Empathy List Archives

R

runsun

Tue, Jul 28, 2015 5:15 PM

nophead wrote

Here is a version that works. I added the vertices at 30 to the facets
they
touch on the top and bottom faces.

This means that my idea "can't render multiple polyhedra at the same time"
is wrong.

Gotta go back to check everything I have and re-write my history :( ....
Thx.

At the mean time. this raises another question, though: If it's not valid,
why the rendering is successful w/o error when there's only one poly ?

$ Runsun Pan, PhD

$ -- libs: doctest , faces ( git ), offliner ( git );

tips: hash( 1 , 2 ), sweep , var

$ -- Linux Mint 17.1 Rebecca x64 + OpenSCAD 2015.03.15/2015.04.01.nightly

--
View this message in context: http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13347.html
Sent from the OpenSCAD mailing list archive at Nabble.com.

nophead wrote > Here is a version that works. I added the vertices at 30 to the facets > they > touch on the top and bottom faces. This means that my idea "can't render multiple polyhedra at the same time" is wrong. Gotta go back to check everything I have and re-write my history :( .... Thx. At the mean time. this raises another question, though: If it's not valid, why the rendering is successful w/o error when there's only one poly ? ----- $ Runsun Pan, PhD $ -- libs: doctest , faces ( git ), offliner ( git ); tips: hash( 1 , 2 ), sweep , var $ -- Linux Mint 17.1 Rebecca x64 + OpenSCAD 2015.03.15/2015.04.01.nightly -- View this message in context: http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13347.html Sent from the OpenSCAD mailing list archive at Nabble.com.

NH

nop head

Tue, Jul 28, 2015 5:30 PM

I guess when there is just one tree node containing a polyset OpenScad
simple renders it. As soon as there is more than one node it uses CGAL to
do at least a top level implicit union and CGAL needs manifold objects.

On 28 July 2015 at 18:15, runsun runsun@gmail.com wrote:

nophead wrote

Here is a version that works. I added the vertices at 30 to the facets
they
touch on the top and bottom faces.

This means that my idea "can't render multiple polyhedra at the same time"
is wrong.

Gotta go back to check everything I have and re-write my history :( ....
Thx.

At the mean time. this raises another question, though: If it's not valid,
why the rendering is successful w/o error when there's only one poly ?

$ Runsun Pan, PhD

$ -- libs: doctest , faces ( git ), offliner ( git );

tips: hash( 1 , 2 ), sweep , var

$ -- Linux Mint 17.1 Rebecca x64 + OpenSCAD 2015.03.15/2015.04.01.nightly

--
View this message in context:
http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13347.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

I guess when there is just one tree node containing a polyset OpenScad simple renders it. As soon as there is more than one node it uses CGAL to do at least a top level implicit union and CGAL needs manifold objects. On 28 July 2015 at 18:15, runsun <runsun@gmail.com> wrote: > nophead wrote > > Here is a version that works. I added the vertices at 30 to the facets > > they > > touch on the top and bottom faces. > > This means that my idea "can't render multiple polyhedra at the same time" > is wrong. > > Gotta go back to check everything I have and re-write my history :( .... > Thx. > > At the mean time. this raises another question, though: If it's not valid, > why the rendering is successful w/o error when there's only one poly ? > > > > > ----- > > $ Runsun Pan, PhD > > $ -- libs: doctest , faces ( git ), offliner ( git ); > > tips: hash( 1 , 2 ), sweep , var > > $ -- Linux Mint 17.1 Rebecca x64 + OpenSCAD 2015.03.15/2015.04.01.nightly > > > > > -- > View this message in context: > http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13347.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 >

T

Trygon

Tue, Jul 28, 2015 9:13 PM

Hi,
Thanks for all your input Peter, Nophead & Runsun, much appreciated!
Interesting explanation from Nophead as to why it renders OK with only one
Polyhedron, but not with two.

Here is version 2 of the problem, now no vertices that are mid-facet on the
adjacent face. Same problem as before, renders fine with one polyhedron,
fails with two.

Thanks again for your comments...

// If set to false, draws 1 polyhedron, renders OK, imports to MeshLab OK.
// If set to true, draws 2 identical polyhedra, on render:
// "WARNING: Object may not be a valid 2-manifold and may need repair!"
Show_Rendering_Error=true;

p=[[0,-10,-5],[2,-10,-5],[8,-10,-5],[8,10,-5],[2,10,-5],[0,10,-5],[2,0,-5],
[0,-10,5], [2,-10,5], [8,-10,5], [8,10,5], [2,10,5], [0,10,5], [2,0,5]];

f=[[8,7,13],[13,12,11],[9,8,10],[8,11,10],
[6,0,1],[4,5,6],[3,1,2],[3,4,1],
[7,8,1],[8,9,2],[9,10,3],[10,11,4],[11,12,5],[12,13,6],[13,7,0],
[1,0,7],[2,1,8],[3,2,9],[4,3,10],[5,4,11],[6,5,12],[0,6,13]];

polyhedron(points=p,faces=f);
if (Show_Rendering_Error) translate([15,0,0]) polyhedron(points=p,faces=f);

--
View this message in context: http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13358.html
Sent from the OpenSCAD mailing list archive at Nabble.com.

