Sparse CSV - Delaunay3D filter Single to Multi threaded

ParaView Support
1

I happen to be using this filter alot and it seems to be single threaded. Is there a way to recompile paraview and utilise multithreading for this?

I am also in need to use this filter with a sparse csv which I would like to pad with 0 in order to be able to use this filter which otherwise simply fails. Is there a good way to do this?

Many thanks

2

AFAIK, the Delaunay3D filter has no multithreading capability at all.

3

Thank you for the reply.

Do you also have a recommendation for the second part of my question. It would help alot. here is some sample data.
inox.csv (188.8 KB)
I use table to points and visualise succesfully. I plot a histogram fine as well. But I would like to create a heatmap similar to the delaunay3d filter but due to the low number of points it doesn’t work.
Do you have a reccomendation to plot this data? Can you superimpose it as a heatmap on a second volume?
Many thanks

4

I’m not sure exactly what you are looking for. But you can map these columns to 3D points and apply Delaunay3D just fine. Please see the attached state file created with ParaView 5.13.0.
inox.pvsm (318.4 KB)

5

Sorry I am new to this and I am having a few issues.
So I am trying to load the state

image
image741×540 15.1 KB

and i get this error message
image
image974×204 24.1 KB

What am I doing wrong?
Ok I pointed it to the file directory and I am getting this
image
image1409×1048 178 KB

Apart from the mesh quality warning this isnt exactly what I want. I would like to get those coordinates resampled as a population in a wider space and then plot the delauney3d heat map from that.

6

I’m afraid I don’t understand the request. Could you post something that looks like the heat map you are looking for?

7

I have shared this previously but this is one example of a structured grid heat map.

image
image840×673 43.3 KB

from this array which consists of binned values across the range.
image
image959×824 66.2 KB

and I would like to do something similar with this data here:
image
image959×824 4.34 KB

which are coordinates of events im interested and somehow generate a similar heatmap as above. perhaps superimposed on to the volume in which this data originates.
image
image959×824 223 KB

Maybe this is a better image
image
image959×824 32.1 KB

image
image1746×1052 72.2 KB

This is the data histogram along one axis. I would like this in 3d over the volume if possible.

8

Thank you for clarifying. One option:

Point Set to Octree Image produces

image
image2856×1684 107 KB

It bins points into an image grid that you can then volume render to show density.

1 Like
9

Cheers. That’s quite nice. Thanks!
Is there perhaps a way to alter the range and size of the bins perhaps correlated to the volume of origin? Would I need to import that as a mesh of some kind?

1 Like