Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!lll-crg!caip!princeton!allegra!ulysses!mhuxr!mhuxt!houxm!ihnp4!cbosgd!cbatt!cbdkc1!cbnap!whp
From: whp@cbnap.UUCP (W. H. Pollock x4575 3S235)
Newsgroups: net.micro.mac
Subject: Re: The "Next Mac" (bitmap conpression)
Message-ID: <161@cbnap.UUCP>
Date: Wed, 13-Aug-86 11:22:05 EDT
Article-I.D.: cbnap.161
Posted: Wed Aug 13 11:22:05 1986
Date-Received: Fri, 15-Aug-86 20:04:20 EDT
References: <5482@topaz.RUTGERS.EDU> <946@unirot.UUCP> <1109@utastro.UUCP> <5963@sun.uucp>
Reply-To: whp@cbnap.UUCP (W. H. Pollock x4575 3S235)
Distribution: net
Organization: AT&T Bell Laboratories, Columbus
Lines: 21

In article <5963@sun.uucp> chuq@sun.uucp (Chuq Von Rospach; Lord of the OtherRealms) writes:
>> I was reading an post on the local bulletin board that rumored Apple had come
>> up with an algorithm for shrinking paint documents in a 31:1 ratio.  Imagine
>> a 31K picture saved as 1K on disk.  Wow.
>
>Wow is right.  The rumor is almost definitely wrong.  Compression technology
>of a random bitstream is lucky to get 50% compaction. The compress program
>used for passing news around averages between 35% and 45%.

Ahh, but bitmaps are not "random bitstreams"!  I read an article in Abacus
last month which describes a method to compress bitmaps to 0.18 bits/pixels,
which is an 82% gain. (The authors also claimed 2.2 bits/character for text).
The method involves adoptive modeling and dynamic Markov models.

Wayne H. Pollock,
UUCP:	...{ihnp4,cbatt}!cbnap!whp
DELPHI:	WHP
GEnie:	W.POLLOCK

	"The opinions expressed above are ficticious.  Any resemblance
	to the opinions of persons living or dead is purely coincidental."