[Paraview] particletracer
Martin Uppman
f04mu at student.lth.se
Thu Feb 19 02:31:45 EST 2009
Hi John,
I downloaded your dataset last night and just got tested it. But I get the
exact same error with your dataset as mine! Let me describe exactly
what/how I'm doing:
1. Loading the dataset
2. Selecting the dataset and create filter 'temporal cache'. Click Apply.
3. Creating a line from menu 'sources'. Modifying the line where I want to
have it. Click Apply.
4. Selecting my dataset go to filters and particletracer. In the new
window I'm choosing 'source' as input port. Under 'select source(s)' I'm
holding down the shift key and click on temporal cache and line. Then I
click OK.
5. Error appears! same as before:
ERROR: In ..\..\..\ParaView3\VTK\Filtering\vtkDemandDrivenPipeline.cxx,
line 822
vtkCompositeDataPipeline (0D4892A8): Input for connection index 0 on input
port index 1 for algorithm vtkTemporalStreamTracer(0D433300) is of type
vtkTemporalDataSet, but a vtkDataSet is required.
Am I doing something totally wrong??
Cheers,
Martin
> Martin
>
> I've put a dataset here
> ftp://ftp.cscs.ch/out/biddisco/flowviz/damBreakMarin.zip
>
> try loading this in paraview, connecting a cache and then a line source.
> If you are able to generate animated particles then all is well and there
> is something wrong with your data.
> If not, then maybe somethig is wrong with paraview cvs and I'd better
> take a look at it.
>
> NB. The dataset supplied is not a very good example as the time variable
> increments as 0,1,2 etc, and in fact should be much smaller, the
> particles travel too far each step. I only just remembered after I
> uploaded it.
>
> JB
>
>
> Hi John, the error messages about Multiblock Dataset is showing when
> I dont use a temporal cache on my input. I tried using just a line
> source for seed points and I dont get any errors about that. So the
> problem is: If I just use particletracer on my input and line soure I
> get error saying Dataset is Multiblock Dataset which is not required.
> If a use a temporal cache on my input and then make a line source as
> seed points and connecting the two using the particletracer, it gives
> error saying dataset is TemporalDataset which is not required. I dont
> have any idea have to come around this problem. Is there another way
> to follow 3d flow in a good manner? I've tried streamtracer but dont
> think you get a clear view. ERROR: In
> /Users/kitware/Berk/ParaView3/VTK/Filtering/vtkDemandDrivenPipeline.cxx,
> line 822 vtkCompositeDataPipeline (0x1ff6cd60): Input for connection
> index 0 on input port index 1 for algorithm
> vtkTemporalStreamTracer(0x1ff6cdf0) is of type vtkTemporalDataSet, but
> a vtkDataSet is required. Cheers, Martin Uppman
> --------------------------------------------------------------------------
> Martin Looking at your earlier message it seems
> like you are setting the seed points using a MultiBlock datastructure.
> I don't think I've enabled support for multiblock seed points (I will
> do so after clicking send, but you won't get it for some time). Please
> try converting your seed points to a simple dataset and retry. The
> error message saying DataSet is required is on port 1 (seeds) - not
> port 0 (vector field) - I didn't notice this before. Sorry. Try unsing
> a line source to start and see if everything works. then move on to
> more sophisticated stuff. JB Sorry, but this doesnt make any
> sense to me. I have a 3D dataset with magnitude and velocity (X,Y,Z)
> for 13 time frames. And as you wrote, paraview internally converts
> multiple vktDataSets into a TemporalDataSet. But particle tracer
> requires multiple time steps, why does it then give error that a
> vktDataSet is required? A vktDataSet, according to what you wrote, is
> just a single timestep. Martin Martin A single
> timestep is a vtkDataSet, StremTracer generates streamline through a
> vector field for this data. The ParticleTracer requires multiple time
> steps to advect particles through the field over time. The particle
> tracer requests 2 time steps per iteration from it's input, internally
> paraview converts multiple vtkDataSets from the input into a
> TemporalDataSet. JB Thanks you, I can now conenct the two sources. I
> tried on Windows and found this alternative if I want to use "input"
> or "source" in the particle tracer. When choosing "source" I could
> select both dataset and seed source. Problem on my Mac was that the
> dialog box was to small so I didnt find this option. Btw what is the
> difference between vtkTemporalDataSet and vktDataSet? ParticleTracer
> requires vktDataSet which I dont have, is it possible to convert the
> dataset somehow. I can use filter StreamTracer on my dataset and
> because StreamTracer and ParticleTracer is based on the same physics,
> I should be able to visualize particles. ERROR: In
> /Users/kitware/Berk/ParaView3/VTK/Filtering/vtkDemandDrivenPipeline.cxx,
> line 822 vtkCompositeDataPipeline (0x225e2020): Input for connection
> index 0 on input port index 1 for algorithm
> vtkTemporalStreamTracer(0x225e20b0) is of type vtkMultiBlockDataSet,
> but a vtkDataSet is required. Cheers, Martin Uppman Can you tell us
> exactly (detailed, step-by-step) what you are doing? That dialog works
> fine for me (running OS X 10.5). -berk On Tue, Feb 10, 2009 at 4:10
> AM, Martin Uppman wrote: Hi again, thank you for fast
> answer John! I'm aware of the problem that my particle tracer doesnt
> have any seed source, but I have been working for quite awhile sloving
> this problem. I did as you said, created temporalcache and seed (point
> seed) but I cant connect the 2 with my particle tracer. I found your
> (John's) great powerpoint presentation on the web explaining
> everything there is to explain. Though I cant make it work!!! In one
> of the pictures in the presentation there are some boxes where you can
> choose which sources to use in the particle tracer (in object
> inspector -> properties -> source). I dont have that option in my
> program. Does someone know why (using v3.4.0 on a Mac)? About v3.4.0,
> isnt it the lastest version? Cant fint anything newer on paraview.org.