Hi, Thanks for all your input Peter, Nophead & Runsun, much appreciated! Interesting explanation from Nophead as to why it renders OK with only one Polyhedron, but not with two. Here is version 2 of the problem, now no vertices that are mid-facet on the adjacent face. Same problem as before, renders fine with one polyhedron, fails with two. Thanks again for your comments... // If set to false, draws 1 polyhedron, renders OK, imports to MeshLab OK. // If set to true, draws 2 identical polyhedra, on render: // "WARNING: Object may not be a valid 2-manifold and may need repair!" Show_Rendering_Error=true; p=[[0,-10,-5],[2,-10,-5],[8,-10,-5],[8,10,-5],[2,10,-5],[0,10,-5],[2,0,-5], [0,-10,5], [2,-10,5], [8,-10,5], [8,10,5], [2,10,5], [0,10,5], [2,0,5]]; f=[[8,7,13],[13,12,11],[9,8,10],[8,11,10], [6,0,1],[4,5,6],[3,1,2],[3,4,1], [7,8,1],[8,9,2],[9,10,3],[10,11,4],[11,12,5],[12,13,6],[13,7,0], [1,0,7],[2,1,8],[3,2,9],[4,3,10],[5,4,11],[6,5,12],[0,6,13]]; polyhedron(points=p,faces=f); if (Show_Rendering_Error) translate([15,0,0]) polyhedron(points=p,faces=f); -- View this message in context: http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13358.html Sent from the OpenSCAD mailing list archive at Nabble.com.

BC

Bob Cousins

Tue, Jul 28, 2015 10:20 PM

On 28 July 2015 at 22:13, Trygon db5765@outlook.com wrote:

Hi,
Thanks for all your input Peter, Nophead & Runsun, much appreciated!
Interesting explanation from Nophead as to why it renders OK with only one
Polyhedron, but not with two.

Here is version 2 of the problem, now no vertices that are mid-facet on the
adjacent face.

Are you sure? I still see some bad triangles. Every edge should be shared
by exactly 2 triangles. It's not enough to have 2 edges shared with 1 edge
of a third triangle.

You need to follow a formal algorithm to triangulate successfully. Have a
look at articles like
http://www.gamedev.net/page/resources/_/technical/graphics-programming-and-theory/polygon-triangulation-r3334

There is an online polygon builder tool, I think we really need a
polyhedron builder tool!

On 28 July 2015 at 22:13, Trygon <db5765@outlook.com> wrote: > Hi, > Thanks for all your input Peter, Nophead & Runsun, much appreciated! > Interesting explanation from Nophead as to why it renders OK with only one > Polyhedron, but not with two. > > Here is version 2 of the problem, now no vertices that are mid-facet on the > adjacent face. > Are you sure? I still see some bad triangles. Every edge should be shared by exactly 2 triangles. It's not enough to have 2 edges shared with 1 edge of a third triangle. You need to follow a formal algorithm to triangulate successfully. Have a look at articles like http://www.gamedev.net/page/resources/_/technical/graphics-programming-and-theory/polygon-triangulation-r3334 There is an online polygon builder tool, I think we really need a polyhedron builder tool!

NH

nop head

Tue, Jul 28, 2015 10:26 PM

Same problem. Change the second vertex from 2,-1-5 to 3,-10,5 to see a hole
open up. That hole is there topologically when it is at 2 as well. It just
has zero area.

On 28 July 2015 at 22:13, Trygon db5765@outlook.com wrote:

Hi,
Thanks for all your input Peter, Nophead & Runsun, much appreciated!
Interesting explanation from Nophead as to why it renders OK with only one
Polyhedron, but not with two.

Here is version 2 of the problem, now no vertices that are mid-facet on the
adjacent face. Same problem as before, renders fine with one polyhedron,
fails with two.

Thanks again for your comments...

