Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!samsung!olivea!uunet!ogicse!zephyr.ens.tek.com!wrgate!copper!michaelk
From: michaelk@copper.WR.TEK.COM (Michael D. Kersenbrock)
Newsgroups: comp.compression
Subject: Re: Z80 compression
Message-ID: <6054@wrgate.WR.TEK.COM>
Date: 22 May 91 21:35:11 GMT
References: <7674@ecs.soton.ac.uk>
Sender: news@wrgate.WR.TEK.COM
Organization: Tektronix
Lines: 35

In article <7674@ecs.soton.ac.uk> jmss89@ecs.soton.ac.uk (Scott JMS) writes:
>Does anyone have any algorithms / assembler to implement file compression for the
>Z80 CPU ?
>
>I could do a port of stuff for 68000, but with 64k memory the hash tables tend to
>gobble it all up - therefore I'd prefer something that didn't need hash tables
>
>Please post to addresses below:
>
>Thanks in advance
>
>  Jon Scott
>
>------------
><jmss89@uk.ac.soton.ecs>
>        OR
> <jon@uk.ac.leeds.geog>
>------------                    :-)


I once implemented 16-bit uncompress for the Z80 (CP/M).  Even with heavy
optimization and tricks (and with the "speed path" enhanced by "dropping"
into assembly language), it still was VERY slow (output was a good bit
faster than you could read, but it wasn't a blur either).  My Z80 system
had 1MB of RAM (but currently resides in a cardboard box in the garage
somewhere).

You *really* don't want to do it unless you'll be satisfied with a much
smaller encoding.  And it'll still likely go sluglike.

--
Mike Kersenbrock
Tektronix Logic Analyzer Division
michaelk@copper.WR.TEK.COM
Aloha, Oregon