[479] in Cypherpunks
Re: cypto + compression
daemon@ATHENA.MIT.EDU (smb@research.att.com)
Fri May 21 17:43:27 1993
From: smb@research.att.com
To: Stanton McCandlish <anton@hydra.unm.edu>
Cc: cypherpunks@toad.com
Date: Fri, 21 May 93 17:23:32 EDT
OK well if you encrypt a compressed file, there are bound to
be lots more new redundencies created in the encryption
process (unless it is something like ROT-13). Why not
compress this AGAIN, squeezing more space out o
Actually, you've got it backwards. A decent encryption algorithm,
including DES, generates something with very little redundancy,
and hence which cannot be compress further.
$ compress </vmunix >vmunix.Z
$ des -e -k foo <vmunix.Z >vmunix.Z.des
des: WARNING: using software DES algorithm
$ compress <vmunix.Z.des >vmunix.Z.des.Z
-rwxr-xr-x 1 root 1875490 Jan 7 16:59 /vmunix
$ des -e -k foo </vmunix | compress >vmunix.des.Z
des: WARNING: using software DES algorithm
$ ls -l /vmunix vmunix*
-rwxr-xr-x 1 root 1875490 Jan 7 16:59 /vmunix
-rw-rw-r-- 1 smb 794374 May 21 17:17 vmunix.Z
-rw-rw-r-- 1 smb 794376 May 21 17:18 vmunix.Z.des
-rw-rw-r-- 1 smb 1066555 May 21 17:18 vmunix.Z.des.Z
-rw-rw-r-- 1 smb 2538235 May 21 17:21 vmunix.des.Z
As you can see, compressing after encrypting *increases* the size of
the file.
