Heck no. Atomic computers like the one described by the article are probably going to be used in really tiny computers. Quantum computers would be much larger, if not huge. Quantum computers need to block out a lot of interference or they will not be able to accuratly function. As of now, radiation shielding involves lotsa really heavy metals, wax, etc. Error correction can only go so far before it becomes not worth it anymore. Also Quantum computers do not already have the functions done, but they are helpful for solving computationally intensive (nasty hard) problems. Right now, DNA computers are also very good in CI problems. Scientific American had a good article on them about a year ago.
Yeah, a cupful of future computers may have the power of all of the processors today, but just imagine the system requirements of Win3k! "MOM, I need another 10 terrabytes of L1 cache or quake will be choppy!":) Then again, my lousy Cyrix desktop system probably has more power than all of the computers just 20 years ago. I remember when 16k of main memory was excessive (TI/99/4A).
On a side note, does anyone really believe that nanotech is that amazing? Look at your cells. Nature has been there and done that, probably more efficiently than we will ever be able to do it.
What I would like to see in the meantime is chip technology moving into a three dimensional arrangement. As it stands now, most chips are basically flat. Imagine the power that could be built into a chip that is not 1cm^2 but 1cm^3 in area! (I worry about heat dissapation though).
Hate to burst your bubble, but they already thought of it, and are using it (in limited amounts) in current processors and chips. Heat is always a problem, as well as parasitic currents causing latch-up problems. There should be examples in VLSI books. I am nowhere near any of mine right now, so I can't give an example. Anyway, making a large circuit like this would be a pain in the butt to design, as well as fabricate. I couldn't see this kind of tech going too far anywhere in the future (with current materials)
What you are saying now is why I would do c before cpp. You can learn all of the basics with C, and then move onto C++ seamlessly. I think it is better to pick up a language and learn 90% of it before moving on. This way one can see the different powers and ideas that went behind forming the language in the first place. I learned TI BASIC, BASIC, PASCAL, C, x86 assembly, 8051 assembly, scheme, C++, ADA, then PERL, in that order. When I got to perl, I could really appreciate the speed at which it would execute simple scripts, and sometimes not so simple scripts! It is powerful, easy to use, platform independentish, and actually does what you want it to do without having to make a lot of functions that do the "real work" in a program. (String manipulations, data structure overhead, etc.) Most of the time, I just want a program to work and be done with it. Perl is great for that, but a beginner should learn structure, orginization, and form before they get into the lazy programming style. Learning perl as a first language would make a large programming effort spaghetti code that would never run quite right.
Slashdot Mirror
PREACH IT LEONARD!!!!!!!
Heck no. Atomic computers like the one described by the article are probably going to be used in really tiny computers. Quantum computers would be much larger, if not huge. Quantum computers need to block out a lot of interference or they will not be able to accuratly function. As of now, radiation shielding involves lotsa really heavy metals, wax, etc. Error correction can only go so far before it becomes not worth it anymore. Also Quantum computers do not already have the functions done, but they are helpful for solving computationally intensive (nasty hard) problems. Right now, DNA computers are also very good in CI problems. Scientific American had a good article on them about a year ago.
:) Then again, my lousy Cyrix desktop system probably has more power than all of the computers just 20 years ago. I remember when 16k of main memory was excessive (TI/99/4A).
Yeah, a cupful of future computers may have the power of all of the processors today, but just imagine the system requirements of Win3k! "MOM, I need another 10 terrabytes of L1 cache or quake will be choppy!"
On a side note, does anyone really believe that nanotech is that amazing? Look at your cells. Nature has been there and done that, probably more efficiently than we will ever be able to do it.
What I would like to see in the meantime is chip technology moving into a three dimensional arrangement. As it stands now, most chips are basically flat. Imagine the power that could be
built into a chip that is not 1cm^2 but 1cm^3 in area! (I worry about heat dissapation though).
Hate to burst your bubble, but they already thought of it, and are using it (in limited amounts) in current processors and chips. Heat is always a problem, as well as parasitic currents causing latch-up problems. There should be examples in VLSI books. I am nowhere near any of mine right now, so I can't give an example. Anyway, making a large circuit like this would be a pain in the butt to design, as well as fabricate. I couldn't see this kind of tech going too far anywhere in the future (with current materials)
I was the AC that posted the flame bait before.
What you are saying now is why I would do c before cpp. You can learn all of the basics with C, and then move onto C++ seamlessly. I think it is better to pick up a language and learn 90% of it before moving on. This way one can see the different powers and ideas that went behind forming the language in the first place. I learned TI BASIC, BASIC, PASCAL, C, x86 assembly, 8051 assembly, scheme, C++, ADA, then PERL, in that order. When I got to perl, I could really appreciate the speed at which it would execute simple scripts, and sometimes not so simple scripts! It is powerful, easy to use, platform independentish, and actually does what you want it to do without having to make a lot of functions that do the "real work" in a program. (String manipulations, data structure overhead, etc.) Most of the time, I just want a program to work and be done with it. Perl is great for that, but a beginner should learn structure, orginization, and form before they get into the lazy programming style. Learning perl as a first language would make a large programming effort spaghetti code that would never run quite right.