Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!know!sdd.hp.com!wuarchive!mit-eddie!media-lab!turk@media-lab.media.mit.edu
From: turk@media-lab.media.mit.edu (Matthew Turk)
Newsgroups: comp.graphics
Subject: Re: Next machine as animation platform (Image compression)
Message-ID: <4031@media-lab.MEDIA.MIT.EDU>
Date: 14 Nov 90 20:36:58 GMT
References: <85866@tut.cis.ohio-state.edu> 	<JIM.90Nov13163758@baroque.Stanford.EDU> <OLSON.90Nov13231633@sax.cs.uiuc.edu> <velasco.658597478@beowulf>
Sender: turk@media-lab.MEDIA.MIT.EDU
Reply-To: turk@media-lab.media.mit.edu
Followup-To: comp.graphics
Organization: MIT Media Lab
Lines: 13
In-reply-to: velasco@beowulf.ucsd.edu's message of 14 Nov 90 15:44:38 GMT

> This is not necessarily true for all forms of compression.  Straight
> run length encoding preserves every bit of detail in the original
> image.  Delta modulation can come pretty close at a high enough
> sampling rate. ...

However lossless compression techniques for static images can't
perform anywhere near 37X compression.  More like 6X or 8X, at best.
I *think* the JPEG standard has a lossless option that performs
somewhere around that.  And by the way, remember that these numbers
are for color images -- the numbers for gray scale images are about 3X
worse.

	Matthew