IntCrv

Hi everyone, I have a collection of points in rhino that I want to interpCrv on, with out having to select them one by one.

I created a simple definition as you can see, and set the point to "multiple points", but I get this result:

Some points are "out of order". Is there any way to reorder them with out having to select them one by one by hand? I have to do A LOT of this curves, I'm beginning to worry :)

thanks a lot!

Attachments:

InterCrv.jpg, 188 KB

Replies to This Discussion

Permalink Reply by Alejandro Ispani on March 5, 2015 at 8:01am

I clearly have a problem with how the points where generated:

They are all done now, and generating them in order would be a real problem... Anyone have any clever ideas?

thanks!

Permalink Reply by peter fotiadis on March 5, 2015 at 8:46am

Post your data (always do that when asking for some help).

The Brain behind GH (David Rutten) had recently posted a rather genius solution on "similar" matters using code and iterative proximity.

Permalink Reply by peter fotiadis on March 5, 2015 at 9:02am

Here's David's solution (I've just converted VBA>C : not a big deal by any means).

Spot the chaos and the order (the polyline/pipe).

Attachments:

IterativeProximity_V2.gh, 18 KB

Permalink Reply by Alejandro Ispani on March 5, 2015 at 9:43am

That works like a charm peter, thank you very much.

I also was able to fix it by drawing a quick curve around the points, and using the "Sort Along curve" component. But your version is much better.

next time I'll remember to attach my file!

Permalink Reply by peter fotiadis on March 5, 2015 at 9:54am

NO NO and NO :

it's NOT my version: it's David Rutten's stuff. I just did the stupid thing (VB>C).

That said I also had some "similar" things (iterative closest pair approach etc etc) but they pale (CPU time) VS Master's solution.

Permalink Reply by peter fotiadis on March 5, 2015 at 10:40am

BTW speaking about code efficiency: here's the most embarrassing thing: I need 10 units of time to find the closest pair (with the fastest of the 3 methods sampled).

Fine ... but David needs 1/10th of that in order to "scan" all the points.

Kinda like doing some racing VS Valentino (the N46 not the movie star).

Attachments: