aboutsummaryrefslogtreecommitdiff
path: root/examples/readme-parameter-ranges.pd
blob: 27fe5f0ad051d508f61c0e9dff01275032251957 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
#N canvas 0 22 499 399 10;
#X text 33 275 The "classification" of an attractor set uses the full
range of the fractals acceptable parameter ranges. As this method of
"classification" is relative \, it will not describe each fractal uniquely
\, but rather can be used to compare attractor sets that have similar
(closely related) parameter values. NOTE: there needs to be a way to
increase the granularity of the classification system...;
#X text 35 21 Parameter Ranges - Once you have an operational fractal
external \, it is important to make sure that the assigned ranges for
each of the parameters are "optimized" to limit the random number generator
to those ranges. This is important when a variable can only range between
(-1 .. 1) and you have declared that it can range from (-100 .. 100).
Because of the much wider search space \, there is less likelyhood
of finding anything useful \, or at least make the search times longer
due to higher failure rates.;
#X text 35 147 If you are unsure of what the ranges are \, you can
either figure them out mathematically \, or use the "brute-force" method
and iterate over ranges of the defined parameters. The points that
return a fractal (even ones that converge) will give an indication
as to the acceptable ranges for each param. Once you have determined
those ranges \, adjust them in your '*.frac' file and re-make the external
for your fractal. (Or you can just edit the Macros in the C code \,
if you are comfortable with that.);