Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
The Health of the RC5 Project
2 views
Skip to first unread message
Eric Cordian
Sep 11, 2002, 10:20:56 PM9/11/02
to
If anyone's interested, I did a quick back of the envelope estimate a
couple weeks ago of the chances that distributed.net's Bovine RC5 project
will exhaust the remainder of the keyspace without finding the key, as a
function of the amount of keyspace thusfar searched, and the confidence
that the project software is error-free.
couple weeks ago of the chances that distributed.net's Bovine RC5 project
will exhaust the remainder of the keyspace without finding the key, as a
function of the amount of keyspace thusfar searched, and the confidence
that the project software is error-free.
http://www.inet-one.com/cypherpunks/dir.2002.08.26-2002.09.01/msg00148.html
Abstract: They should wait until they are over 99% before sticking their
heads in the oven.
--
Eric Michael Cordian 0+
O:.T:.O:. Mathematical Munitions Division
"Do What Thou Wilt Shall Be The Whole Of The Law"
Eric Young
Sep 12, 2002, 3:45:04 AM9/12/02
to Eric Cordian
Many years ago I was involved in one of the first 'challenges',
searching the SSLv2 RC4-40 keyspace.
I started at 1/2 way through on a network of about 100 486/50,
and another person used a maspar to search from 0.
I exausted my range and the maspar got a hit at about %48.
So between us we search %98 of the key space before getting the hit,
so don't give up until you hit %100 :-)
A particular french student beat us by a few hours.
searching the SSLv2 RC4-40 keyspace.
I started at 1/2 way through on a network of about 100 486/50,
and another person used a maspar to search from 0.
I exausted my range and the maspar got a hit at about %48.
So between us we search %98 of the key space before getting the hit,
so don't give up until you hit %100 :-)
A particular french student beat us by a few hours.
eric
Scott Contini
Sep 12, 2002, 12:43:09 PM9/12/02
to
Eric Cordian <e...@artifact.psychedelic.net> wrote in message news:<alotm8$1brq$1...@news.hal-pc.org>...
> If anyone's interested, I did a quick back of the envelope estimate a
> couple weeks ago of the chances that distributed.net's Bovine RC5 project
> will exhaust the remainder of the keyspace without finding the key, as a
> function of the amount of keyspace thusfar searched, and the confidence
> that the project software is error-free.
>
> If anyone's interested, I did a quick back of the envelope estimate a
> couple weeks ago of the chances that distributed.net's Bovine RC5 project
> will exhaust the remainder of the keyspace without finding the key, as a
> function of the amount of keyspace thusfar searched, and the confidence
> that the project software is error-free.
>
A related paper is here:
Philippe Golle and Ilya Mironov, "Uncheatable Distributed
Computations," Topics in Cryptology – CT-RSA 2001, pp. 425–440, 2001
Available at:
http://crypto.stanford.edu/~mironov/research.html
Scott
Eric Braeden
Sep 13, 2002, 12:18:33 PM9/13/02
to
I have long feared that someone was spoofing the blocks
completed. While I have not looked at the code. It seems
obvious that it must be computationally less expensive to
spoof the work completed than to actually do it.
completed. While I have not looked at the code. It seems
obvious that it must be computationally less expensive to
spoof the work completed than to actually do it.
Time will tell.
Scott Contini
Sep 13, 2002, 7:52:11 PM9/13/02
to
"Eric Braeden" <bra...@erinet.com> wrote in message news:<3d820fc3$0$1427$272e...@news.execpc.com>...
Yes: and the method in the paper I referenced above is a very
easy way to more or less defeat spoofers. But let's hope that
the key is still to come! :o)
Scott
0 new messages