[vtkusers] about use vtkPolyDataNormals

[David Doria]

On Thu, Nov 26, 2009 at 10:04 AM, renlishen <renlishen at gmail.com> wrote:

>
> David, thanks~
> I find the key point to my question...
> If I don't use the vtkPolyDataNormals' SplittingOff()
> I get more points and normals like this
> pic1:
> http://old.nabble.com/file/p26530678/11.png
>
> but when I use SplittingOff()
> I get the correct number of points and normals
> pic2:
> http://old.nabble.com/file/p26530678/22.png
>
>
> but I don't know why vtkPolyDataNormals need to "splitting" the edge.....
> anyway~thanks again for your replying...^^
>
>
> p.s. my data has 5 points and 6 cell(face)
> After using SplittingOff() ----> 5 points and 5 normals(pic2)
> If no use SplittingOff()    ----> 16 points and 16 normals(pic1)
>
>
It looks to me like SplittingOff() is finding the average normal at each
point. You can see this by looking at the tip of your pyramid - the normal
that exists in 22.png does not exist in 11.png - it is the average of the 4
vectors. With SplittingOn() - it looks like it is simply using the normal of
the face. Since each point is connected to multiple faces, there are
multiple normals.

Hope that helps.

Thanks,

David
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.vtk.org/pipermail/vtkusers/attachments/20091126/56088e36/attachment.htm>