Question about increasing accurancy of tessellations

[HouXiaoD]

Hello,

I have generated a tesr file by following this Tutorial with my DCT data.

However, when I try to generate a tessellation from the TESR file, the resulting structure looks quite different from the original TESR.

The command I used was:
neper -T -n from_morpho -domain "cube(1.82,1.85,3.36)" -morpho "tesr:file(dct-c.tesr)" -morphooptiobj "tesr:pts(region=surf,res=40)+val(bounddist)" -o dct-f1

$ neper -T -n from_morpho -domain "cube(1.82,1.85,3.36)" -morpho "tesr:file(dct-c.tesr)" -morphooptiobj "tesr:pts(region=surf,res=40)+val(bounddist)" -o dct-f1

========================    N   e   p   e   r    =======================
Info   : A software package for polycrystal generation and meshing.
Info   : Version 4.4.2-33
Info   : Built with: gsl|muparser|opengjk|openmp|nlopt|libscotch (full)
Info   : Running on 16 threads.
Info   : <https://neper.info>
Info   : Copyright (C) 2003-2022, and GNU GPL'd, by Romain Quey.
Info   : No initialization file found (`/home/bob/.neperrc').
Info   : ---------------------------------------------------------------
Info   : MODULE  -T loaded with arguments:
Info   : [ini file] (none)
Info   : [com line] -n from_morpho -domain cube(1.82,1.85,3.36) -morpho
         tesr:file(dct-c.tesr) -morphooptiobj
         tesr:pts(region=surf,res=40)+val(bounddist) -o dct-f1
Info   : ---------------------------------------------------------------
Info   : Reading input data...
Info   : Creating domain...
Info   : Creating tessellation...
Info   :     [i] Parsing file `dct-c.tesr'...
Info   :     [i] Parsed file `dct-c.tesr'.
Info   :       . Number of cells: 559
Info   :   - Setting seeds... 100%
100%
Info   :   - Generating crystal orientations...
Info   :       . Listing cell voxels... 100.0%
Warning:       . Sampling factor along x = 0.845274.  Setting to 1...
Warning:       . Sampling factor along y = 0.845274.  Setting to 1...
Warning:       . Sampling factor along z = 0.845274.  Setting to 1...
Info   :       . Number of voxels per cell: min = 3, avg = 2705, max = 38126.
Info   :   - Running tessellation...
Info   :     > Initial solution: f   =0.032326161
Info   :     > Iteration  88199: fmin=0.022209698 f=0.022224695
Info   :     > Iteration 157896: fmin=0.021421394 f=0.021428791 (2 subplex)
Info   :     > Iteration 160134: fmin=0.021421394 f=0.021421984 (3 subplex)
Info   :     > Iteration 171756: fmin=0.021326032 f=0.021327773 (4 praxis) 
Info   :     > Iteration 204855: fmin=0.021148024 f=0.021160787 (5 subplex)
Info   :     > Iteration 207093: fmin=0.021148024 f=0.021160343 (6 subplex)
Info   :     > Iteration 209331: fmin=0.021148024 f=0.021150233 (7 praxis) 
Info   :     > Iteration 209332: fmin=0.021148024 f=0.021150233 (7 praxis)
Info   :     > Final solution  : f   =0.021148024 (209332 iterations)
Warning:     > Failed due to unknown error.
Info   : Writing results...
Info   :     [o] Writing file `dct-f1.tess'...
Info   :     [o] Wrote file `dct-f1.tess'.
Info   : Elapsed time: 7836.593 secs.
========================================================================

I've attached both the TESR and tessellation images/files for reference.

Is there anything I might be missing, or are there any recommended options to improve the tessellation quality or better preserve the original structure?

attachmentFile.zip

Any advice would be greatly appreciated.

Best,
XiaodiHou

[rquey]

Looking at the two structures, I think it is about as good as it gets. I think the small grayish grain (near the orange one) on top, which "balloons" in the bottom image, is only a few voxels thick. It might be useful to remove it...

In general, removing the very small grains from the input polycrystal helps.

[HouXiaoD]

Thank you very much. I'll remove the very small grains and try again.