On 12/19/2010 6:08 PM, stbenge wrote:
> On 12/19/2010 1:37 PM, Patrick Elliott wrote:
>> On 12/18/2010 8:00 PM, stbenge wrote:
>>>> Frankly, I am no where near good enough to even guess how to do that.
>>>> lol
>>>
>>> Sometimes exploration will let you stumble across a structure you had
>>> intended. Other times, you need a system suited for what you are doing.
>>> For instance, cellular automata can be "grown" from a substrate, and
>>> bounds can be made for it, but the end result is still something
>>> unintended. Fairly recently someone invented a snowflake generator which
>>> produces what appears to be physically-accurate results. Really cool,
>>> but it takes days to grow the things :(
>>>
>>> Sam
>>
>> Well, I meant in the specific sense of being able to "find" a way to
>> constrain it. Someone needs to come up with something like, I don't
>> know, a biologic constraint type thing, which can kind of "switch off"
>> growth, in a way that would still produce the fractal result, but not
>> necessarily something like looking for a specific shape in Mandelbrot,
>> but getting not just *that* shape, but the whole rest of the things you
>> didn't want too, or something.. Like I said, I know what you would need
>> to make it more predictable in terms of size, or even shape (like you
>> might get in some cases where the form is part of an object, but only
>> applies to say, a leg, or something), but actually having a clue how you
>> manage that is way beyond me.
>
> You want controlled unpredictability... something organically grown, but
> unexpected? Things like the Mandelbrot fractal aren't 'grown', but the
> result of a complex number squared and iterated... They can be
> constrained, but not without global consequences... Kaleidoscopic IFSs
> are sort of grown, but not really... Maybe you need L-Systems? Or
> perhaps something involving genetic evolution with fractal qualities?
> Perhaps if you can define your problem to a finer degree, I can help
> point the way to what you want...
>
> There are Framsticks, and there is Spore, and there are plenty of
> L-Systems out there with genetic evolutionary properties...
>
> Sam
Yeah. That is often the problem in a nutshell. Defining the problem. 
Thing is, living things "have" controlled unpredictability, as you put 
it. They can be constrained by what they grow on, in, around, how much 
"fuel" they have to develop, and even their own "code", which can do 
things like give them extra toes, if a stop codon had a mutation. 
Fractals have the coincidence of often generating a lot of complex 
shape, but they are an "unconstrained" version of the same math that 
generates what we see in the real world, usually. Not always, but 
usually. In the case of something like Mandelbrot, you can *find* just 
about any thing you could ever imagine in there, but, its like trying to 
create a beach, and having to manufacture a whole universe along with 
it, because the rules you have to work with don't let you just produce 
that *one* beach. If you see what I am saying.

And, its not always easy to find a subset of rules that produce the same 
result.

-- 
void main () {

     if version = "Vista" {
       call slow_by_half();
       call DRM_everything();
     }
     call functional_code();
   }
   else
     call crash_windows();
}

<A HREF='http://www.daz3d.com/index.php?refid=16130551'>Get 3D Models, 
3D Content, and 3D Software at DAZ3D!</A>

Post a reply to this message