Notices
Results 1 to 16 of 16

Thread: Quantum computing

  1. #1 Quantum computing 
    Forum Freshman
    Join Date
    Oct 2013
    Posts
    9
    Would it be possible to create a quantum computer, even if only of a limited sort, by developing software algorithms based on quantum logic and hardware logic chips that could handle these? In other words could there be a software/hardware route to quantum computing short of using entangled quantum particles? If not, why not?


    Reply With Quote  
     

  2.  
     

  3. #2  
    Brassica oleracea Strange's Avatar
    Join Date
    Oct 2011
    Location
    喫茶店
    Posts
    16,998
    You can simulate quantum computing systems and algorithms. And people do. The problem is that without an actual quantum computer to run the algorithms on, they will be no faster and probably slower than conventional algorithms.

    The point about a quantum computer is that it uses superposition to effectively perform many (potentially) all possible calculations of a problem in parallel (simplifying enormously, because I don't understand the details myself). If you don't have superposition, then you have to do all those operations sequentially and you are no better off than a standard computer.


    ei incumbit probatio qui dicit, non qui negat
    Reply With Quote  
     

  4. #3  
    Forum Freshman
    Join Date
    Oct 2013
    Posts
    9
    Yes, I understand the requirement for superposition. What I don't understand is why this requirement can not be satisfied through software making use of quantum logic. Present day computers use bit logic, but why could not a computer be constructed using qubit logic with superposition but without quantum particles?
    Reply With Quote  
     

  5. #4  
    Brassica oleracea Strange's Avatar
    Join Date
    Oct 2011
    Location
    喫茶店
    Posts
    16,998
    Quote Originally Posted by Mandala View Post
    Yes, I understand the requirement for superposition. What I don't understand is why this requirement can not be satisfied through software making use of quantum logic. Present day computers use bit logic, but why could not a computer be constructed using qubit logic with superposition but without quantum particles?
    Well, one way of doing it, I suppose would be to implement multiple "shadow" bits for each bit in the word. So, for example, a 32 bit word could be, say, four bits "deep". This would then allow each word to take on multiple (16 in this case) values at the same time.

    This is a novel approach to parallel computing that I haven't seen tried. I think this is because the hardware costs would be too high. Every register and datapath in the processor would have to be 4 bits deep as well. The memory (and its data buses) would need to be 4 times as large.

    So, I suppose it could be done, in principle, but would be expensive and hard to implement. It could be an interesting project to build a small version.
    ei incumbit probatio qui dicit, non qui negat
    Reply With Quote  
     

  6. #5  
    Forum Freshman
    Join Date
    Oct 2013
    Posts
    9
    I wonder if part of the problem here is not related to the fact that computer science has been built on the binary mathematics developed by Leibniz based on 1s and 0s rather than on the Taoist yin-yang system of +1s and -1s from which Leibniz took his formulation, changing it in the process. If a computer were based on +1s and -1s two different zeros fall out naturally depending upon whether +1 changes to -1 or -1 to +1. To me it seems this should offer an easy way to get a handle on qubits without great expense or difficulty. Just a different sort of logic based on a different approach to binary arithmetic. But then I am neither a mathematician nor a computer scientist. Just a layman who doesn't always respect entrenched authority without good reason.
    Reply With Quote  
     

  7. #6  
    Brassica oleracea Strange's Avatar
    Join Date
    Oct 2011
    Location
    喫茶店
    Posts
    16,998
    Quote Originally Posted by Mandala View Post
    I wonder if part of the problem here is not related to the fact that computer science has been built on the binary mathematics developed by Leibniz based on 1s and 0s rather than on the Taoist yin-yang system of +1s and -1s from which Leibniz took his formulation, changing it in the process.
    I have never heard of Leibnitz being credited with the invention of binary (nor a connection with Taoism). Boole and, most importantly, Shannon are the usual suspects.

