I was not sure how to tackle the problem with nodes so i wrote it out…
new to all this so I am REALLY hoping there is a more efficient way of doing this.Not only is my way broken, but it gets sluggish with just a few iterations…
My end game is to generate a pattern by randomly tyling the hexagonal patterns (thousands),
and only have the parts of the pattern IN my desired shapes to show.
BTW: The vast majority of people (even “pros”) believe that // is all about finding the magic red button: just press it and get wrap speed. Truth is that there’s no gain without pain. On the other hand and for real-life stuff is worth considering the time required for masterminding a proper scheduling policy: in most of cases is not worthy (unless you are Academic). But since contemporary these days means 666 cores (see the trend in Intel’s newest Coffee Lake generation) … well … get some experience on that matter.
BTW: As an exercise do some “filtering” (bad/wrong data are the norm in our business):
Check if all your trimming curves are coplanar. If not mastermind some sort of “clustering” (meaning sample “groups” of trimming stuff VS curves to trim with coplanarity criteria)
Mastermind a policy for dealing with intersecting trimming curves.
Do the 2 with self intersecting ones as well (see related Method in Curves).