in news:cja### [at] netplexaussieorg Christopher 
James Huff wrote:

> I'm not really sure what you're thinking of here, but it seems 
> to be the other way around...a blob pattern generated on the fly the 
> same way the crackle pattern chooses its points.
> 

That's it yes, with the radius of the blob component determined by the 
(shortest) distance of the center of the component to the enclosing box.

Thinking of it, maybe it is useful to  makes these crackle centeroids 
available to the user through an internal function (f_centeroids()), 
although I have no clue how to use them ...

Ingo

Post a reply to this message