> Obviously you know something about fractal IMAGE compression. There is such a
> thing as fractal DATA compression too!!! It is only hard to imagine because
> you can't visualise it. It has been successfully been used before - I have
> heard of a game (can't remember the name) which used fractal DATA
> compression to store all the adventures you could go on.
So when is a fractal not a fractal? Whan it's a rabbit, of course. Okay,
so you've redefined the fractal? That's not what you said yesterday.
I'm not talking about image compression. I talking about the applicability
of fractals to data. It just isn't the suitable compression method. Now,
instead of saying "well you're just talking about images", why don't you
now say "go and have a look at this URL where you can be enlightened about
fractal compression and its application to data". If you believe there is
a case here for fractal compression to be used generally for data
compression, then let's see it!
What's so hard about visualising fractal data compression? It's like image
compression, except the patterns are not represented graphically. That
doesn't mean that the concept cannot be visualised.
> For the rest of you who are interested and want to see an example of fractal
> IMAGE compression at a ratio of 100:1 without downloading any new software,
> there is an example at:
>
> http://www.mitsubishi.co.jp/jp/fractal/perfor.html
>
> It shows you two images - the first one was compressed with JPEG and the
> second was compressed with fractal compression. Both were compressed at a
> rate of 100:1 - you don't really need a comparison between the two to be
> amazed.
Nothing amazing here. Two lossy compression methods that display different
artefacts when pushed passed their limits. One cartographer says "I've
created a new projection method. Distances are correct, as are areas, but
for some reason, Antarctica looks like a cat and Australia, like a whale!".
The other says "Mine look okay, except that Greenland looks about 5 times
the size of Australia". Which one's better, considering neither are accurate?
Granted, we don't think this one is solvable, so we use different methods for
different applications. However, don't try to make out as if you've got
this new fantastic, whizz bang method. It still has major drawbacks, and
they are not necessarily any better than those experienced by other
compression methods (although the artefacts may seem to have more
accentuated artefacts).
I'll stop here for now as I think you might have the gist of it....
Peter
------------------------------------------------------------------------
Peter Meric pmeric@nospam.socs.uts.edu.au
pmeric@nospam.progsoc.uts.edu.au
pmeric@nospam.acs.itd.uts.edu.au
"Stupidity cannot be cured with money, or through education, or by
legislation. Stupidity is not a sin, the victim can't help being
stupid. But stupidity is the only universal capital crime; the sentence
is death, there is no appeal, and execution is carried out
automatically and without pity."
Robert Heinlein