Ok, "invalid positions" will separate the boundary points from the
volume points but in respect to the boundary conditions the boundary
points are not equivalent. Is it possible to mark the boundary points
with a tag/property  that specify the type of boundary conditions
imposed on the surface this point belongs?

Thanks for your comments,
Ted 

Chris Pelkie wrote:
> 
> DX supports the concept of "invalid positions" (this is much easier
> to implement using the native format, in your case).
> 
> So, consider making all non-boundary positions in the data have the
> value byte(01) (invalid) and all boundary points valid (byte(00)).
> That way, you can preserve the full grid structure as a regular, thus
> compact, representation and ignore the invalid positions by
> Remove'ing them from the branch in your network that displays the
> entire grid.
> 
> For a parallel branch that displays only the boundary info, the
> invalid positions will cause all interior points to not display.
> 
> There are 2 invalid declarations: the byte style I mentioned requires
> a value at each grid point. The "by reference" style permits you to
> note the indices of only those positions that are to be "invalid",
> but in your case that number is probably much larger than the valids,
> plus it's probably more work to calculate, so I suggest the simpler
> byte mask, that is one invalid position entry (either 00 or 01) for
> each position; the invalid positions array is thus to be declared
> "position dependent". If you already know the indices of the
> boundary, you're almost done.
> 
> Chris Pelkie
> Vice President/Scientific Visualization Producer
> Conceptual Reality Presentations, Inc.
> 30 West Meadow Drive
> Ithaca, NY 14850
> [EMAIL PROTECTED]

-- 
Ted Sariyski

Combustion Research and Flow Technology, Inc.
174 North Main Street
Building 3, P.O.Box 1150
Dublin, PA 18917
Tel: (215) 249-9780
Fax: (215) 249-9796
[EMAIL PROTECTED]
[EMAIL PROTECTED]

Reply via email to