a whole lot of other crap came tumbling out. We don't know how small matter can get (despite what physicists say, they have been wrong countless times before (think: conventional current)).

And second, there is no limit to how big a computer can be. If the micro-parts can't get any smaller, make the macro-part as big as you want.

I admit there has to be a limit but it is far down the line and we will never know if we've reached it because there is no telling what new discovery could be made.
Click to expand...

+1

I think its highly arrogant of humans to think we can predict with certainty whats going to happen +50 years from now. Let alone proclaiming to know for certain all physical limitations and boundaries. Any theory we have about anything, is only valid for as long as 1 observation disproves it....and that has happened sooooo many times
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X