X-Message-Number: 15094
Date: Fri, 08 Dec 2000 12:59:37 +0100
From: Henri Kluytmans <>
Subject: quantum dot computer <> quantum computer

I wrote : 

>>Please don't call this a quantum computer. This creates
>>misconceptions. A real quantum computer is a computer
>>which operates using quantum logic (superpositions of
>>qubits).

Joseph Kehoe wrote :

>According to the article that is what it is
>quantum dots used to store qubits!
>
>The idea is to use their technique to set up arrays of qubits.
>If they can combine that with the work of other researchers in 
>interrogating qubits then we are well on the way to quantum >computing.

I was sloppy, I didnt read the article. Now I've read it...

However Im still not convinced. The article is not clear about it.

A quantum dot architecture alone does not imply a quantum computer 
(using quantum logic). All descriptions for quantum dot architectures 
I've seen so far were not real quantum computers. 

I've done a survey on the net, and still couldnt find any architecture 
using quantum dots to perform quantum logic.

I found this quote :

"Although Lent et al. used the term "quantum cellular automata" to describe
their model, the term was already in use to describe a wholly different
method of computing. The term seems to have been coined by Groessing in
1988. Groessing's quantum cellular automata are quantum computers in the
same sense as Benioff's. Unlike quantum computers, where are all the
"qubits" are quantumly entangled together, distant cells in Lent's QCA are
not coupled. Although the cells switch via quantum mechanical means,
neighboring cells are coupled through classical Coulombic interaction."

The quantum dot architecture the article in EETimes is referring to 
is Lent's QCA. And this architecture does not perform any quantum 
logic. (Otherwise all cells would need to be coupled.) Lent's QCA 
architecture is equivalent to a conventional cellular automata 
hardware architecture.

(i.e. The terms : "quantum dot cellular automata" and "quantum 
cellular automata" are not equivalent! )

For more information about Quantum Dots Cellular Automata see :

http://www.nd.edu/~qcahome/

=====

I think the person who wrote the article mixed things up. However
this Bandyopadhyay is also (maybe on purpose?) not being very 
clear. And he even made a huge mistake :

In the article I found this quote from Bandyopadhyay :

"You could build a quantum computer that has 2 to the 1,000th bits 
of data, which you could never do with a binary computer, because 
the number 21,000 is larger than the number of atoms in the known 
universe. But a quantum computer could store that many bits with 
just 1,000 atoms," said Bandyopadhyay. 

This is plain wrong !!!

You cannot store 2^1000 bits of data in 1000 qubits.
You can only store a superposition of 2^1000 possible 
combinations of 1000 bits in a 1000 qubits. And then 
you can perform operations on all those 2^1000 
combinations at once in one single operational step.
This is a hugely big fundamental difference.

Quantum computers are especially interesting for cracking public 
key systems, because they can find the right secret key 
in only one operation (because they can operate on all possible 
combinations at once), they dont have to scan all possible 
keys step by step (as conventional computers have to do).
They can also be used for computing some other special problems.

However secret key systems cannot be cracked this easily 
using quantum logic! 

The breaktrough, that will follow when serious quantum 
computers can made, is highly exaggerated.


You also wrote :

>Quantum computers nearer. If we can get these working then a 
>lot of speed and space problems will simply disappear!!

Quantum computers will not solve space problems. Because 
you cannot store a 2^1000 bits of information in a 1000 
qubits.

Quantum computers will only solve speed problems for some 
special cases (like factoring a product of two large primes).


Grtz,
>Hkl

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=15094