Re: Basically a sieve method, relation to quantum

jstevh_at_msn.com
Date: 01/22/05 

Date: 22 Jan 2005 08:42:17 -0800

Durk van Veen wrote:
> jstevh@msn.com wrote:
>
>
> >
> > The original algorithm in my program, will, my current analysis
shows,
> > factor about 50% of the time, which is astounding.
> >
>
> Wouldn't trial and error division find the factors 100% of the time?
Now it
> might take a while (maybe longer than the universe takes to reach
> heat-death) but the question is whether your algorithm would be
faster or
> not.

My algorithm is in polynomaial time.

It is significantly faster than what you're suggesting.

I did one test calculation of the number of combinations and found that
a number made up by multiplying the first 1,000 primes together would
give only around 100 million combinations.

That is, with a number that huge, my algorithm would probably loop
through all the combinations on a decent pc in a couple of hours.

In comparison, it could factor an RSA type number, if the theory is
correct, and all the implementation details can be worked out, in
seconds.

Make no mistake, we can talk here as if it's nothing, but if the basic
theory is right, then someone else may already be successfully breaking
Internet encryption as we speak, with some fancy program or programs on
a distributed system that can factor VERY large numbers in a few
seconds, based off the math.

If my research holds up, and I think it will, then the basic algorithm
itself, which is already out there, will factor an RSA type number 50%
of the time, with maybe a few million combinations to iterate through
for any given j that you try.

That's why this work should be taken seriously.

You can talk about it, or downplay it or whatever, but if it's correct
and someone less silly than you people pays attention, they can start
walking through Internet security IMMEDIATELY as in right now, not
later, not in a few months, but RIGHT NOW.

James Harris