[rc5] yet more rc5-64 vs Something Else

Daniel Baker dbaker at wrangler.cuckoo.com
Wed Oct 1 22:24:21 EDT 1997

> >> Daniel Baker <dbaker at wrangler.cuckoo.com> wrote:
> >> Having gracefully made our point about encryption, let's retire from that
> a
> >> bit (unless someone can suggest another algorithm that merits attention -
> >> perhaps some of those used in cellphones or some such?) and find
> something
> >> to do. Ideally, I'd like to see several different projects happening in
> >> parallel, possibly including rc5-64, but let's not limit ourselves to the
> >> very sterile (IMHO) exercise of breaking an algorithm a second time just
> >> because we can.
> >
> >Right, but then we're saying "56 bit is hard, so 64 is enough".  All we're
> >saying is "we've giving up when the going gets tough".  We need to continue
> >proving the fact that we can break encryption standard after encryption
> >standard.
> As far as I'm concerned 64-bit IS enough for what I use encryption for.  I
> don't think any of us can dictate to anyone else how they should interpret
> the results of our effort.  Whether or not we try the RC5-64bit contest (I
> won't) we have shown that it takes months to find a single key for a single
> tranmission encrypted with RC5-56.  I don't know about you, but the most
> sensitive thing I will probably ever protect with encryption is my credit
> card number and I only have a $10,000 limit so I doubt it would be worth it
> for anyone to spend several times that for the effort needed to find my
> 56-bit key, let alone my 64-bit key.  On the other hand, the prize for this
> contest is only $10,000...so I could be wrong.
> I think we've given enough data to whoever wants to know about how much
> effort is required to find a 56-bit key that they could extrapolate to
> figure out what would be required to find a 64-bit key.  We wouldn't really
> be providing any new or useful information, except possibly how easy or hard
> it is to recruit additional computing power.

This is with free clients on IDLE cpus running regular operating
systems.  For _very_ little money and a little research, a company
or government group could crack 64 bit in a few weeks.


Daniel Baker (dbaker at cuckoo.com)
Chief Technology and Executive Officer -- CuckooNet! (http://www.cuckoo.com)
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.

More information about the rc5 mailing list