Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!brl-adm!adm!W8SDZ@SIMTEL20.ARPA
From: W8SDZ@SIMTEL20.ARPA (Keith Petersen)
Newsgroups: comp.os.cpm
Subject: Compression schemes and no source code
Message-ID: <803@brl-adm.ARPA>
Date: Sat, 15-Nov-86 09:31:55 EST
Article-I.D.: brl-adm.803
Posted: Sat Nov 15 09:31:55 1986
Date-Received: Sun, 16-Nov-86 00:41:26 EST
Sender: news@brl-adm.ARPA
Lines: 27

> I had retrieved a copy of <CPM.CCP>conix.lbr and found many of
> the files had .DZC or .CZM extensions.  Can you tell me what
> compression scheme was used to squeeze these files?

> Also, if there is some handy CP/M-80 utility to unsqueeze them,
> could you pass its name along as well?

Yes, there is a program called CRUNCH (and companion UNCRunch) which
does Lempel-Zev-Welch style file compression on CP/M-80.  It is far
more efficient than SQueeze/UnSQueeze which uses the Hoffman
algorithm.  The efficiency of CRUNCH exceeds that of MSDOS`s
ARC512.EXE utility.

[Flame on] - Too bad the authors of LU and NULU never released the
source code.  We **COULD** have had a combined LBR/CRUNCH which would
have been better than MSDOS's ARC files.  Another stifling of software
expansion/development caused by those who will not share the basic
tools we need to keep CP/M alive. [end of flame].

Filename		Type	 Bytes	 CRC

Directory PD:<CPM.SQUSQ>
CRUNCH20.DAT.1		ASCII	  1984  E4B5H <--performance info
CRUNCH22.LBR.1		BINARY	 26880  FAB1H <--Z80 crunch/uncrunch
						 (no source code available)

--Keith