Voronoi offset failure

Replies to This Discussion

Permalink Reply by Manuel on December 14, 2009 at 7:49pm

If you don't want to change the points, maybe you can cull the correct curves (maybe with the curve length or null elements) and then apply a offset with a smaller distance.

Permalink Reply by Jaroslaw Sakowicz on December 14, 2009 at 10:27pm

Hi,
problematic cels are those where curve seam get outside offseted curve. just move curve seam and rhino will offset it, could get tedious depend how many you got.

Permalink Reply by WoodDuck on December 14, 2009 at 11:12pm

Thanks for the replies. I don't mind changing the points or curve seams manually but I'm not sure how... Is there a way to select / move them in Rhino without baking or do I need to bake them first?

Permalink Reply by Manuel on December 15, 2009 at 12:04am

I think that the problem is when you have points that are too close each other, so maybe you need to remove those points (http://www.grasshopper3d.com/forum/topics/need-help-filter-from-list)

or maybe you can use the series component with a small step (a slider?) to create the points, then you use the jitter component so you get a random order... the step slider will control how close are the points.

Permalink Reply by WoodDuck on December 15, 2009 at 1:17am

Thanks Manuel, I tried the first idea but it seemed to generate the bad cells again. The second idea, I'm not sure if I'm implementing properly as I can't get any points at all. Here is the screen print:

Thx

Attachments:

• Random Points - Voronoi.png, 194 KB

Permalink Reply by Manuel on December 15, 2009 at 8:32am

on the second idea you probably are feeding the wrong uv points, but I tried myself and i still getting some offset problems so the points to close each other are not the problem...

try using the fillet component after the voronoi component and then the offset component, that solved some of the cells, but it still depend on the offset distance...

Attachments:

• Voronoi Offset.gh, 42 KB

Permalink Reply by WoodDuck on December 15, 2009 at 10:06am

Thanks very much Manuel, this does seem more stable. I get the occasional failed cell but it's nothing I can't fix in Rhino. I also added a slider for the seed shuffle on the jitter components. Then if I get a bad cell I can shuffle things around with the same number of points til I get a complete set of good cells.

Just out of curiosity, I looked at the cells that fail and it seems they always fail at a sharp corner with a short line. Here are some images with a good cell, then as the offset increase, you can see where it fails. It seems to have trouble offsetting the short line just as it comes to a point.

Thx again

Attachments:

• Voronoi Offset Failure 1.png, 19 KB
• Voronoi Offset Failure 2.png, 18 KB
• Voronoi Offset Failure 3.png, 16 KB

Permalink Reply by Björn Syse on January 10, 2012 at 5:31am

Hi,

Have you tried doing the voronoi without the containment box argument?