Hi Michael,
is it possible to take the align to surface which is part of array along curve and make a standalone script of this feature.
I know we have the orient tool but I'm thinking when an object is moved by bounding box centre it snaps to surface but does not align.
Cheers
Barry

Hi Barry, well the move tool is intended to only do translation, not rotation. The Transform > Orient tool is the one that does translation and also rotation.

Would a "Bounding box center" option for Transform > Orient do what you are describing? If not I think I'll need more information.

- Michael

Surface normals at points on surface are now available by methods.
- B

Reading the v4 beta release notes:

10/10/2019, face.evaluateNormal(uv).
The normal is a vector.

1/22/2020, face.dropPoint(pt), yields 2D uv of surface point, closest to input pt.
So pt just has to be close to surface, or could be a point on an isoline.

I figure that a point (say the center point of a circle), could be acted on by a force, either attractive or repulsive, move to pt2, and a new point on surface found, with new circle aligned to new surface normal. What is the time overhead?

- Brian

Hi Michael,
<<Would a "Bounding box center" option for Transform> Orient do what you are describing? If not I think I'll need more information.>>
Yes that's what I'm after.
Perhaps the photo illustrates what I mean better.
Cheers
Barry

Hi Brian,

re:
> 1/22/2020, face.dropPoint(pt), yields 2D uv of surface point, closest to input pt.
> So pt just has to be close to surface, or could be a point on an isoline.

The point given to face.dropPoint() can actually be anywhere, it will find the closest point on the surface from the input point.


> I figure that a point (say the center point of a circle), could be acted on by a force, either
> attractive or repulsive, move to pt2, and a new point on surface found, with new circle
> aligned to new surface normal. What is the time overhead?

Well it depends a lot on the complexity of the surface and also the complexity of the trimming boundaries. It does have to do quite a bit of work but you could probably do a few hundred calls.

- Michael

Hi Barry, please try the attached plugin command.

- Michael

Thank you Michael. The surface would have to be locally flat-ish, relative to the radius of the circle. Circle centerpoints that are 2D far away would have to be 3d far away, (no nearby folding). D3- force has the force equations, but only does boundaries for 2d constant x or 2d constant y. Or huge 2d arc boundaries, with external centers. The only way to repel from a surface curve boundary, AFAIK, would be to replace boundary curve with lots of fixed points, close together relative to circle radius...which drives up overhead.. Or many small arcs with external centers.. And boundary curves need to be "reasonable". Surfaces with internal boundaries, (e.g. holes) similarly. Have not located efficient boundary curve repulsion algorithm.

Re surfaces, points on curve (e.g. isocurve), on surface, can be used to get normals, for alignment to surface.

Brian

Maybe dropPoint could be used on boundary curves to advantage?

- Brian

Hi Michael,
thank you that works fine.
Have a good Xmas.
Barry