Re: Hey - Livermore knows how to crack 3DES!

From: Anon E. Maus (anon_at_maus.duh)
Date: 06/29/03 

Date: Sun, 29 Jun 2003 13:07:48 GMT

Still, I'll keep using 3DES for at least a couple more years anyway. The
length + (length % 8) padding is better than % 16 padding, especially for
UDP. Small stream packets are bad enough... miilions of packets per minute
constitutes a major payload overhead.even with % 8 padding. CPU horsepower
isn't much of an issue, but traffic payloads are. Even a couple of years
isn't going to diminish this issue, unfortunately. Rather the opposite!

Back to the point of this thread (I digress) if a crack is possible every 45
minutes, a key change would be prudent at a faster rate. We can also
shuffle coding sequences (EDE, EEE, DED, etc.) without any impact to the
issues mentioned above. In additio to coding sequences we can also shuffle
the key sequences (321->123, etc.) all within 8 bits. Hey - 64 - 8 = 56...

"Runu Knips" <I@hate.spam> wrote in message news:3efc3104@news.celox.de...
> AFAIK Bruce Schneier said (immediately after Rijndael was voted for
> being AES) there will be very likely theoretical breaks of Rijndael in
> some years. That is, the 128 bit key will then be only 125 or 120 bits,
> effectively.
>
> Nothing to worry about.
>