[Paraview] Intra-zonal surfaces

[Moreland, Kenneth]

Uh, sure.  Sounds good.  D3 clearly needs some updating, so hack away.

 

-Ken

 

BTW, D3 will find duplicate points and merge them together.  If you have
global point ids, that can make the process faster and more reliable.

 

________________________________

From: John Biddiscombe [mailto:biddisco at cscs.ch] 
Sent: Friday, January 11, 2008 9:02 AM
To: Moreland, Kenneth
Cc: paraview at paraview.org
Subject: Re: [Paraview] Intra-zonal surfaces

 

Ken

Thanks for confirming what I suspected. As it is, I'm just finishing
vtkMultiBlockUnstructuredGridAppend
which iterates over all the leaf nodes and appends all Unstructured
Grids into a single dataset (dumps any other dataset types that may be
present). Immensely wasteful of space, but I'll just run it once and
save the surface generated by vtkGeometry filter for later use.

Since D3 will find ghost cells for a single UnstructuredGrid that is
split across processors - I will consider making a modified version
which accepts a multiblock dataset (i.e. different leaf nodes of a
compound dataset) and operates that way (D3 doesn't expect points to be
duplicated does it?). Not sure how much work that would be - but in
principal we only need to fool D3 into taking data from leaf nodes of a
dataset rather than pieces of a dataset.

That will work, yes?

JB






John,
 
I am afraid that there is no good way to do what you want right now.
There is no filter that is able to directly determine which external
faces of each block are actually abutting or are inside another block.
You could use tools like D3 or Clean to Grid, but as you suspect you
first have to cram everything in a single unstructured grid.  Right now
there is no filter to do that either.  Writing one would be trivial, and
there has been talk on this list for doing it.  I think no one has done
it because it is generally best to avoid doing so whenever possible.
 
Does your CFD dataset happen to look like a brick?  If so, you can fake
the operation with 6 slice filters on the boundaries of the data.
 
-Ken
 
  

	-----Original Message-----
	From: paraview-bounces+kmorel=sandia.gov at paraview.org
	    

[mailto:paraview-
  

	bounces+kmorel=sandia.gov at paraview.org] On Behalf Of John
Biddiscombe
	Sent: Friday, January 11, 2008 7:10 AM
	To: paraview at paraview.org
	Subject: [Paraview] Intra-zonal surfaces
	 
	When I load a complex CFD dataset with several dozen zones, I
create a
	multiblock data structure which has one dataset per zone/volume.
	 
	When visualizing this as a surface only and with transparency
turned
	    

on.
  

	The faces between zones are visible as artifacts inside the
volume
	    

where
  

	the zones abut. I was wondering if back face culling would
remove
	    

these,
  

	but of course, there are font faces there too.
	 
	Is there any way of coaxing d3 or similar filter to generate
ghost
	    

cells
  

	that can be used to overcome these artifacts.
	or
	I am thinking that if I join all cells into a single
unstructured grid
	(in parallel if necessary) then these surface cells will
disappear
	    

(but
  

	perhaps there is a solution I have missed). Is there a filter to
	    

append
  

	all the zones into one already present in ParaView - like
AppendFilter
	for a multiblock->single unstructured grid.
	 
	Thanks
	 
	JB
	 
	--
	John Biddiscombe,                            email:biddisco @
cscs.ch
	http://www.cscs.ch/about/BJohn.php
	CSCS, Swiss National Supercomputing Centre  | Tel:  +41 (91)
610.82.07
	Via Cantonale, 6928 Manno, Switzerland      | Fax:  +41 (91)
610.82.82
	 
	 
	_______________________________________________
	ParaView mailing list
	ParaView at paraview.org
	http://www.paraview.org/mailman/listinfo/paraview
	    

 
 
  






-- 
John Biddiscombe,                            email:biddisco @ cscs.ch
http://www.cscs.ch/about/BJohn.php
CSCS, Swiss National Supercomputing Centre  | Tel:  +41 (91) 610.82.07
Via Cantonale, 6928 Manno, Switzerland      | Fax:  +41 (91) 610.82.82
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://public.kitware.com/pipermail/paraview/attachments/20080111/6490de43/attachment.html