MK
Marius Kintel
Thu, Nov 17, 2016 3:53 AM
Thx!
You’re aware that this mesh is inside-out, right (check F12)?
Not that it changes anything related to triangulation..
-Marius
> On Nov 16, 2016, at 20:31, Ronaldo <rcmpersiano@gmail.com> wrote:
>
> Kintel, you asked for it:
>
Thx!
You’re aware that this mesh is inside-out, right (check F12)?
Not that it changes anything related to triangulation..
-Marius
MK
Marius Kintel
Thu, Nov 17, 2016 5:48 AM
Kintel, you asked for it:
This is a minimal test case:
polyhedron(points = [
[78.1831, 62.349, 10],
[97.4928, -22.2521, -5],
[-97.4928, -22.2521, 5],
[-78.1831, 62.349, -10],
[78.1831, 62.349, 200],
[97.4928, -22.2521, 200],
[-97.4928, -22.2521, 200],
[-78.1831, 62.349, 200]
],
faces = [
[0, 1, 4],
[1, 2, 5],
[2, 3, 6],
[3, 0, 7],
[5, 4, 1],
[6, 5, 2],
[7, 6, 3],
[4, 7, 0],
[4, 5, 6],
[4, 6, 7],
[3, 2, 1, 0],
]);
translate([200,0,0]) cube();
-Marius
On Nov 16, 2016, at 20:31, Ronaldo <rcmpersiano@gmail.com> wrote:
>
> Kintel, you asked for it:
>
This is a minimal test case:
polyhedron(points = [
[78.1831, 62.349, 10],
[97.4928, -22.2521, -5],
[-97.4928, -22.2521, 5],
[-78.1831, 62.349, -10],
[78.1831, 62.349, 200],
[97.4928, -22.2521, 200],
[-97.4928, -22.2521, 200],
[-78.1831, 62.349, 200]
],
faces = [
[0, 1, 4],
[1, 2, 5],
[2, 3, 6],
[3, 0, 7],
[5, 4, 1],
[6, 5, 2],
[7, 6, 3],
[4, 7, 0],
[4, 5, 6],
[4, 6, 7],
[3, 2, 1, 0],
]);
translate([200,0,0]) cube();
-Marius
MK
Marius Kintel
Thu, Nov 17, 2016 5:59 AM
> On Nov 16, 2016, at 20:31, Ronaldo <rcmpersiano@gmail.com> wrote:
>
> Kintel, you asked for it:
>
..and here’s an issue to follow: https://github.com/openscad/openscad/issues/1867
-Marius
P
Parkinbot
Thu, Nov 17, 2016 11:53 AM
I'm amazed to see that this error also occurs with a quad.
--
View this message in context: http://forum.openscad.org/Can-t-find-what-s-wrong-here-use-import-stl-and-difference-tp19123p19193.html
Sent from the OpenSCAD mailing list archive at Nabble.com.
I'm amazed to see that this error also occurs with a quad.
--
View this message in context: http://forum.openscad.org/Can-t-find-what-s-wrong-here-use-import-stl-and-difference-tp19123p19193.html
Sent from the OpenSCAD mailing list archive at Nabble.com.
RP
Ronaldo Persiano
Thu, Nov 17, 2016 3:03 PM
And even with quads almost planar:
polyhedron(points = [
[78.1831, 62.349, 1e-6],
[97.4928, -22.2521, 0],
[-97.4928, -22.2521, 0],
[-78.1831, 62.349, 0],
[78.1831, 62.349, 200],
[97.4928, -22.2521, 200],
[-97.4928, -22.2521, 200],
[-78.1831, 62.349, 200]
],
faces = [....
2016-11-17 9:53 GMT-02:00 Parkinbot rudolf@parkinbot.com:
And even with quads almost planar:
polyhedron(points = [
[78.1831, 62.349, 1e-6],
[97.4928, -22.2521, 0],
[-97.4928, -22.2521, 0],
[-78.1831, 62.349, 0],
[78.1831, 62.349, 200],
[97.4928, -22.2521, 200],
[-97.4928, -22.2521, 200],
[-78.1831, 62.349, 200]
],
faces = [....
2016-11-17 9:53 GMT-02:00 Parkinbot <rudolf@parkinbot.com>:
> I'm amazed to see that this error also occurs with a quad.
>
>
>
> --
> View this message in context: http://forum.openscad.org/Can-
> t-find-what-s-wrong-here-use-import-stl-and-difference-tp19123p19193.html
> Sent from the OpenSCAD mailing list archive at Nabble.com.
