Path: utzoo!utgpu!watserv1!watmath!att!pacbell!pacbell.com!ucsd!usc!zaphod.mps.ohio-state.edu!uakari.primate.wisc.edu!aplcen!haven!uflorida!stat!sun13!sun8.scri.fsu.edu!nall
From: nall@sun8.scri.fsu.edu (John Nall)
Newsgroups: comp.os.minix
Subject: Comic
Message-ID: <228@sun13.scri.fsu.edu>
Date: 10 Jul 90 12:46:22 GMT
Sender: news@sun13.scri.fsu.edu
Reply-To: nall@sun8.scri.fsu.edu (John Nall)
Organization: Supercomputer Computations Research Institute
Lines: 26

This may be of interest to some, so I'll post it.  (We have
a fairly aggressive version of Pnews here, which makes one
feel guilty if one posts something which may not be of general
interest.  It keeps asking questions like "Are you *sure* that
people want to see this?" and "This costs money, you know.  Are
you *sure* you should do it??")

The fairly significant increase in the compression ratio of
Comic is attractive, so I played with it some on our Cray Y-MP
to see how it does there.  It is horrible!  (In terms of speed.
Worked OK, though, with no changes, under Unicos).

The compression is good enough, however, so that it would bear
some further investigation.  Two possibilities arise (which are
not necessarily non-exclusive):  (a) vectorizing the code, and
(b) profiling it and then going to assembly language for those
parts which show up as chewing up all the execution time.

If anyone else is doing the same thing, I would be interested
in their results.


--
John W. Nall		| Supercomputation Computations Research Institute
nall@sun8.scri.fsu.edu  | Florida State University, Tallahassee, FL 32306
     "Los dioses eran computadores.  ?Que otra cosa podian ser??"