Re: A92: compression

To: assembly-92@lists.ticalc.org
Subject: Re: A92: compression
From: Famille SCHERRER <p.c.scherrer@wanadoo.fr>
Date: Sun, 04 Apr 1999 13:49:02 +0200
Delivered-To: assembly-92-outgoing@towerguard.unix.edu.sollentuna.se
Delivered-To: assembly-92@lists.ticalc.org
In-Reply-To: <199904032013.NAA27770@smtp2.asu.edu>
References: <Version.32.19990403184811.00de1f00@pop.wanadoo.fr> <86a564af.3703d3f6@aol.com>
Reply-To: assembly-92@lists.ticalc.org

At 13:11 03/04/99 -0700, you wrote:
>
>>I made a LZW compression algorithm a long time ago for the 
>>TI92. I am now decided to use it to make a PCTOOLS98
>>compression plugin.
>>But first I will make a compression plugin using the huffman
>>algorithm in ziplib
>
>Unless I'm confused, huffman is more efficient than LZW, particularly for
>text or similar data with some characters that repeat more often than others.
>--Cliff Biffle

No I don't think. I compare myself compression ratio between
my routine and the huffman one from huffman.exe (when
there was not ziplib yet) and each time it compressed a
little better. 

The LZ method was made by Lempel and Ziv in 1978.
Welch terminates it in 1984, and created the LZW algorithm.

According to a french web site, the LZW is the best method. It
is the more famous one, but it is not the most used, because
of expansive copyright. That's why the ZIP or ARJ files
only use LZ method.

LZ? (LZ77, LZ78... ) is also used by RAR or ARC achives,
or GIF files.

Benoit

References:

Re: A92: compression

From: Famille SCHERRER <p.c.scherrer@wanadoo.fr>

Re: A92: compression

From: NDStein@aol.com

Re: A92: compression

From: Cliff Biffle <cbiffle@safety.net>