Re: Someone said 256 bits is not enough


It depends. What are your assumptions?

*Because even if you were doing the best you could, irreversibly on a digital processor,...

*So it's not possible on a digital machine or a practically possible network of them.

It sounds like your friend was listening to a proponent of quantum
computing - some of them often put out such "facts".

I was trying to describe my assumptions as "not a quantum computer".

Ok. Here's something I don't understand about quantum computing...
I've heard that they can reduce a symmetric cipher to square root
complexity but something about that doesn't make a lot of sense. I
understand that they work on superposition and all. It just seems
like ... well, I can't really put it succinctly right now so I'll give
you an example.

Ugh.. well.. I read a lot about DNA computing where they cleverly
craft complimentary strands of DNA to model their problem and they go
together to form an exhaustive set of strings over the problem set and
then apply what are essentially filters to keep strings that match
what they expect from a valid solution and then they let them incubate
for a while or something to let those probably-more-correct strings
duplicate. I kind of see quantum computing doing the same thing.
You've got your superposition, an exhaustive set of strings over the
problem space, and you basically apply an optimization algorithm;
applying filters that keep waveform constituents that agree with what
you expect to be characteristic of the correct answer. I don't know
if you can amplify the signal after that or not, but you end up with
an answer or subset of answers that's increasingly more probable while
other answers' probabilities converge with zero...

I could be fundamentally off base here. I'm just not sure how that
extends to practice. I guess what I'm saying is if you are going to
universally reduce symmetric ciphers to square root complexity, that
includes ciphers that use the key to generate their internal states
and that use data dependent functions and stuff. I guess you're
already generating the key schedule in superposition.

When you say you're attacking with a quantum computer and that
complexity is reduced to square root, what are you assuming (if
anything) about the algorithm itself? How do you narrow in on what is
a more probably correct answer from a field of random looking stuff?
Does the quantum computer actually reverse the steps the algorithm
took..? I guess I'm actually in the dark about the pretty much all
the details of the implementation. So yeah, got any reading material
or references or anything?

2**128 is still a lot. Does this, in your opinion Mr Eather, warrant
a move to 512 bit keys or kilobit keys? I mean if we're facing such
eminent threats, why take chances, right?
.

Relevant Pages

  • Re: Someone said 256 bits is not enough
    ... computing - some of them often put out such "facts". ... Here's something I don't understand about quantum computing... ... then apply what are essentially filters to keep strings that match ... and you basically apply an optimization algorithm; ...
    (sci.crypt)
  • Quantum Computer Algorithms
    ... internal state (where these operations behave as in classical physics). ... time differed from computing device to computing device...) ... Our current best quantum mechanical ... which produce the same result as classical algorithms but ...
    (sci.physics.research)
  • For physicists and computer persons (but not exclusively)
    ... Subject: Quantum Security ... quantum entities known as qubits. ... computing capabilities. ...
    (uk.business.agriculture)
  • Re: The logic of atheism
    ... Now go search that site for any references to quantum ... however, just like with cryptography, the demand for computing ... >> water molecule (three small atoms) than they are with a beta ... let the experiment run, the better your predictions. ...
    (talk.atheism)
  • Re: Problem demonstrating that the set of binary strings is uncountable.
    ... > algorithm for computing it can't really exist? ... Computability has nothing to do with the definition of uncountability. ... > must be in the list and there is no algorithm for modifying it ... strings that all algorithms can produce. ...
    (sci.math)
Quantcast