    If a computer were based on +1s and -1s two different zeros fall out naturally depending upon whether +1 changes to -1 or -1 to +1.
    You then need two bits to represent every value. And, off the top of my head, I can't see any advantage. There have been computers that used different encoding schemes for numbers but, in the end, everyone comes back to binary for simplicity and flexibility.
    ei incumbit probatio qui dicit, non qui negat
    Reply With Quote  
     

  8. #7  
    Forum Freshman
    Join Date
    Oct 2013
    Posts
    9
    Quote Originally Posted by Strange View Post
    I have never heard of Leibnitz being credited with the invention of binary (nor a connection with Taoism).
    The modern binary number system was discovered by Gottfried Leibniz in 1679 and appears in his article:Explication de l'Arithmétique Binaire[1](1703). Leibniz's system uses 0 and 1, like the modern binary numeral system. As a Sinophile, Leibniz was aware of the Yijing (or I-Ching) and noted with fascination how its hexagrams correspond to the binary numbers from 0 to 111111, and concluded that this mapping was evidence of major Chinese accomplishments in the sort of philosophical mathematics he admired. [Binary number - Wikipedia, the free encyclopedia]
    See also Explanation of binary arithmetic.
    Reply With Quote  
     

  9. #8  
    Brassica oleracea Strange's Avatar
    Join Date
    Oct 2011
    Location
    喫茶店
    Posts
    16,998
    Interesting. I have never come across that before. I wonder if Boole et al. were familiar with his work on this...
    ei incumbit probatio qui dicit, non qui negat
    Reply With Quote  
     

  10. #9  
    Forum Freshman
    Join Date
    Oct 2013
    Posts
    9
    It seems unlikely that Boole, who published his paper introducing his algebraic system of logic in 1854, would not have been well familiar with Leibniz's work of the late 17th century.

    I am developing a mandalic geometry based upon the I Ching hexagrams and Descartes' coordinate system. This is intimately related to the question at hand. If interested you can check it out at

    http://mandalicgeometry.tumblr.com and

    http://blindmen6.tumblr.com
    Last edited by Mandala; October 22nd, 2013 at 08:09 PM.
    Reply With Quote  
     

  11. #10  
    Forum Freshman
    Join Date
    Oct 2013
    Posts
    9
    You then need two bits to represent every value. And, off the top of my head, I can't see any advantage. There have been computers that used different encoding schemes for numbers but, in the end, everyone comes back to binary for simplicity and flexibility.
    Well, yes. Computing with two bits to represent every value would be more complicated than computing with one bit for each value as we do today. The point is a qubit represents a two-state quantum-mechanical system so necessarily requires more information to designate the state of the system. Take as an example

    . . . polarization of a single photon: here the two states are vertical polarization and horizontal polarization. In a classical system, a bit would have to be in one state or the other, but quantum mechanics allows the qubit to be in a superposition of both states at the same time, a property which is fundamental to quantum computing. [Qubit - Wikipedia, the free encyclopedia]
    But if this could be accomplished by means of a slightly more complex gate logic involving superposition of bits rather than difficult and capricious superposition of quantum particles, what's to complain? Now I don't know that it can, but I think the possibility is worth considering.
    Reply With Quote  
     

  12. #11  
    Forum Isotope
    Join Date
    Feb 2012
    Location
    Western US
    Posts
    2,852
    Quote Originally Posted by Mandala View Post
    But if this could be accomplished by means of a slightly more complex gate logic involving superposition of bits rather than difficult and capricious superposition of quantum particles, what's to complain? Now I don't know that it can, but I think the possibility is worth considering.
    You are missing a rather fundamental point. It's not that such ideas haven't been considered -- they have. Many times. In many proposed implementations. So, it's not as if a simple algorithm running on "slightly more complex" logic would confer the benefits of QC. The part that you are missing is energy per operation. The appeal of QC is that the superposition of states comes for "free." The incremental energy cost is correspondingly low. And thus, the overall energy to perform a computation is -- or should be -- much less than that consumed by an ordinary binary-logic-based alternative.

    When, instead, you mildly perturb the macroscopic hardware of today's machines to mimic QC, the energy doesn't go down. It only goes up. Why? If you want to minimize energy consumption in an ordinary binary system, you lower the voltage until you can just barely discern two states with acceptable error rate. Interpose another level to mimic superposition, and you must raise the voltage to be able to identify that new state unambiguously. That increases energy.

    That's precisely why what you propose is a non-starter. An algorithmic overlay and an incremental tweak of the existing computational fabric will only get us incremental improvements. A true QC needs to operate in the quantum realm. Period.
    Reply With Quote  
     

  13. #12  
    Forum Freshman
    Join Date
    Oct 2013
    Posts
    9
    You are missing a rather fundamental point.
    Thank you for this explanation. That makes sense and I was overlooking energy consumption. I will think on this further and try to digest it. But just to be clear here about what I was suggesting: if there were some way to compute with +1s and -1s rather than 1s and 0s then the "new state" might come free because it is just the sum of +1 and -1 to give 0 (actually two zeros) which would just occur automatically. No "new state" would be introduced externally. Yeah, this sounds crazy. Off the top of my head though I can cite at least a dozen other crazy ideas once considered scientific heresy which are now universally accepted. I'm done now until I read up on energy cost of computing.
    Reply With Quote  
     