>
> _______________________________________________
> OpenSCAD mailing list
> Discuss@lists.openscad.org
> http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
>
MK
Marius Kintel
Thu, Nov 17, 2016 3:35 PM
On Nov 17, 2016, at 06:53, Parkinbot rudolf@parkinbot.com wrote:
I'm amazed to see that this error also occurs with a quad.
On Nov 17, 2016, at 10:03, Ronaldo Persiano rcmpersiano@gmail.com wrote:
And even with quads almost planar:
It’s a regresssion - I’m on it.
The most amazing is that nobody caught this regression until now (or it was masked by larger non-manifold issues).
-Marius
> On Nov 17, 2016, at 06:53, Parkinbot <rudolf@parkinbot.com> wrote:
>
> I'm amazed to see that this error also occurs with a quad.
>
> On Nov 17, 2016, at 10:03, Ronaldo Persiano <rcmpersiano@gmail.com> wrote:
>
> And even with quads almost planar:
It’s a regresssion - I’m on it.
The most amazing is that nobody caught this regression until now (or it was masked by larger non-manifold issues).
-Marius
JJ
Johan Jonker
Thu, Nov 17, 2016 6:19 PM
I don'know if it is still relevant, but I ran a test following the list
above;
test 6: replaced sweep with skin:
Output:
--
View this message in context: http://forum.openscad.org/Can-t-find-what-s-wrong-here-use-import-stl-and-difference-tp19123p19197.html
Sent from the OpenSCAD mailing list archive at Nabble.com.
I don'know if it is still relevant, but I ran a test following the list
above;
test 6: replaced sweep with skin:
Output:
--
View this message in context: http://forum.openscad.org/Can-t-find-what-s-wrong-here-use-import-stl-and-difference-tp19123p19197.html
Sent from the OpenSCAD mailing list archive at Nabble.com.
K
kintel
Thu, Nov 17, 2016 6:29 PM
This is now fixed in master.
I've only tested Ronaldo's example and my derivations thereof.
For those curious, this was a bug which caused a vertex to separate into
both its double and its downscaled to float version, in certain
circumstances, leaving a tiny gap which caused the resulting polyhedron to
be non-manifold.
-Marius
--
View this message in context: http://forum.openscad.org/Can-t-find-what-s-wrong-here-use-import-stl-and-difference-tp19123p19198.html
Sent from the OpenSCAD mailing list archive at Nabble.com.
This is now fixed in master.
I've only tested Ronaldo's example and my derivations thereof.
For those curious, this was a bug which caused a vertex to separate into
both its double and its downscaled to float version, in certain
circumstances, leaving a tiny gap which caused the resulting polyhedron to
be non-manifold.
-Marius
--
View this message in context: http://forum.openscad.org/Can-t-find-what-s-wrong-here-use-import-stl-and-difference-tp19123p19198.html
Sent from the OpenSCAD mailing list archive at Nabble.com.
R
Ronaldo
Thu, Nov 17, 2016 7:42 PM
Johan,
Both, Rudolf's sweep and Linde's skin, make assumptions on the order the
sections are input. I have not checked, but possibly they have adopted
opposed circulation order for the sections. The same happens with the module
show_closed_mesh I have defined some messages above and, as Kintel
mentioned, the surfaces were inside out for Rudolf's tricky model.
That is the reason why I usually add an additional parameter to the module
that build the polyhedron prescribing a reversion (or not) of the circular
order for each surface patch.
--
View this message in context: http://forum.openscad.org/Can-t-find-what-s-wrong-here-use-import-stl-and-difference-tp19123p19201.html
Sent from the OpenSCAD mailing list archive at Nabble.com.
Johan,
Both, Rudolf's sweep and Linde's skin, make assumptions on the order the
sections are input. I have not checked, but possibly they have adopted
opposed circulation order for the sections. The same happens with the module
show_closed_mesh I have defined some messages above and, as Kintel
mentioned, the surfaces were inside out for Rudolf's tricky model.
That is the reason why I usually add an additional parameter to the module
that build the polyhedron prescribing a reversion (or not) of the circular
order for each surface patch.
--
View this message in context: http://forum.openscad.org/Can-t-find-what-s-wrong-here-use-import-stl-and-difference-tp19123p19201.html
Sent from the OpenSCAD mailing list archive at Nabble.com.
