Attractor Pattern junction

jean_philippe_de

Hi,
I'm trying the Attractor pattern on a solid of revolution. It's OK but the problem is that the pattern doesn't join correctly.
See the screenshot.

https://cad.onshape.com/documents/a5f5be574d6a5634babad56c/w/b224ed94db346ffded968616/e/3e37ea518e7fcf68281240a9

Thanks.

S1mon

@Evan_Reese

I played with this a bit, and I can't easily tweak the parameters to solve the issue either. I suspect that I could hack it by dividing up the revolved bottle shape into a wedge and tweaking the angles until I could get something that would pattern nicely. 

I vaguely remember these kinds of challenges from using Attractor Pattern in the past. Perhaps the default should be to use the full parameter space with Border=0 even if the instances will only partially be on the surface. There could also be some sort of special accommodation for periodic surfaces.

jean_philippe_de

"dividing up the revolved bottle shape into a wedge and tweaking the angles until I could get something that would pattern nicely"
i've tried but the result is the same.
Thanks

S1mon

@jean_philippe_de

Here's what I meant.

https://cad.onshape.com/documents/ef809f172ecc65fc66a9c8d7/w/ed608413499b83c0aaa64068/e/e51feea3e485a49d59489d7f

The trick is that I split the surface, and tweaked the 7.4 degree angle until the instances visually lined up with one 1/4 of the whole. 

Ideally this hack wouldn't be necessary. It would also be tricky to get this to be exact and parametric.

jean_philippe_de

Thanks !!
and yes, this hack wouldn't be necessary.

EvanReese

This issue is on my backlog of things to fix with this feature, but I can't guarantee a timeframe for it. My priorities are elsewhere at the moment. What it needs is a way to tell when the base surface is periodic in the u or v direction, then update the grid accordingly. 

jean_philippe_de

OK and thanks Evan.