Normals errors - unify

Hello,

I am having trouble with normals vector on my definition :

The element from the same tree branch have different plane orientation.

I would like to unify them to match a single vector direction.

Would you know how to do such thing ?

Thank you very much fro your time.

Marc.

Replies to This Discussion

Permalink Reply by Marc Johnson on April 7, 2013 at 11:06am

Thank you Ralf for your help.

Yes I would like to reorient the wrong normals to match the initial mesh normal.

However I got mix up with all the planes components in my definition.

Do you think you will be able to "repair" that mistakes ?

thank you.

Marc.

Permalink Reply by Ralf Lindemann on April 7, 2013 at 11:49am

Ok. I see. Could you isolate (internalise data) the mesh geometry, the elipses and the part that creates the ellipses and post the definition and I'll have a look at it.

But I guess it's going to be something similar to what David suggested below.

Permalink Reply by Marc Johnson on April 7, 2013 at 1:47pm

Thank you for your help. It will be very helpful.

Attachments:

thank you so much.gh, 438 KB

Permalink Reply by Ralf Lindemann on April 8, 2013 at 3:20pm

Not quite there yet but I didn't have more time at the moment.

Have a look. You need to rotate the ellipses by 90deg and flip them into the right direction as David explained. Let me know how your're getting on. I'll can have another quick look tomorrrow.

Attachments:

thankyousomuch.gh, 1.5 MB

Permalink Reply by Marc Johnson on April 9, 2013 at 2:28am

Dear Ralf, Thank you for your help.

The orientation is ok.

However the resulting curves (ellipses) do not match with the alignement of the original lines.

Do you know why and/or how to fix this ?

Best Regards.

Marc.

Permalink Reply by Ralf Lindemann on April 13, 2013 at 6:30am

Marc,

you just need to change the x and y axis to get the right alignment.

Hope that's what you want.

Best,

Ralf

Attachments:

thankyousomuch_02.gh, 1.5 MB

Permalink Reply by Marc Johnson on April 13, 2013 at 8:38am

Thank you very much Ralf both for your patience and performance.

Have an excellent week end.

Marc

Permalink Reply by Marc Johnson on April 14, 2013 at 6:03am

Dear Ralf,

would you know ?

http://www.grasshopper3d.com/forum/topics/error-data-conversion-fai...

Permalink Reply by J[D] on May 29, 2013 at 9:54am

Hi Ralf...I saw your C# component used to flip normals...I have a similar problem:

I'm trying to unify the normals of a mesh built from a metaball function.

When I move the points the geometry changes and also the normals (I also analyzed the output with the plug-in ladybug that works only with the right normals)

I would like to modify the script to obtain a good result but my attemps failed...could you help me please?

Thanks in advance

Attachments:

Unify_normals.3dm, 65 KB
Unify_normals.gh, 106 KB

Permalink Reply by David Stasiuk on April 5, 2013 at 4:22am

If you can use your base mathematica mesh as a guide, then you can evaluate the normal on the mesh at the closest point to each of your ellipse center points, and then determine the angle between the base mesh closest-point normal and your ellipse-center normals. If that angle is greater than Pi/2, you flip it, if not, you leave it as is.

Permalink Reply by Ron Shvartsman on May 29, 2013 at 10:29am

I have seen this problem and experienced it several times (if I am not mistaken as to your problem). It seems to me you simply need to orient your planes in the proper direction. From what I can tell you are using mesh.normal component as well as a plane.normal component later on. Dependent on your input geometry, simply using these components may not give you a dersirable result (i.e., you planes may be normal to to the vector, however, they may be skewed along the rotation axis). You need to use more plane components to define specifically the orientation (if the orientation is of such importance). Try 3 point plane for instance. It will require a bit of data management but doable I think.

Does this help??

Ron

Permalink Reply by J[D] on May 29, 2013 at 1:53pm

First of all thanks for the quick reply,

but I still have some doubt about what you said:

I need to define the orientation of each mesh face in a dynamic-way because using the metaball component, when you change the position of the points (and I need to work with that) it generates isocurves but their orientation changes in a non predictable way.

In my definition the geometry is already defined, could you explain more how can I use the 3 point plane to define a NEW orientation of each face mesh?Or I need to define again all the geometries in a different way?

I attached the definition on the last post.

Really thanks for your time