"Gg" sorry for the name ^^.
Look into the icetree that is on the "Grid" object. That's the one that
calculate the "tangent" I called "Gg".

To select all edges border :
Select edge mode (right menu)
then select "border edge" in the menu just below.
Now if you rectangle select with the mouse in the view, only the border
edges (the light blue one) will be selected.

On Thu, Mar 10, 2016 at 3:24 PM, Morten Bartholdy <x...@colorshopvfx.dk>
wrote:

> Wow thanks a lot Olivier - that looks like it will do the trick. The
> reason I
> didn't want to use curves is merely that I will get a lot of polygon
> islands and
> don't know how to automagically select all border edges for extraction -
> otherwise curve tangency would be fine.
>
> The Get Gg node just before Align Particle to Vector looks like it is the
> one
> that provides the edge vector, correct!? But I fail to understand what it
> is
> doing though.
>
> Cheers
> Morten
>
>
> > Den 10. marts 2016 klokken 14:17 skrev Olivier Jeannel
> > <facialdel...@gmail.com>:
> >
> >
> > If you don't want to use curve (and curve tangent), I made this, but you
> > need then to subdivide your edges to get good result.
> > It calculate a "tangent" stored per point on the polygonal object.
> > Then your Pointcloud get the closest location and take that tangent data
> > and aligns vector.
> >
> >
> > On Thu, Mar 10, 2016 at 2:01 PM, Olivier Jeannel <facialdel...@gmail.com
> >
> > wrote:
> >
> > > You don't want to work with extracted edge curves ?
> > >
> > > On Thu, Mar 10, 2016 at 1:44 AM, Ed Manning <etmth...@gmail.com>
> wrote:
> > >
> > >> I think there might be something in Eric Mootz's emTools that does
> > >> alignment to edges?
> > >>
> > >> On Wed, Mar 9, 2016 at 7:22 PM, Morten Bartholdy <x...@colorshopvfx.dk
> >
> > >> wrote:
> > >>
> > >>> I am tinkering with a simple method for creating a cityscape in ICE
> using
> > >>> scatter tools. Julien Carmagnac SiScatter does most of what I need
> with
> > >>> insetting instances from edges and even scattering, but I am missing
> one
> > >>> important functionality. I need to be able to align the x or z axis
> of
> > >>> the
> > >>> instances to the tangent of the closest border edge, and then add a
> > >>> little
> > >>> randomization to the y rotations. I have looked deep into the
> compounds
> > >>> and
> > >>> tried different ways to get the edge tangent, but alas my ICE skills
> > >>> come up
> > >>> short here. Any scattering tool would do I guess - Milan Vaseks
> Scatter
> > >>> Tools or
> > >>> even simple nonsimulated emissions, but I need this extra alignment
> > >>> method. I
> > >>> tried Juliens Generate instance on Curve too, but it requires edge
> > >>> curves to be
> > >>> selected and extracted, plus it seems the instances seem to be
> flipped
> > >>> 180 deg
> > >>> on the y-axis between isntances without any way to control it.
> > >>>
> > >>> I have attached a screenshot showing a simple test where the A
> example
> > >>> is what I
> > >>> need, and B is what I get with Generate instance on Curve.
> > >>>
> > >>> Any ideas on this will be much appreciated.
> > >>>
> > >>> Best
> > >>> Morten
> > >>> ------
> > >>> Softimage Mailing List.
> > >>> To unsubscribe, send a mail to
> softimage-requ...@listproc.autodesk.com
> > >>> with "unsubscribe" in the subject, and reply to confirm.
> > >>>
> > >>
> > >>
> > >> ------
> > >> Softimage Mailing List.
> > >> To unsubscribe, send a mail to
> softimage-requ...@listproc.autodesk.com
> > >> with "unsubscribe" in the subject, and reply to confirm.
> > >>
> > >
> > >
> > ------
> > Softimage Mailing List.
> > To unsubscribe, send a mail to softimage-requ...@listproc.autodesk.com
> with
> > "unsubscribe" in the subject, and reply to confirm.
> ------
> Softimage Mailing List.
> To unsubscribe, send a mail to softimage-requ...@listproc.autodesk.com
> with "unsubscribe" in the subject, and reply to confirm.
>
------
Softimage Mailing List.
To unsubscribe, send a mail to softimage-requ...@listproc.autodesk.com with 
"unsubscribe" in the subject, and reply to confirm.

Reply via email to