Re: DES: left "circular shift" of key bits
- From: Mike McNally <mmcnally@xxxxxxxxxx>
- Date: Sat, 25 Nov 2006 12:14:07 -0500
On Sat, 25 Nov 2006 06:09:44 +0000 (UTC), ggr@xxxxxxxxxxxxxxxx (Greg Rose) wrote:
I think from your discussion that you're
misunderstanding something. The key is *56* bits.
Each round key is 48 of those bits selected the
same way each time. But it's the 56-bit key that
gets rotated.
Hey Greg, the '48 bits' part was a typo on my part, where it should have been 56. I'm
just beginning in learning modern cryptography and am using DES as my first algo
learning platform. By the way, after being stripped of every 8th 'parity checking'
bit, it is the 56 bit key that gets rotated, as you said. But that 56 bit key first
gets split into two 28 bit halves ('L'eft and 'R'ight), and it is the 2 halves that
both get their own separate rotation before being integrated back together into a 56
bit key... after 16 splits, rotations, and reintegrations generating the 16
subkeys...next on to Permuted choice 2 where it will be permuted into a 48 bit key
for XORing, etc.
Ok, now are you impressed with my learning! Ha, just kidding :)
I'm a beginner and am now with the E, S and P boxes inside the feistel-Chiffre
function. This is the most fun a person can have frying their brain! heh.
.
- References:
- DES: left "circular shift" of key bits
- From: Mike McNally
- DES: left "circular shift" of key bits
- Prev by Date: Re: Interesting paper: On the Power of Simple Branch Prediction Analysis
- Next by Date: Re: Programs for breaking Classic Ciphers
- Previous by thread: Re: DES: left "circular shift" of key bits
- Next by thread: sorry guys
- Index(es):