Re: Chaos Theory - Yes!

Anton Blanchard ((no email))
Thu, 18 Jul 1996 12:23:13 +1000 (EST)

> > I'll just type it again in case you think it was a typo - 50:1

> > This compression method can be used (I have heard) to compress data files
> > up to 200 times smaller ie. you could fit a whole encyclopedia onto a floppy
> > disk. The only problem is that the algorithm to extract the real data is
> > almost purely iterative and you could be waiting a long time for a 100k
> > file to decompress to a 5 megabyte file.

> Sorry, this wouldn't work (IMHO). There are theoretical limits to how far
> you can push lossless compression (which is what you want for data, not
> images). An encyclopedia isn't fractal in nature (self similar at
> different scales), hence would not compress using this technique.

I agree. Shannon theory places restrictions on the maximum compression
ratio you can get with a perfect algorithm. You go past this and you start
losing information.

If you were to identify 1 thing in 8 using a binary code, it would take
3 bits. I'd like to see it done in less, on average. (this is what 200:1
compression suggests to me)

Anton

-- 
Anton Blanchard                         anton@nospam.progsoc.uts.edu.au
Programmers' Society CSO and            anton@nospam.lister.flex.com.au
Computer Systems Engineering Student    http://www.progsoc.uts.edu.au/~anton/
University of Technology
Sydney, Australia

OK, quark brain.