[RC5] RC5 question

Elektron elektron_rc5 at yahoo.ca
Sun Oct 5 21:52:23 EDT 2003


On Sunday, Oct 5, 2003, at 19:06 Asia/Hong_Kong, Richard Menedetter  
wrote:

> Hi Bruce!
>
> 04 Oct 2003, "Bruce Wilson" <bwilson at distributed.net> wrote:
>
>  BW> The point for RSA to offer the contest is
> THIS is clear :)
> Customers fear that somebody has (partly) solved the factorization  
> problem, or
> came up with something much better than brute force.
> The contest helps RSA to calm these fears down.

RC5 encryption has nothing to do with primes (That's RSA encryption,  
which confusingly has the same name as the company, which IIRC is the  
initials of the people who invented it). The contest we're doing  
(RC5-72) is the secret key challenge [1].

There is, actually, an RSA factoring challenge [2], where you factor a  
576 to 2048-bit number into two primes (I believe) [3], but d.net  
currently doesn't participate in this one (and I'm not sure how well  
this works for distributed computing).

The 576-bit number is approximately 2^575.6, and since it's not prime,  
a factor must be less than it's square-root, approximately 2^287.8, so  
it must be under 2^288. The number of primes in the first N natural  
numbers is approximated by N/ln(N), with N=2^288 we get over 2^280  
primes.

That's a lot of primes.

- Purr

[1] http://www.rsasecurity.com/rsalabs/challenges/secretkey/index.html
[2] http://www.rsasecurity.com/rsalabs/challenges/factoring/index.html
[3]  
http://www.rsasecurity.com/rsalabs/challenges/factoring/numbers.html  
e.g. for the 576-bit one,  
188198812920607963838697239461650439807163563379417382700763356422988859 
715234665485319060606504743045317388011303396716199692321205734031879550 
656996221305168759307650257059



More information about the rc5 mailing list