Path: utzoo!attcan!uunet!husc6!bloom-beacon!bu-cs!purdue!i.cc.purdue.edu!j.cc.purdue.edu!pur-ee!uiucdcs!uiucdcsp!gillies
From: gillies@uiucdcsp.cs.uiuc.edu
Newsgroups: comp.sys.mac
Subject: Re: some (should-be) ground-rules f
Message-ID: <76000218@uiucdcsp>
Date: 12 Jun 88 05:29:00 GMT
References: <2758@utastro.UUCP>
Lines: 22
Nf-ID: #R:utastro.UUCP:2758:uiucdcsp:76000218:000:1082
Nf-From: uiucdcsp.cs.uiuc.edu!gillies    Jun 12 00:29:00 1988


I have been reading the compress source code extensively, and the
original paper by Terry Welch.  Here are some of the many non-standard
modifications to "vanilla" LZW:

(1) adaptive reset to restart compression at a certain threshold
    (suggested, but not standard).  I've read the code, and I suspect 
    the reset may always be arriving 100% too late, but I'm not positive.
(2) variable-length codes, minimum = 9 bits, max = 16, increasing
    as the compression goes on.  "Vanilla" LZW is 12 bits (constant).
(3) some kind of special processing for low-density bit images.
    This may be a significant performance hit -- it may significantly
    overuse (abuse?) the adaptive reset [HOG_CHECK code].

Someone said we should standardize compress because it's decent and
nearly every UNIX machine has it.  Well, I believe that IBM PCs are
decent and nearly every business has one.  So while we're at it, why
don't we just give up on the macintosh altogether?  This would save
even more time!

Don Gillies {ihnp4!uiucdcs!gillies} U of Illinois
            {gillies@cs.uiuc.edu}