[RC5] [ADMIN] DES-II-2 is over...

Pete Krawczyk petek at mc.net
Fri Jul 17 10:09:29 EDT 1998

At 03:49 AM 7/17/98 -0500, Alan Beberg wrote:

>In a press release (http://www.eff.org/descracker.html) the Electronic
>Frontier Foundation (EFF) announced today that it had beat DES-II-2
>using a machine it developed for US$250,000 designed specifically to
>crack DES.

I'd like to point out that the EFF claims that their key was found
at approximately 25% of the keyspace, or after searching over 17
quadrillion keys.  This is approximately the same point that
distributed.net is at now.

Which means someone should crunch the "winning" block today.

If that's true, then also the point should be made that this
computer is apparently 2 times as fast as the whole d.net machine
at this point.

One thing I'd like to bring up, though, is that they may have beat us
on DES-II-2, and we can certainly challenge them in the future.  But
I would like to see a coordinated "attack" on the remaining crypto
standards.  The attack on DES is certainly powerful.  But imagine
what would happen to the d.net RC5 attack if we effectively
tripled our effort.  Working with the EFF, we can hand them huge size
blocks (probably like 2^50 blocks) and have them crunch them.
And if they get the right key, we can donate 8 of the 10k to them -
for I'm sure if EFF tripled our power, people would be more inclined
to vote for them as the $6k recipient.

And if you read the press release, they said the DES cracking was
software based.  Thus, we should be able to get them an RC5 core.

+- Alan Beberg said on July 9 in the State of d.net address:
| For the week ending July 4, 98, we did 49.3 million blocks for an
| average keyrate of 21,883 Mk/s. 

Assuming the EFF machine is twice as fast as the d.net machine,
this adds 43 Gk/s to our effort.  Yow.

A standard 2^30 block (1,073,741,824 keys) to the EFF would take 
1/43 of 1s.  A 2^40 block (1,099,511,627,776 keys) would take 25
seconds.  A 2^48 block (281,474,976,710,656 keys) would take
about 1.8 hours.  And a 2^56 block (7,205,759,4037,927,936 keys)
would take 19.4 days.  To research the whole keyspace, the EFF 
would need 13.6 years.

Let's look into this and prove to everyone that standard 56 bit and
64 bit cyphers just don't cut it anymore.  Even a 72 bit cypher, which
is the next step after 64, looks "weak" at this point - not now, but
in a couple years from now.

-Pete K
Pete Krawczyk  pkrawczy at uiuc dot edu -or- petek at mc dot net
 http://www.uiuc.edu/ph/www/pkrawczy  Finger for PGP public key
If you attempt to mail me at pkrawczy at mc.net, I will not get it.
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest

More information about the rc5 mailing list