[vtk-developers] vtkArrayCalculator enhancement

Moreland, Kenneth kmorel at sandia.gov
Mon Aug 28 17:47:26 EDT 2006


I don't think deforming the mesh by changing the point coordinates will
"wreak major havoc" with the Pipeline.  The point coordinates have no
effect on ghost cells.  Ghost cells are determined by connectivity,
which still cannot be changed by the array calculator.  I'm pretty sure
the pipeline will automatically recompute the geometric bounds of the
data as necessary.  And although it is true that changing point
coordinates will invalidate normals, the same is true for any filter
that modifies the geometry.

 

Overall, the behavior would not be much different than any of the warp
filters.  True, it would not be hard to create a bizarre or invalid
output, but VTK provides many easier and more spectacular ways to shoot
yourself in the foot.

 

-Ken

 

________________________________

From: vtk-developers-bounces+kmorel=sandia.gov at vtk.org
[mailto:vtk-developers-bounces+kmorel=sandia.gov at vtk.org] On Behalf Of
Randall Hand
Sent: Monday, August 28, 2006 3:11 PM
To: Clinton Stimpson
Cc: vtk-developers
Subject: Re: [vtk-developers] vtkArrayCalculator enhancement

 

Sounds intruiging.  I previously toyed with the idea of adding the
ability to use a point's location in the formula but never implemented
it.  In my case I needed to compute a cartesian vector's cylindrical R
component for an isosurface extraction (radial field strength).  I wound
up just doing it manually. 

But allowing the user to "deform" the mesh through the Array
Calculator... I'm not an expert here, but won't that wreak major havoc
with the vtk Pipeline (Ghost Cells, normals, range/extents)?

On 8/28/06, Clinton Stimpson <clinton at elemtech.com> wrote:

Just thought I'd let others know...

I'm enhancing vtkArrayCalculator so that it can compute on point
coordinates as well as scalars and vectors.
And also have an option to output results as coordinates.
I hear I'm not the only one that wanted this. 

I'll wait a few days till the dashboard clears up, of course.

Clint

_______________________________________________
vtk-developers mailing list
vtk-developers at vtk.org 
http://www.vtk.org/mailman/listinfo/vtk-developers




-- 
----------------------------------------
Randall Hand
Visualization Scientist
ERDC MSRC-ITL 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/vtk-developers/attachments/20060828/67c7de7b/attachment.html>


More information about the vtk-developers mailing list