Recreational Programming.

[Ryan_Burnside]

Hello!

I've casually browsed this forum for some time enjoying the various projects people have been working on.
As a programmer, I wanted to provide a little place for people who do recreational programming to give some algorithms in pseudocode.

I've got a few effects and fractals already documented.
It is not strictly for fractals we have things like Perlin Noise too.

It is mostly for programmers.
Take a look, I'd welcome some more pseudocode and effects. I don't know many fractals beyond Lindenmayer Systems.
Thanks for your consideration.

You might want to bookmark the site, we don't have the easiest host name.
If you would like to join please do, there may be a bit of a wait as I approve your account. ( I do sleep a few hours each day)
http://pixeloutlaw.smakweb.com/forums/

P.S. Moderator:
Hopefully this post does not break the rules here regarding website solicitation, if so please lock/delete it and you have my apologies.

[cKleinhuis]

hello, and thank you for your kind request

recreational programming ... here are some hints:

- animate sine dots by mixing up different frequency waves ( classic sine dot effect in oldschool demos )
- program in an 8 bit assembly language and solve various problems, e.g. like clearing the screen on a c64 emulator
- perlin noise is a good starter, but quite high maths ... do you have a game of life ?
- game of life can be extended to the classic fish-shark simulation, where sharks eat the fishes and depending on the rate they extinct or create waves ov more sharks or more fishes ...
- random game - iterated function systems for creating the fern fractal
  - can then be easily extended to flame fractals
- basic towers of hanoi shouldnt be missing on such a site


so far, best wishes for your project!

[cKleinhuis]

i see many topics i suggested you have already
the pseudo code notation is great to get an algorithm ....
the mandelbox is equally elegant algorithm to the mandelbrot, might worth a pseudo code notation as well
just google for it

[Ryan_Burnside]

Thanks for the further thoughts cKleinhuis!

I can't believe I forgot Conway's Game of Life... :

[David Makin]

Something related to fractals and similar:

Assume you have a random generator with a given distribution - could be just 1D or 2D or 3D - and then assume you have 2 sets of data say each generated from a different seed and you want to morph from set A to set B. Now if using a pseudo-random you can't just smoothly change the seed so an initial idea is use result = s*An + (1-s)*Bn where An and Bn are the members of the sets and s varies from 1 to 0. Unfortunately this results in a change in the distribution e.g. if both linear like a single dice they end up as both with the distribution of a 2-dice roll when s is 0.5.
We want to change the combination so that if the two original sets are linear then the set at s=0.5 is also linear as are all resulting sets for s=0 to s=1.
To simplify the algorithm assume we have sets of data with all values from -1 to 1 and our mix is s where we want an undisturbed distribution as s varies.
Now obviously in this case our max range is always -1 to 1 and the central value is always 0.
Using:
  r = 2.0*(s*An+(1-s)*Bn)
  if r<-1
    r = -2-r
  else if r>1
    r = 2-r
  endif
Should give us morphing with much reduced change of character/distribution.

Oh look - folding

Edit: Using the above if there are n possible results in each source set of linear distribution i.e. such that the probability of each result is 1/n then using the above fudge to try to keep linear distribution in the resultant combined set results in n-3 values with probability (n+1)/n*n, two with probability (n+1)/2*n*n and one with probability (n+2)/n*n i.e. total probability =  ((n-3)*(n+1)+2*(n+1)/2+(n+2))/n*n i.e. (n^2-2*n-3)+n+1+n+2)/n*n = (n^2-2*n-3+2*n+3)/n*n = 1.

[Ryan_Burnside]

David, I've been busy work on attractors today but your code looks like it produces some interesting results.
Thanks

Here is my implementation of "Orderly Attractors" where each point in a list seeks out the next in the list while leaving a trail behind. The last seeks the first forming a loop.
More stuff like this to come folks.