Tim May writes:
Not to split universes here, but it is interesting to consider that some ciphers may not be breakable in _our_ universe, in any amount of time.
Our universe presumably has some finite number of particles (currently estimated to be 10^73 particles). This leads to the "even if every particle were a Cray Y-MP it would take..." sorts of thought experiments.
But I am considering _energy_ here. Ignoring reversible computation for the moment, computations dissipate energy (some disagree with this point). There is some uppper limit on how many basic computations could ever be done with the amount of free energy in the universe. (A rough calculation could be done by calculating the energy output of stars, stuff falling into black holes, etc., and then assuming about kT per logical operation. This should be accurate to within a few orders of magnitude.)
The above analysis may be incorrect... there may be no limit to the amount of computation that can be done with a given finite amount of energy. The late Nobel laureate Richard Feynman became very interested in the subject of computation and physics towards the end of his life. My understanding is that he concluded that there was no apparent limitation to the amount of computation that could be completed with a given amount of free energy. Computation may indeed always dissipate energy, but Feyman's conclusion was that this dissipated energy can be made arbitrarily small -- that there is no fundamental quantum limitation on the amount of computation that can be performed at any given mass-energy scale. The kT per logical operation can always be reduced to finer and finer scales. Presumably, this would require advances to ever new technologies, based on new physical forces that are relevant at finer scales (down to computation based on the interactions of quarks as in QCD, gravitons, etc.) Of course, since I can't give you references, you have to take this with a brick of salt... can anyone else comment on whether they have heard this about Feynman's conclusions? This is distinct from the issue of "quantum computers" and Shor's recent results... that issue has to do with whether quantum mechanics can be used to produce *qualitatively* different types of computation. In the above, I am simply discussing the use of quantum mechanical principles to produce fully "classical" computers, but with every greater computational powers using a given amount of energy, based on physics of the ultra-small. In fact, classical computers today rely on quantum mechanics, as the transistor cannot be described without it (electron tunneling, etc.) __ __ __ __ Doug Cutrell / ) /__) /_ /\ / /| /| / /\ / / ) doug@OpenMind.com \_/ / (_ / \/ / |/ | / / \/ /__/ ===================================================================