// If set to false, draws 1 polyhedron, renders OK, imports to MeshLab OK.
// If set to true, draws 2 identical polyhedra, on render:
// "WARNING: Object may not be a valid 2-manifold and may need repair!"
Show_Rendering_Error=true;

p=[[0,-10,-5],[2,-10,-5],[8,-10,-5],[8,10,-5],[2,10,-5],[0,10,-5],[2,0,-5],
[0,-10,5], [2,-10,5], [8,-10,5], [8,10,5], [2,10,5], [0,10,5], [2,0,5]];

f=[[8,7,13],[13,12,11],[9,8,10],[8,11,10],
[6,0,1],[4,5,6],[3,1,2],[3,4,1],
[7,8,1],[8,9,2],[9,10,3],[10,11,4],[11,12,5],[12,13,6],[13,7,0],
[1,0,7],[2,1,8],[3,2,9],[4,3,10],[5,4,11],[6,5,12],[0,6,13]];

polyhedron(points=p,faces=f);
if (Show_Rendering_Error) translate([15,0,0]) polyhedron(points=p,faces=f);

--
View this message in context:
http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13358.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

Same problem. Change the second vertex from 2,-1-5 to 3,-10,5 to see a hole open up. That hole is there topologically when it is at 2 as well. It just has zero area. On 28 July 2015 at 22:13, Trygon <db5765@outlook.com> wrote: > Hi, > Thanks for all your input Peter, Nophead & Runsun, much appreciated! > Interesting explanation from Nophead as to why it renders OK with only one > Polyhedron, but not with two. > > Here is version 2 of the problem, now no vertices that are mid-facet on the > adjacent face. Same problem as before, renders fine with one polyhedron, > fails with two. > > Thanks again for your comments... > > // If set to false, draws 1 polyhedron, renders OK, imports to MeshLab OK. > // If set to true, draws 2 identical polyhedra, on render: > // "WARNING: Object may not be a valid 2-manifold and may need repair!" > Show_Rendering_Error=true; > > p=[[0,-10,-5],[2,-10,-5],[8,-10,-5],[8,10,-5],[2,10,-5],[0,10,-5],[2,0,-5], > [0,-10,5], [2,-10,5], [8,-10,5], [8,10,5], [2,10,5], [0,10,5], [2,0,5]]; > > f=[[8,7,13],[13,12,11],[9,8,10],[8,11,10], > [6,0,1],[4,5,6],[3,1,2],[3,4,1], > [7,8,1],[8,9,2],[9,10,3],[10,11,4],[11,12,5],[12,13,6],[13,7,0], > [1,0,7],[2,1,8],[3,2,9],[4,3,10],[5,4,11],[6,5,12],[0,6,13]]; > > polyhedron(points=p,faces=f); > if (Show_Rendering_Error) translate([15,0,0]) polyhedron(points=p,faces=f); > > > > > > -- > View this message in context: > http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13358.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 >

T

Trygon

Tue, Jul 28, 2015 10:43 PM

Hi Bob,

Whoops... having re-checked it in MeshLab, yes you are right.

Thanks, I had better tweek my mesh builder.

--
View this message in context: http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13361.html
Sent from the OpenSCAD mailing list archive at Nabble.com.

Hi Bob, Whoops... having re-checked it in MeshLab, yes you are right. Thanks, I had better tweek my mesh builder. -- View this message in context: http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13361.html Sent from the OpenSCAD mailing list archive at Nabble.com.

AP

Andrew Plumb

Tue, Jul 28, 2015 10:52 PM

Wherever you have four or more co-planar points, you can group them into multi-point faces instead of just triangles:

f=[ [0,1,2,3,4,5,6]
,[13,12,11,10,9,8,7]
,[7,8,9,2,1,0]
,[10,11,12,5,4,3]
,[12,13,6,5]
,[13,7,0,6]
,[9,10,3,2]
];

…although collapsing the angled face ([12,13,7,0,6,5]) seems to work too:

f=[ [0,1,2,3,4,5,6]
,[13,12,11,10,9,8,7]
,[7,8,9,2,1,0]
,[10,11,12,5,4,3]
,[12,13,7,0,6,5]
,[9,10,3,2]
];

I forget which release started supporting this, but it’s been there for a while now.

Andrew.

On Jul 28, 2015, at 5:13 PM, Trygon db5765@outlook.com wrote:

Hi,
Thanks for all your input Peter, Nophead & Runsun, much appreciated!
Interesting explanation from Nophead as to why it renders OK with only one
Polyhedron, but not with two.

Here is version 2 of the problem, now no vertices that are mid-facet on the
adjacent face. Same problem as before, renders fine with one polyhedron,
fails with two.

Thanks again for your comments...

