Re: DES & RC4-48 Challenges

On Mon, 21 Aug 95 23:13:06 -0400 you wrote:

>
>Hello all,
>
>	Lets face it the real challenge is DES, but those 56 bits are quite a 
>bit harder than 40. 65536 times harder in fact.
>
>40 bit code a tad and set to work. It sounds as if it is comparable to the 
>RSA-129 prime that was cracked (OK its probably a touch harder but machines are 
>faster now).
>
>	So who wants to make an RC-48 and a DES challenge?

Not having my copy of The Differential Cryptanalysis of the Data
Encryption Standard handy, I'd like to know about the distributability
of this type of DES attack.  Done right, we could significantly reduce
the time complexity.
	The main problem, of course, would be coordinating such an effort.  I
seem to recall this attack requiring lots of known plaintexts.  Time
to review the text, I suppose....
	Does the Federal Reserve still use single-key DES?
						Dan  
******************************************************************************
"I think, therefore I am" - Descartes                            Dan Bailey
"I don't think, therefore I'm a moustache." - Sartre		    dan@milliways.org
Worcester Polytechnic Institute and The Restaurant at the End of the Universe
******************************************************************************

Dan wrote:

|Not having my copy of The Differential Cryptanalysis of the Data
|Encryption Standard handy, I'd like to know about the distributability
|of this type of DES attack.  Done right, we could significantly reduce
|the time complexity.
|	The main problem, of course, would be coordinating such an effort.  I
|seem to recall this attack requiring lots of known plaintexts.  Time
|to review the text, I suppose....
|	Does the Federal Reserve still use single-key DES?

The forms of differential cryptanalysis that I'm aware of require The
cracker to adaptively atack the encrypting or decrypting device. I
therefore do not believe that they are especially applicable to
financial transactions schemes, most of which change keys quite
frequently.

JWS

I don't think the differential Cryptanalysis attack helps very much. You
need a somewhat improbable quantity of chosen plaintext to attempt it.

You would get further with a direct attack I suspect. In any case lets do 
RC4-48 or RC4-44 first as proof of ability.


Another idea, lets crack CMDF, IBMs weakened key DES.

After DES the only other crack of interest thats in range is the meet in the 
middle attack on two key DES. Clipper looks a bit off in the distance still.
(We will get it one day, probably not until 2010 though).

I think we can probably do DES within two years.

	Phill