Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!ucbvax!agate!gandalf.Berkeley.EDU!ofer
From: ofer@gandalf.Berkeley.EDU (Ofer Licht)
Newsgroups: comp.sys.amiga.tech
Subject: Wanted: specs for Eric Graham's Movie anim file format
Summary: Wanted: "J-compression" algorithm for anims
Message-ID: <22556@agate.BERKELEY.EDU>
Date: 3 Apr 89 22:02:20 GMT
Sender: usenet@agate.BERKELEY.EDU
Reply-To: ofer@gandalf.Berkeley.EDU (Ofer Licht)
Distribution: usa
Organization: Statistics Dept., U. C. Berkeley
Lines: 28



Does anyone out there have the specifications to Eric Graham's movie
anim file format?  Is his "J-compression" algorithm available for
perusal?
--------
I am now able to make the movie package (dilbm, pilbm, movie) function
consistently (read: no gurus) and find the resulting anims to be 
significantly faster than those created by makeanim,grabanim,etc.
The movie anims, however, take longer to create and require more storage
space.  The storage and speed factors suggest that "J-compression" is
not quite as good as Jim Kent's "vertical encoding" compression used in
ANIMs of type 5.  Then why do the movie anims take so much longer to compute?
--------
I am willing to sacrifice the storage requirements of movie anims but would
like to speed the process up a bit.  Currently, I produce animation frames
in ilbm format on a network of suns (using DBW render ported back to UNIX and
a custom local network job distributer).  I download the frames and process
them into an anim on the amiga.  I would like to be able to process
the frames into an anim with a "J-compression" algorithm running on a sun4.
If you have any suggestions for other compression algorithms, email me.

Note: anyone interested in my ray-tracing setup on the suns is welcome to 
	a copy of all sources (including the network job distributer and
	the modified DBW render package).

Ofer Licht
ofer@gandalf.berkeley.edu