Re: Cell-centered data

• To: Kapper Michael G Contr AFRL/PRSA <Michael.Kapper.ctr@xxxxxxxxxxxxxx>
• Subject: Re: Cell-centered data
• From: Bill Hibbard <billh@xxxxxxxxxxxxx>
• Date: Mon, 31 Jul 2006 14:14:50 -0500 (CDT)

Hi Mike,

While such data could fit in the VisAD data model,
we have not implemented the necessary classes. We
would need Set classes defining appropriate
interpolation (constant values over cell polygons),
probably a new extension of Field or Function to
alter assumptions in some methods of point samples,
and probably a new extension of DataRenderer. This
would be a lot of work.

You may be able to improve the quality of approximation
to your data by using a higher resolution grid, but of
course at the cost of performance.

Good luck,
Bill

On Mon, 31 Jul 2006, Kapper Michael G Contr AFRL/PRSA wrote:

> I have a simple question concerning GriddedSets.  If I initialize a
> Gridded2DSet using coordinates of cell vertices, is there a way to
> setSamples with data from the cell centers?  i.e. is VisAD capable of
> filling in the values enclosed by the vertices?  Right now I average the
> values from neighboring cells and setSamples of the vertices, but this
> degrades the quality of discontinuous solutions.

==============================================================================
To unsubscribe visad, visit:
http://www.unidata.ucar.edu/mailing-list-delete-form.html
==============================================================================

• References:
  • Cell-centered data
    • From: Kapper Michael G Contr AFRL/PRSA