blubee has joined #solvespace
<
blubee>
hello, i'm running solvespace: version 3.0~8f2af5b2
<
blubee>
on Linux localhost.localdomain 4.19.4-300.fc29.x86_64 #1 SMP Fri Nov 23 13:03:11 UTC 2018 x86_64 x86_64 x86_64 GNU/Linux
<
blubee>
the exported mesh has really bad triangles
<
blubee>
is there anything that I can do to get a better output mesh?
cr1901_modern has quit [Ping timeout: 246 seconds]
cr1901_modern has joined #solvespace
<
whitequark>
blubee: can you upload the slvs file somewhere?
<
whitequark>
blubee: what do you mean by bad triangles?
<
whitequark>
is the model too coarse?
<
whitequark>
i.e. the circles aren't circular?
<
whitequark>
go to "settings" and lower "export chord tolerance" in that case
<
blubee>
the wireframe after exporting triangle mesh
<
blubee>
tried both obj and stl
<
whitequark>
that's... not exactly bad
<
whitequark>
i mean, it is a legal mesh, not self-intersecting or anything
<
whitequark>
if you need to feed it to a tool that requires more regular triangles, you need to clean it up with something else
<
whitequark>
but afaik most tools don't care
<
whitequark>
it just looks weird
<
blubee>
i wanted to use solvespace to design parametric models
<
blubee>
which solvespace does very well
<
blubee>
from this mesh, I need to create quad or hexahedral volumetric mesh for FEA simulation
<
whitequark>
ok, i see
<
whitequark>
there is a number of tools that can clean up meshes. there is unfortunately no built-in solvespace function for that
<
whitequark>
you can click the "display mesh of solid model" button on the toolbar
<
whitequark>
and see that this is exactly the same as mesh used internally for rendering
<
blubee>
when i try to hover over the icons the text blinks
<
whitequark>
this is an issue with gtk3 unfortunately
<
whitequark>
either it is a bug in gtk3, or a change in gtk3 exposing a bug in solvespace
<
whitequark>
not sure which
<
blubee>
okay well i'll have to look into something that can better clean up the meshes
<
blubee>
i'll keep looking for a solution to this issue then
<
blubee>
solvespace is pretty solid otherwise
<
blubee>
well it's solid at parametric modeling with constraints
<
blubee>
i have meshlab
<
whitequark>
does it not clean up the mesh well?
<
whitequark>
iirc that's what i used
<
blubee>
i'll have to go back to it and try
<
blubee>
there are some algorithms to create normalized hexahedral meshes
<
whitequark>
hexahedral meshes?
<
whitequark>
what are those?
<
blubee>
regular cubes
<
blubee>
instead of triangles
<
blubee>
meshlab is still having a hard time dealing with the exported mesh
<
whitequark>
unfortunately i don't think there is anything i can easily do with solvespace
<
blubee>
i'll have to keep on working at this
<
whitequark>
you can tick "force nurbs surface to triangle mesh"
<
whitequark>
but from what i can see, the mesh becomes even worse
<
whitequark>
well, it is different, and worse.
<
blubee>
they all make the meshes more complex
<
blubee>
inside of solvespace how are the geometric objects represented?
<
whitequark>
either directly as NURBS surfaces or directly as triangle meshes
<
whitequark>
well, it always triangulates for rendering
<
whitequark>
but it operates on either NURBS or meshes depending on the group option
<
whitequark>
I think your only viable option is to clean up the existing mesh it creates
<
blubee>
that's cool that nurbs are used under the hood
<
whitequark>
it's a blessing and a curse...
<
whitequark>
NURBS operations tend to be fragile
blubee has quit [Ping timeout: 240 seconds]
blubee has joined #solvespace
<
blubee>
whitequark: thanks for the help
blubee has quit [Quit: leaving]
_whitelogger has joined #solvespace
castaway has quit [Remote host closed the connection]
castaway has joined #solvespace