RES: [Paraview] Particle Simulation and Paraview modification?

Renato N. Elias rnelias at nacad.ufrj.br
Sat Jan 21 10:36:13 EST 2006


 

Viksit asked: >>> does the ensight reader work in parallel? <<<

Yes, Ensight/Paraview readers can do it through the server-of-server format
(an extension for the case format). Take a look in the examples distributed
with Paraview
(http://www.paraview.org/files/v2.4/paraview-tutorial-data-2.4.tar.gz). The
server-of-server files are those with SOS extension. Furthermore, you'll
have to compile and launch PV in MPI mode and load the SOS file that will
point to other CASE files and these geometries and results respectively.

If you're interested in using Ensight's format, get the Fortran routines
available in
http://www.nacad.ufrj.br/~rnelias/public/EnsightFortranRoutines.zip (the
Ensight's format description is within the zip file too). 

About your particles: you can write only the particle coordinates (a mesh
node, strictly speaking) or associate each particle to an arbitrary ID and
save "point" elements, like this:

C-------------------------------
      buffer = 'part   '
      write (buffer(5:7),'(i3)') 7
      write (iens, '(A20)') buffer

	buffer = 'PART NAME'
      write (iens, '(A20)') buffer

      buffer = 'point'
      write (iens, '(A )') buffer
      write (iens, '(i8)') k

      do no = 1, nnos
         write (iens, '(4i8)') no, no
      enddo
C-------------------------------

Cheers

Renato N. Elias
===================================================
Researcher
High Performance Computing Center - NACAD/COPPE
Federal University of Rio de Janeiro, RJ -Brazil
http://www.nacad.ufrj.br/~rnelias
 


-----Mensagem original-----
De: paraview-bounces+rnelias=nacad.ufrj.br at paraview.org
[mailto:paraview-bounces+rnelias=nacad.ufrj.br at paraview.org] Em nome de
Viksit Gaur
Enviada em: sexta-feira, 20 de janeiro de 2006 22:34
Para: SamuelKey; Paraview ML
Assunto: Re: [Paraview] Particle Simulation and Paraview modification?

Hi Samuel,

Sorry, I was unable to respond earlier - was away. I checked out ensighy as
you suggested, and I'm not sure about a couple of things which I thougt you
might answer for me.

Ensight is a separate application, and I assume its expensive as well :)
  When you say that I can use either Ensight 6, 8 or Gold formats, I guess
you mean that my data can be exported in the desired format, using a custom
program? VTK-XML seems to be something not very well suited for my work, the
reason for which is given below.

Secondly, you're right about me having to look at a lot of datasets -
mostly, rather large particle data sets over time. Are there any tutorials
out there which might help me write a reader which can work in Parallel? Or
does the ensight reader work in parallel?

Thanks,
Viksit

SamuelKey wrote:
> Greetings Viksit,
> 
>>From my use of PV, it does what you want already.
> Use either the Ensight reader ( http://www.ensight.com download their 
> user's manual to get format documentation) or the VTK XML-format if 
> you really have a small number of particles (2-200). Both support 
> points as cell objects.
> 
> The multiple-time-step support with the ensight-format is much better.
> 
> I don't know if PV can automatically add glyphs to pionts on start-up 
> -- maybe with a script (which it will write on request.).  Scripts are 
> OK if you are looking at lots of data sets and want to see the same 
> pictures each time.
> 
> The more you use PV the better it gets.
> 
> Good viewing,
> 
> Sam Key
> 
> ----- Original Message -----
> From: "Viksit Gaur" <vik.list.vtk at gmail.com>
> To: <paraview at paraview.org>
> Sent: Saturday, January 07, 2006 11:03 PM
> Subject: [Paraview] Particle Simulation and Paraview modification?
> 
> 
> Hi all,
> 
> I'm developing a particle simulation system for some physics 
> experiments, and I'm exploring VTK and Paraview to do so. Till now, 
> the extensive (and excellent) documentation available for both leads 
> me to believe that since both are being used extensively, they should 
> solve my purpose.
> 
> That said, I wanted to gauge an opinion on the method I've thought I 
> should use -
> 
> The simulation is of certain particles whose coordinates are given in 
> time steps, and they need to be represented by glyphs - spheres, 
> boxes, and so on. I would then like to be able to see these particles 
> move over a given time period, and have the ability to track a certain 
> particle through its sequences.
> 
> How I think I could go about it, is:
> We use a custom format to represent the simulation data, and thus I 
> would need to create a VTKReader for this. Next, this reader has to be 
> included with Paraview, and data can be loaded. The next part is what 
> I'm unsure about - to be able to zero in and select a particular 
> particle from the simulation, does Paraview source need to be 
> modified? Or can i write TCL scripts to do so? Or, better still, is 
> this functionality provided already?
> 
> Thank you,
> Cheers!
> 
> Viksit
> _______________________________________________
> ParaView mailing list
> ParaView at paraview.org
> http://www.paraview.org/mailman/listinfo/paraview
> 
> 
> 


--
Viksit Gaur
Deptt. of Computer Science
Yale University

me at viksit dot com
http://viksit.com

_______________________________________________
ParaView mailing list
ParaView at paraview.org
http://www.paraview.org/mailman/listinfo/paraview



More information about the ParaView mailing list