  14. #13  
    Forum Isotope
    Join Date
    Feb 2012
    Location
    Western US
    Posts
    2,852
    Quote Originally Posted by Mandala View Post
    if there were some way to compute with +1s and -1s rather than 1s and 0s then the "new state" might come free because it is just the sum of +1 and -1 to give 0 (actually two zeros) which would just occur automatically. No "new state" would be introduced externally. Yeah, this sounds crazy. Off the top of my head though I can cite at least a dozen other crazy ideas once considered scientific heresy which are now universally accepted. I'm done now until I read up on energy cost of computing.
    Sorry, but that won't work either, but nice try. To understand the fundamentals of the problem, you can't think of these levels as noiseless. Think of them explicitly as noisy values. Then adding them won't produce zero, it will produce noise. The +1 and -1 states have to be separated by enough to tell them apart from each other, even in the presence of noise. That precludes being able to detect zero unambiguously.
    Reply With Quote  
     

  15. #14  
    Brassica oleracea Strange's Avatar
    Join Date
    Oct 2011
    Location
    喫茶店
    Posts
    16,998
    Quote Originally Posted by Mandala View Post
    if there were some way to compute with +1s and -1s rather than 1s and 0s then the "new state" might come free because it is just the sum of +1 and -1 to give 0 (actually two zeros)
    I missed this before. You are effectively using two bits (whether as physical bits or different energy levels) to represent 3 values (as there is no benefit to having two zero values). Which is even less efficient in hardware terms.
    ei incumbit probatio qui dicit, non qui negat
    Reply With Quote  
     

  16. #15  
    Forum Freshman
    Join Date
    Oct 2013
    Posts
    9
    You are effectively using two bits (whether as physical bits or different energy levels) to represent 3 values (as there is no benefit to having two zero values). Which is even less efficient in hardware terms.
    Well it seems the point is moot now as it has been put forth that noise would preclude being able to detect zero unambiguously. However if one could do so, a left zero and a right zero would exist and be different depending on whether -1 or +1 led to it. So two qubits would effectively represent 4 values, not 3. The benefit to having two zero values is that it opens a (logic) door to superposition.
    Last edited by Mandala; October 24th, 2013 at 04:13 PM.
    Reply With Quote  
     

  17. #16  
    Forum Freshman
    Join Date
    Oct 2013
    Posts
    9
    One more try because I see I have failed to adequately describe the crucial point here. Current computer technology is based upon a single dimension of digital input. I am suggesting use of two dimensions of input.

    Basically the idea is to use two simultaneous data streams which are separate and independent of one another. These data streams could be either electron-based or photon-based. Each stream individually would be similar to the single data stream of 1s and 0s in current usage, i.e., current flow on = 1, current flow off = 0. But the readout would be made to vary according to the sum of the readouts of the two separate data streams divided by 2.

    If we allow current flow to equal +1 and no current flow to equal -1 then the possible readouts would be [(-1) + (-1)]/2 = -1; [(+1) + (+1)]/2 = +1; [(-1) + (+1)]/2 = left 0; and [(+1) + (-1)]/2 = right 0. This logic arrangement allows for the superposition of on and off states demanded by quantum computing and in fact provides two different superimposed on/off states.


    I just can't see why we couldn't do this. A little electronic or photonic noise should not be an insurmountable spoiler here since we've already put a man on the moon and manufactured DVD recording and playback devices that work faithfully. The challenge involved here should be of a lesser degree of difficulty than either of those achievements.
    Last edited by Mandala; October 24th, 2013 at 04:17 PM.
    Reply With Quote  
     

Similar Threads

  1. How far away is personal quantum computing?
    By jakesyl in forum Computer Science
    Replies: 7
    Last Post: August 16th, 2013, 06:11 PM
  2. Replies: 0
    Last Post: November 19th, 2012, 02:55 AM
  3. quantum computing
    By Gnostic_Pneumatic in forum Computer Science
    Replies: 1
    Last Post: April 5th, 2011, 07:11 PM
  4. Quantum Computing
    By CarpeDiem in forum Computer Science
    Replies: 12
    Last Post: February 19th, 2010, 06:45 AM
  5. Quantum computing and the future of cryptography
    By Jarek Duda in forum Computer Science
    Replies: 30
    Last Post: May 8th, 2009, 05:28 AM
Tags for this Thread

View Tag Cloud

Bookmarks
Bookmarks
Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •