Conformal secondary tessellations

[xinitol]

Hello everyone!

I am new to Neper and I am currently trying to understand some of the hierarchical tessellation examples available in the official documentation/website (for instance, https://neper.info/generation.html).
In particular, I am trying to understand how to generate a two-scale tessellation where each parent grain contains secondary subdivisions (https://neper.info/doc/neper_t.html).

Considering a simple example, neper -T -n 100::10, I noticed that the secondary tessellations are not conformal across adjacent parent grains. In other words, the secondary domains on opposite sides of a grain boundary do not share the same vertices/edges on the common interface, leading to non-matching boundaries. For example (picture taken here https://neper.info/generation.html, where red circles shows where this non conformity happens):

I think this becomes problematic for the kind of meshing workflow that are coarser in comparison to the one proposed here https://neper.info/doc/tutorials/simple_model.html.

I was wondering whether there is some more advanced way to control this behavior, possibly through more advanced usage of -reg,
or constrained regions, or some other option that I may have missed. I also tried looking at -sel and -rsel, but from what I understand these options only seem to control the size/scale of the secondary tessellations, not the conformity of interfaces.
So my main question is if there is a way in Neper to generate conformal secondary tessellations across adjacent parent grains.
Or alternatively, is this currently not supported by the hierarchical tessellation approach?

If anyone could point me to relevant documentation, examples, references, or alternative workflows, I would really appreciate it.

Thanks a lot!

[rquey]

Hi @xinitol. No, it is not supported. Even if you tried to enforce it, it would lead to a degenerated configuration that the current algorithm would not be able to handle properly.