X-Message-Number: 6130
From:  (Tad Hogg)
Date: Tue, 30 Apr 1996 17:38:18 PDT
Subject: SCI.CRYONICS  Re: Quantum Computers

In article <4lfhqm$>, John K Clark <>
writes:

> When a conventional 64 bit single processor computer performs an
> operation, it does it on ONE 64 bit number at a time. When a 64
> bit (actually a 64 qubit) single processor QUANTUM computer
> performs an operation it does it on ALL 64 bit numbers at the
> same time, all 2^64 of them, more than a billion billion,
> and any increase in the number of qubits the computer can handle
> will increase it's already astronomical power exponentially.

Even if these machines can be built reliably, there's the problem of how
to program them. Operating simultaneously on ALL 64 bit numbers is the
easy part for these machines! The real trick is to get a useful answer
from the computation. This requires "measuring" the quantum state.
Unfortunately this just gives you the result for ONE number, chosen
randomly with probabilities determined by the final state. This
measurement destroys the quantum state, so no further information is
available on the rest of choices.

Programs must use a second aspect of quantum computers, interference, to
combine different ways to compute the each result. Properly done this
gives a high probability to get a desired SINGLE answer from among ALL the
choices. But this requires clever design of algorithms, e.g., to allow
different ways of computing undesirable results to "cancel" each other
out.

So far as I know, only two specific algorithms using interference have
been proposed -- one for factoring integers and another, more complex, one
for general search problems. A lot more work is needed to invent more
algorithms and see how well they might work when subject to errors and
noise.

Further information is available at
   ftp://parcftp.xerox.com/pub/dynamics/constraints.html#quantum
including pointers to other quantum computing pages at Stanford and
Oxford.

While it is too soon to tell if quantum computers will be practical, they
could substantially increase the computer power available for determining
the correct way to repair frozen tissue. Additional discussion of the
computational issues of repair is at
   http://merkle.com/merkleDir/cryptoCryo.html


Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=6130