You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I noticed that the number of nodes printed by gmsh (when reading a .msh-file for example) doesn't match the "number of points" displayed in Paraview's information tab. The same behavior happens if we use the built-in CartesianDiscreteModel.
The below code shows, that exporting a (2,2)-quad triangulated unit square leads to a VTU file, which has some nodes duplicated. Instead of the required 9 nodes (see attached image), we get 16 nodes.
I think that instead of using connectivity information, the writevtk(Ω, "trian")-function simply places every element inside the "world". In contrast to this, the writevtk(model, "model") works as expected.
Is this behavior intended? Couldn't we save a lot of memory by using the connectivity information as well and not writing nodes multiple times?
Or do I have a fundamental error in thinking about unstructured meshes? 😄
I noticed that the number of nodes printed by
gmsh
(when reading a.msh
-file for example) doesn't match the "number of points" displayed in Paraview's information tab. The same behavior happens if we use the built-inCartesianDiscreteModel
.The below code shows, that exporting a (2,2)-quad triangulated unit square leads to a VTU file, which has some nodes duplicated. Instead of the required 9 nodes (see attached image), we get 16 nodes.
I think that instead of using connectivity information, the
writevtk(Ω, "trian")
-function simply places every element inside the "world". In contrast to this, thewritevtk(model, "model")
works as expected.Is this behavior intended? Couldn't we save a lot of memory by using the connectivity information as well and not writing nodes multiple times?
Or do I have a fundamental error in thinking about unstructured meshes? 😄
I use Julia 1.8.2 and Gridap 0.17.16.
The text was updated successfully, but these errors were encountered: