Path: utzoo!attcan!uunet!ginosko!usc!polyslo!cosmos.acs.calpoly.edu!mdeale
From: mdeale@cosmos.acs.calpoly.edu (Myron Deale)
Newsgroups: comp.graphics
Subject: Re: SigGraph Fractal Compression
Message-ID: <13665@polyslo.CalPoly.EDU>
Date: 18 Aug 89 04:58:17 GMT
References: <444@mit-amt.MEDIA.MIT.EDU> <20400001@inmet> <1934@uceng.UC.EDU>
Sender: news@polyslo.CalPoly.EDU
Reply-To: mdeale@cosmos.acs.calpoly.edu.UUCP (Myron Deale)
Organization: ACS, Cal Poly, San Luis
Lines: 29

In article <1934@uceng.UC.EDU> mfinegan@uceng.UC.EDU (michael k finegan) writes:
>rich@inmet writes:
>>Remember decompressing really means applying fractal equations over and over 
>>again (like in painting a Mandelbrot).  The amazing thing is that he was 
>>"decompressing" the pictures at video rate: 22 pics per second.  
>
>    While MIPS and MFLOPS are different animals, the DSP16's could also be
>used in parallel, and at 820 MIPs the DCT methodology would yield ~54
>pics per second (compress or uncompress). This makes the fractal decompression
>look pretty good (1/2 performance of current methods); what about fractal
>compression? I was under the impression that encoding the image using 'fractal
>analysis' took many orders of magnitude longer than the decoding ...
>				      -	Mike Finegan
>					mfinegan@uceng.UC.EDU

   Hey, not to mention the Inmos A300 (or some such) DCT chip. Supposed
to be rated at 320 MIPS, because it has something like 8 mult/acc's on-chip.
I can't remember if I read it right or not, but they may have mentioned
performing the transform fast enough to handle real-time video rates,
eg. 20MHz (chip clock rate). Cost was around $80 (1000's) expected to go
down to $25.

   Not to put down the "fractal compression" technique, but it seems to
need more development, and be more open to public and/or scientific
scrutiny. On the flip side, DCT is a known.


-Myron
// mdeale@cosmos.acs.calpoly.edu