// If set to false, draws 1 polyhedron, renders OK, imports to MeshLab OK.
// If set to true, draws 2 identical polyhedra, on render:
// "WARNING: Object may not be a valid 2-manifold and may need repair!"
Show_Rendering_Error=true;

p=[[0,-10,-5],[2,-10,-5],[8,-10,-5],[8,10,-5],[2,10,-5],[0,10,-5],[2,0,-5],
[0,-10,5], [2,-10,5], [8,-10,5], [8,10,5], [2,10,5], [0,10,5], [2,0,5]];

f=[[8,7,13],[13,12,11],[9,8,10],[8,11,10],
[6,0,1],[4,5,6],[3,1,2],[3,4,1],
[7,8,1],[8,9,2],[9,10,3],[10,11,4],[11,12,5],[12,13,6],[13,7,0],
[1,0,7],[2,1,8],[3,2,9],[4,3,10],[5,4,11],[6,5,12],[0,6,13]];

polyhedron(points=p,faces=f);
if (Show_Rendering_Error) translate([15,0,0]) polyhedron(points=p,faces=f);

--
View this message in context: http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13358.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

Wherever you have four or more co-planar points, you can group them into multi-point faces instead of just triangles: f=[ [0,1,2,3,4,5,6] ,[13,12,11,10,9,8,7] ,[7,8,9,2,1,0] ,[10,11,12,5,4,3] ,[12,13,6,5] ,[13,7,0,6] ,[9,10,3,2] ]; …although collapsing the angled face ([12,13,7,0,6,5]) seems to work too: f=[ [0,1,2,3,4,5,6] ,[13,12,11,10,9,8,7] ,[7,8,9,2,1,0] ,[10,11,12,5,4,3] ,[12,13,7,0,6,5] ,[9,10,3,2] ]; I forget which release started supporting this, but it’s been there for a while now. Andrew. > On Jul 28, 2015, at 5:13 PM, Trygon <db5765@outlook.com> wrote: > > Hi, > Thanks for all your input Peter, Nophead & Runsun, much appreciated! > Interesting explanation from Nophead as to why it renders OK with only one > Polyhedron, but not with two. > > Here is version 2 of the problem, now no vertices that are mid-facet on the > adjacent face. Same problem as before, renders fine with one polyhedron, > fails with two. > > Thanks again for your comments... > > // If set to false, draws 1 polyhedron, renders OK, imports to MeshLab OK. > // If set to true, draws 2 identical polyhedra, on render: > // "WARNING: Object may not be a valid 2-manifold and may need repair!" > Show_Rendering_Error=true; > > p=[[0,-10,-5],[2,-10,-5],[8,-10,-5],[8,10,-5],[2,10,-5],[0,10,-5],[2,0,-5], > [0,-10,5], [2,-10,5], [8,-10,5], [8,10,5], [2,10,5], [0,10,5], [2,0,5]]; > > f=[[8,7,13],[13,12,11],[9,8,10],[8,11,10], > [6,0,1],[4,5,6],[3,1,2],[3,4,1], > [7,8,1],[8,9,2],[9,10,3],[10,11,4],[11,12,5],[12,13,6],[13,7,0], > [1,0,7],[2,1,8],[3,2,9],[4,3,10],[5,4,11],[6,5,12],[0,6,13]]; > > polyhedron(points=p,faces=f); > if (Show_Rendering_Error) translate([15,0,0]) polyhedron(points=p,faces=f); > > > > > > -- > View this message in context: http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13358.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

T

Trygon

Wed, Jul 29, 2015 5:48 AM

Nophead,
Thanks for the clear explanation of the problem - zero area hole, move
vertex & hole opens up - I like that.

--
View this message in context: http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13363.html
Sent from the OpenSCAD mailing list archive at Nabble.com.

Nophead, Thanks for the clear explanation of the problem - zero area hole, move vertex & hole opens up - I like that. -- View this message in context: http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13363.html Sent from the OpenSCAD mailing list archive at Nabble.com.

T

Trygon

Wed, Jul 29, 2015 3:41 PM

"Ear clipping" algorithm now incorporated into my mesh building code as
suggested by BobC
(http://www.gamedev.net/page/resources/_/technical/graphics-programming-and-theory/polygon-triangulation-r3334).

Now all working correctly.

--
View this message in context: http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13366.html
Sent from the OpenSCAD mailing list archive at Nabble.com.

"Ear clipping" algorithm now incorporated into my mesh building code as suggested by BobC (http://www.gamedev.net/page/resources/_/technical/graphics-programming-and-theory/polygon-triangulation-r3334). Now all working correctly. -- View this message in context: http://forum.openscad.org/Rendering-error-using-polyhedron-tp13336p13366.html Sent from the OpenSCAD mailing list archive at Nabble.com.

discuss@lists.openscad.org

Rendering error using polyhedron()