> Cheers, Martin Uppman Martin ERROR: In
> /Users/kitware/Berk/ParaView3/VTK/Filtering/vtkDemandDrivenPipeline.cxx,
> line 722 vtkCompositeDataPipeline (0x217c7fb0): Input port 1 of
> algorithm vtkTemporalStreamTracer(0x1e345e00) has 0 connections but is
> not optional. Input 0 should be a time
> dependent dataset. It is best to use a TemporalCache before the
> particle tracer to stop multiple updates for the same time step. Input
> 1 should be some seed points. Ensight -> TemporalCache ->
> ParticleTracer Some Other data eg.line/plane -> the ignore
> pipeline time option is provided to get around some animation issues
> that used to plague paraview. You should not need it, but the correct
> way to use it, is set IgnorePipelineTime to true, then set an
> animation keyframe to animate TimeStep from 0->N - the particle tracer
> will then use the timestep to generate update times for the Ensight
> input, instead of usigin the GUI generated time. If you are using the
> particle tracer from prior to 3.4 then you should switch to a newer
> one as it is contiunually being improved. I will soon (months, not
> days) be checking in a lot of changes which enable it to handle
> dynamic meshes better. JB I've been looking for information on this
> and a found a thread which I think concerns the same problem. In that
> thread Berk made a patch to fix this problem. I also tried to ignore
> pipeline time, but this doent work. I'm working with .case files and I
> want to animate the blood flow in a human heart using particle
> tracers. I'm using a Mac with ParaView version 3.4.0. Thanks for
> looking into this. Cheers, Martin Uppman
> _______________________________________________ Powered by
> www.kitware.com Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html Please keep
> messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView Follow this link to
> subscribe/unsubscribe:
> http://www.paraview.org/mailman/listinfo/paraview -- John
> Biddiscombe, email:biddisco @ cscs.ch http://www.cscs.ch/ CSCS, Swiss
> National Supercomputing Centre | Tel: +41 (91) 610.82.07 Via
> Cantonale, 6928 Manno, Switzerland | Fax: +41 (91) 610.82.82
> _______________________________________________ Powered by
> www.kitware.com Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html Please keep
> messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView Follow this link to
> subscribe/unsubscribe:
> http://www.paraview.org/mailman/listinfo/paraview
> _______________________________________________ Powered by
> www.kitware.com Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html Please keep
> messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView Follow this link to
> subscribe/unsubscribe:
> http://www.paraview.org/mailman/listinfo/paraview -- John
> Biddiscombe, email:biddisco @ cscs.ch
> http://www.cscs.ch/ CSCS, Swiss National Supercomputing Centre | Tel:
> +41 (91) 610.82.07 Via Cantonale, 6928 Manno, Switzerland | Fax:
> +41 (91) 610.82.82 _______________________________________________
> Powered by www.kitware.com Visit other Kitware open-source projects
> at http://www.kitware.com/opensource/opensource.html Please keep
> messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView Follow this link to
> subscribe/unsubscribe:
> http://www.paraview.org/mailman/listinfo/paraview -- John
> Biddiscombe, email:biddisco @ cscs.ch
> http://www.cscs.ch/ CSCS, Swiss National Supercomputing Centre | Tel:
> +41 (91) 610.82.07 Via Cantonale, 6928 Manno, Switzerland | Fax:
> +41 (91) 610.82.82
> _______________________________________________ Powered by
> www.kitware.com Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html Please keep
> messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView Follow this link to
> subscribe/unsubscribe:
> http://www.paraview.org/mailman/listinfo/paraview
>
> -- John Biddiscombe, email:biddisco @ cscs.ch
> http://www.cscs.ch/ CSCS, Swiss National Supercomputing Centre | Tel:
> +41 (91) 610.82.07 Via Cantonale, 6928 Manno, Switzerland | Fax:
> +41 (91) 610.82.82
More information about the ParaView
mailing list