How to make sure the minimun scale larger than 0.001 in two-scale 3D Tess?

[mqc007]

Hi,
I'm trying to build a tess with two-scale in a cylinder (0.050,0.100). I hope it can contain more 4000 cells.
so here is my code

neper -T -n 1000::4 -domain "cylinder(0.1,0.05)" -morpho gg -reg 1 -fmax 0 -rsel 3 -group "(scaleid(1))" -statcell diameq -format tess,3dec -o c1000x4

the point is i have to know the exactly two numbers of the two-scale for example 1000::4, because i want to group the cells in 3DEC.
when i use the V-Module to picture the tessellation i found that the minimum scale is about 0.0005. the maxmum is larger than 0.013.
another thing is, if the scale of the minimun cell is less than 0.001 it will be very hard to continue my analyse in 3DEC.

so is there any possible to make sure the minimum scale larger than 0.001 under these conditions?

best regards
many thanks

[rquey]

First thing first, there is no reason why you would not be able to group cells (for 3DEC) while having different numbers of subcells from one primary cell to another.

For example:

neper -T -n 100::1000*area -dim 2 -morpho gg -group "scaleid(1)" -o test
neper -V test.tess -datacellcol group -print img

In this example, the number of subcells of a cell is proportional to its surface area.

You can even generate equal-size subcells (centroidal is a fast way to do it):

Both examples will clearly lead to cell sizes that are closer to each other than what you get with 1000::4.

(Also note that -reg 1 -fmax 0 is almost equivalent to no regularization at all).

[mqc007]

many thanks.
I will try

[mqc007]

@rquey Sir, I did try with the code diameq:interval(0.001,1) but I went something wrong, the minimum sacle is still samller than 0.001
here is my code
neper -T -n 1000::4 -morpho "diameq:interval (0.002,0.012)" -domain "cylinder(0.1,0.05)" -morpho gg -reg 1 -fmax 0 -rsel 3 -group "(scaleid(1))" -statcell diameq -format tess,3dec -o 1000x4new
and V module code neper -V 1000x4new.tess -datacellcol "real:file(1000x4new.stcell)" -datacellscale 0.001:0.003:0.005:0.007:0.009:0.011:0.013 -imagesize 1920:1080 -datacellscaletitle "Cell size(m)" -datacelltrs 0.5 -print 1000x4new

Sorry to bother you.

[rquey]

Please copy text as raw text, not images...

You have to specify diameq:interval(0.002,0.012) at scale 2, so, assuming you would keep gg at scale 1: -morpho "gg:diameq:interval(0.002,0.012)".

[mqc007]

right. of course i need to specify two scales. thank you very much!

[mqc007]

@rquey Sir, I'm back. the code doesn't work, it says i have discovered a bug in neper.
Info : [ 2] 797/800 Info : - Setting seeds... 100% Info : - Generating crystal orientations... Info : - Running tessellation... Info : > Initial solution: f =0.974178833 Info : > Iteration 491: fmin=0.793700527 f=0.793700535 Info : > Final solution : f =0.793700527 (491 iterations) Info : > Reached eps' criterion.
Info : [ 2] 798/800 Info : - Setting seeds... 100%
Info : - Generating crystal orientations...
Info : - Running tessellation...
Info : > Initial solution: f =0.961398256
Info : > Iteration 566: fmin=0.793700526 f=0.793700532
Info : > Final solution : f =0.793700526 (566 iterations)
Info : > Reached eps' criterion. Info : [ 2] 799/800 Info : - Setting seeds... 100% Info : - Generating crystal orientations... Info : - Running tessellation... Info : > Initial solution: f =0.994133089 Info : > Iteration 525: fmin=0.793700574 f=0.793700645 Info : > Final solution : f =0.793700574 (525 iterations) Info : > Reached eps' criterion.
Info : [ 2] 800/800 Info : - Setting seeds... 100%
Info : - Generating crystal orientations...
Info : - Running tessellation...
Info : > Initial solution: f =0.993893260
Info : > Iteration 329: fmin=0.793701397 f=0.793701486
Info : > Final solution : f =0.793701397 (329 iterations)
Info : > Reached `eps' criterion.
Info : Flattening 2-scale tessellation...
Info : [level] Flattening 2-scale tessellation...
Info : [ 1] 1/1: id=1
Error : You have discovered a bug in Neper! Please file an issue at
https://github.com/rquey/neper/issues. Thank you.

`

I tied 1000::4 and 800::4, both failed.
the code i used:
neper -T -n 800::4 -morpho "gg::diameq:interval (0.001,0.10)" -domain "cylinder(0.1,0.05)" -reg 1 -fmax 0 -rsel 3 -group "(scaleid(1))" -statcell diameq -format tess,3dec -o 1000x4new