Reorganize Intersection Points

Replies to This Discussion

Permalink Reply by Andrea on February 28, 2015 at 2:22pm

another screen

Attachments:

• Schermata 2015-02-28 alle 20.52.14.png, 63 KB

Permalink Reply by David Rutten on February 28, 2015 at 3:11pm

Since you're using equations to generate the curves, why not also use equations to find the corner points directly?

Permalink Reply by Andrea on February 28, 2015 at 4:27pm

i don't know how to do it , the equation gave me the first curve, than i need to duplicate the curve several times around an axis, than to mirror the result. 

I don't understand how to work with the equations to find the point that i need, in group of for like i need.

:( 

Permalink Reply by David Rutten on March 1, 2015 at 4:06am

I understand how the file works, it's just that the way it works seems to combine the worst of both worlds*. It's using expressions to generate exact points. Then those points are interpolated, ruining the accuracy that existed before. Then those inaccurate curves are intersected, which results in intersection points that may themselves add a small amount of inaccuracy yet again (not to mention it's slow to intersect many curves).

It sounds as though you're almost at a perfect mathematical solution, but you've stopped just short and are now adding enormous layers of complexity (and inaccuracy, and computational expense) instead.

I realise this is not what you want to hear necessarily, but if you want to make this shape, you need to understand the mathematics of it. 

At any rate, I'd recommend only intersecting a single curve with all counter-curves, then polar arraying the intersection points. That gives you a much more organised data structure:

* One world being the mathematically explicit, the other one being the intuitively geometric.

Attachments:

• OsakaMaritimeMuseumForPANELS.gh, 17 KB