[RC5] Historical perspective

Chris Arguin Chris.Arguin at unh.edu
Fri Nov 21 16:45:24 EST 1997

On Fri, 21 Nov 1997, Robert Brooks wrote:

> Out of curiosity, I wonder what a distributed supercomputer like ours would
> have been able to do with the "uncrackable" mechanically encrypted messages
> of the WWII era?  Has anyone subjected those old messages to '90s
> computerized decryption to see how the science has evolved... surely
> there's a web site out there somewhere.

I imagine that Enigma would go pretty quickly... Probably a matter of
minutes on an average PC.

> It's interesting, though, that part of our decryption method depends on the
> 16 or so characters of known text at a known position in the message we're
> trying to break.  What about the real-world scenario of a message encrypted
> first with a simple cypher (even a quickie rot-13), then run through
> RC5-64?  How long would it have taken to crack RC5-56 if we didn't know any
> of the contents of the message -- if we had had to apply the additional
> step of looking for common English-language letter frequencies and patterns?
> (quick, someone tell me it would have only been a matter of time!)

Certainly. At the least, you save the partially decrypted versions for
human perusal. Slow, and error-prone, but simple.

Along the same lines... Is there an encryption technique in which the
given the encrypted message, you could generate any decrpyted message,
by varying the key? What I mean is, if you had the message
"IUhiosahfi879*^*(^hs;", and you tried brute force on it, you could get
back (amoung a lot of garbage) "Sell my stock", "Don't sell my stock", and
"Millenium Hand and Shrimp"

It just seems to me that would make a good encryption technique.

Chris Arguin          | "Multiple exclamation marks are a sure sign of a 
Chris.Arguin at unh.edu  |  diseased mind." - Eric, Terry Pratchett

To unsubcribe, send 'unsubscribe rc5' to majordomo at llamas.net
rc5-digest subscribers replace rc5 with rc5-digest

More information about the rc5 mailing list