Newsgroups: comp.compression,sci.image.processing
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!nntp.sei.cmu.edu!cis.ohio-state.edu!math.ohio-state.edu!howland.reston.ans.net!news.sprintlink.net!news.clark.net!rahul.net!a2i!bug.rahul.net!a2i!olivea!charnel.ecst.csuchico.edu!csusac!csus.edu!netcom.com!tgl
From: tgl@netcom.com (Tom Lane)
Subject: Re: HARC Compression claims... oversights, or designed to mislead??
Message-ID: <TGL.95Jul29075247@netcom18.netcom.com>
Sender: tgl@netcom18.netcom.com
Organization: Netcom Online Communications Services
References: <3vcb31$6j6@net.auckland.ac.nz>
Date: Sat, 29 Jul 1995 14:52:47 GMT
Lines: 22
Xref: glinda.oz.cs.cmu.edu comp.compression:22144 sci.image.processing:16162

yensoy@Glue.umd.edu (Sridhar Srinivasan) writes:
> Whether by design or by oversight, some of the claims made by Houston
> Advanced Research Center (HARC) with regard to their image compression
> technology are highly misleading.  The only information they are willing
> to divulge at this point to the public is contained in their www page
> (http://www.harc.edu/).
> [ much snipped ]
 
Yeah, I wondered about that too.  The last I looked at HARC's web page,
it was possible to see some compression artifacts on the small thumbnail
image of the "300:1" case.  But those artifacts were *not* visible on
the larger GIF it was linked to.  I think they have the wrong GIF linked
there.  At best, very sloppy work; at worst, deliberate deception.

In any case, reducing and color quantizing the presumably-full-color
test images is certain to obscure any compression artifacts pretty
thoroughly.  Thus these "sample" GIFs are 100% useless for evaluating
HARC's technology.  If they really wanted to make believers out of us,
they should put up full-size full-color images, as TIFFs or PNGs perhaps.

			regards, tom lane
			organizer, Independent JPEG Group
