[rc5] Some Deep questions
dmoerenh at reference.be
Mon Aug 18 15:37:19 EDT 1997
What they describe there is something I taught of too. Probably many of
us did. Especially the fact that if you make a crack-chip you can
easily crack this thing in a short period. It's like people with a K5
noticing they've got a very cheap but still fast crack-chip. Chipmakers
should be well capable of making like standard-8086 compatible chips that
are capable of doing 2MKeys/s each. Those chip would use very little
transistors and can't cost more than 10$. 10.000 of those do 20Gkeys/s
and that's 6 times the current speed of Bovine and Cyberian combined.
Governments are of course well capable of spending more than $100.000 as are
many companies. Spending $1.000.000 would give 200Gkeys/s or 60
times the combined speed of Cyberian and Bovine. When using real
dedicated chips made just to crack rc5 you would get better results. Of
course you 've to pay for the design too but when spending over
$1.000.000 that can't be a problem and you would get more bang for the
buck. Imagine that $10.000.000 could give you something like 5000Gkeys/s...
maybe even more.
On Mon, 18 Aug 1997, Arvin Meyer wrote:
> > From: Brian Foddy <bkfoddy at skypoint.com>
> > To: rc5 at llamas.net
> > Subject: [rc5] Some Deep questions
> > Date: Sunday, August 17, 1997 6:01 PM
> > I've been thinking about this whole process.
> > And there are some questions I have.
> Anyone interested in the feasibility of how easy it is to crack a 56 bit
> algorithm check out:
> where you'll find that a dedicated *government* sponsored effort can take
> as short as 12 seconds. A description of what it would take to do that,
> and how, can be found there written by some very knowledgeable
> cryptographers and mathematicians.
> Arvin Meyer
> onsite at esinet.net
> To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.
More information about the rc5