Flow Along Srf in Grasshopper

Replies to This Discussion

Permalink Reply by Chris Tietjen on February 18, 2012 at 1:17am

Hi Renata,

Use the 'Map to Surface' component.  Let me know if I'm missing some detail of what you're trying to do.

Chris

Attachments:

• CurveMapping.gh, 5 KB

Permalink Reply by Nick on February 27, 2012 at 12:24pm

Hi Chris ...I'm trying to solve the exact same solution in Maya but haven't been successful so looking at other applications now.  From your screen capture it appears that there is some distortion from the original. Is there a way to do this where the original curve(s) retain their profile when deforming over the target mesh?

Cheers

Nick.

Permalink Reply by Michael Pryor on February 27, 2012 at 12:57pm

project?

Permalink Reply by Nick on February 27, 2012 at 3:51pm

I suspect that whatever underlying process that is going on in order to use the "Map to Surface" component is actually doing some sort of plane/camera projection to begin with. The problem with this is it doesn’t respect the actual profile of the original curve itself in its local space in comparison to the target mesh in world space where there is wildly differing depth values …so you end up with a varying degree of distortion. I was hoping there might have been a way to normalise those values.

Permalink Reply by Michael Pryor on February 27, 2012 at 5:25pm

project follows the form exactly in my case, where as map seems more "approximated" yet both have the exact same plan. 

Permalink Reply by Nick on February 28, 2012 at 2:45pm

wow ok interesting ...how do I read your image? The orthographic ...are they the curves pre deformation and that image plane a displacement map? Or are we seeing them being projected already? Is the green/red where it is deviating from the original??

Thanks for the reply :)

Permalink Reply by Michael Pryor on February 28, 2012 at 3:00pm

hi nick, the green like is a result of the project component, the red line is the result of map to surface. the curves in both cases start as flat 2d curves, mapped to the surface (red) and projected to the surface (green). in the plan view you see both curves seem identical in profile, in the perspective on the right we see that in red, map to surface has come off the surface in areas in the z direction, where as project, in green, perfectly follows it. 

green, using project, is where it should be, red, using map, is a deviation. 

Permalink Reply by Nick on February 28, 2012 at 3:57pm

ahhh I see ok great! I don't actually know grasshopper/rhino at all so I was just assuming perhaps this surface mapping was doing some funky projected math.

So my actual question remains then... can you show the original 2D curve before the projection next to the projected result?

What I am essentially trying to solve is the issue I have here. Image on the left is projected onto a sphere ...the right is that projection translated along the surface normal ...the distortion comes from the fact it is a nurbs surface and essentially open so tapering at the poles ...however you get a similar problem if you try and project to a poliginal surface. Limitations of the software so hoping your solution is a winner. I am ultimately wanting to move a curved surface over another surface and the curve not deform... i.e. it retains its local space profile while attached to the surface

Permalink Reply by Michael Pryor on February 28, 2012 at 4:58pm

but if u wrap a curve around another surface it will always distort in some way except when viewed from the projection view. so i dont understand what you want to achieve.

Permalink Reply by Nick on February 29, 2012 at 3:52pm

Hi Michael ...for some reason I cannot reply to your below post.

I found an alternative solution ...went the texture route to get my profiles on the surface the way I want by mapping via lat-long hdr. Just means I loose the interactivity I was after.

I realise the curve will always deform when meshed to a surface that is passing a new profile ...however that said, it is still possible to retain it's local space orientations relative to one another so when it is projected on a surface, the actual world space values remain constant. I think I will just have to write something. I can see this being useful in the future.

Thanks for all the help. Much appreciated.

Permalink Reply by Sean Madigan on August 8, 2012 at 7:50pm

There is a portion of this definition that does the "flow along surface" command. with a variable level of accuracy by dividing the curves and then mapping their uv coordinates along the new surface and recreating new curves with the newly mapped points,  good luck!

http://www.sean-madigan.com/2012/08/09/flow-along-surface/