Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!husc6!mit-eddie!mit-amt!bc
From: bc@mit-amt.MEDIA.MIT.EDU (bill coderre)
Newsgroups: comp.graphics
Subject: Re: Chaotic Compression
Message-ID: <1743@mit-amt.MEDIA.MIT.EDU>
Date: Fri, 13-Nov-87 14:38:51 EST
Article-I.D.: mit-amt.1743
Posted: Fri Nov 13 14:38:51 1987
Date-Received: Sun, 15-Nov-87 09:50:25 EST
References: <619@applix.UUCP>
Reply-To: bc@media-lab.MEDIA.MIT.EDU (bill coderre)
Organization: MIT Media Lab, Cambridge MA
Lines: 21

About Hypercard compression:

First off, it is soon to be patented, so you will be able to read the
algorithm soon. No one is currently allowed to talk about it, for the
specific reason that this might invalidate the patent claims.

Second, Hypercard "cards" are 1 bit pelmaps 512x342.

Third, compression peaks at 30:1, for "normal" images. Thus, a card,
which is 28k uncompressed, averages 1k-2k compressed.

Fourth, "standard" compressions (like they use in FAX transmission)
average 5:1 or maybe even 10:1 if you are lucky.

Fifth, one of the features is that compressed cards are very fast to
uncompress. A MacII can display 5-10 cards per second, including
setting up all the data structures to actually use the card, if the
cards are in memory.

DISCLAIMER: I worked for Apple this summer, and signed those horrible
non-disclosures. I now work for MIT as a graduate student. The above
does not represent the opinions of anybody but me. OK, OK, OK, OK???bc