Top view after render <http://forum.openscad.org/file/n15824/2D_-Z_top.jpg> Side view - note negative Z <http://forum.openscad.org/file/n15824/2D_-Z_Right.jpg> Angled views, make it look like it is not aligned to the axis <http://forum.openscad.org/file/n15824/2D_-Z_angle_same.jpg> <http://forum.openscad.org/file/n15824/2D_-Z_angle_opposite.jpg> Is this fixable? ----- Newly minted Admin - PM me if you need anything, or if I've done something stupid... Unless specifically shown otherwise above, my contribution is in the Public Domain; to the extent possible under law, I have waived all copyright and related or neighbouring rights to this work. Obviously inclusion of works of previous authors is not included in the above. The TPP is no simple “trade agreement.” Fight it! http://www.ourfairdeal.org/ time is running out! -- View this message in context: http://forum.openscad.org/2D-view-negative-Z-tp15824.html Sent from the OpenSCAD mailing list archive at Nabble.com.

I just translated it to look OK... for the wiki <http://forum.openscad.org/file/n15825/Rotate_extrude_wiki_2D_B.jpg> ----- Newly minted Admin - PM me if you need anything, or if I've done something stupid... Unless specifically shown otherwise above, my contribution is in the Public Domain; to the extent possible under law, I have waived all copyright and related or neighbouring rights to this work. Obviously inclusion of works of previous authors is not included in the above. The TPP is no simple “trade agreement.” Fight it! http://www.ourfairdeal.org/ time is running out! -- View this message in context: http://forum.openscad.org/2D-view-negative-Z-tp15824p15825.html Sent from the OpenSCAD mailing list archive at Nabble.com.

On 23. jan. 2016 03:17, MichaelAtOz wrote: > Top view after render > <http://forum.openscad.org/file/n15824/2D_-Z_top.jpg> > Side view - note negative Z > <http://forum.openscad.org/file/n15824/2D_-Z_Right.jpg> > Angled views, make it look like it is not aligned to the axis > <http://forum.openscad.org/file/n15824/2D_-Z_angle_same.jpg> > <http://forum.openscad.org/file/n15824/2D_-Z_angle_opposite.jpg> > > Is this fixable? This seems paradoxical, or perhaps not? The problem relates to the value of the non-existent z coordinate in a 2d [x,y] coordinate system? I think 2d coordinates on OpenSCAD are slightly (just slightly) scizophrenic. 2d shapes are sometimes drawn/interpreted/used for a value of z != 0, showing that it really has a z value as in your example. When you rotate a 2d shape around for example the y-axis to give the coordinates z != 0, strange things start to happen. First, it makes a lot of sense: rotate([0,-45]) translate([15,0])circle(5); http://arnholm.org/tmp/openscad/circle2d_rotated.png Now you have a 2d circle placed in 3d space where the points on the circumference of the circle have different z-coordinates in the global system. It is also drawn correctly with a small unit thickness in the direction perpendicular to the transformed local 2d coordinate system. It is the local coordinate system of the circle that is 2d, not its placement or orientation in 3d space. The display this in OpenSCAD appears to support this understanding. But try this "extrude1": module extrude1() { color("green") linear_extrude(height=10) rotate([0,-45]) translate([15,0])circle(5); } extrude1(); For some reason, the (unstated) projection of the rotated circle down to the global coordinate system x-y plane is now taken as the starting point of an extrusion in global z direction. This is where the thing is (slightly) scizophrenic, the circle has and it hasn't values for its z-coordinates, and the direction of its thickness/direction of extrusion has changed. One could perhaps argue that a more consistent behaviour would be for the above to cause the same result as this "extrude2": module circle3d(r=1,dx=1,ry=1) { rotate([0,ry]) translate([15,0]) linear_extrude(height=0.001) circle(r); } module extrude2() { color("red") hull(){ circle3d(r=5,dx=15,ry=-45); translate([0,0,10]) circle3d(r=5,dx=15,ry=-45); } } extrude2(); That is, it is more consistent if you insist that linear_extrude operates only in z of the global 3d coordinate system. However, it is equally logical to expect that linear_extrude extrudes in the perpendicular direction of its 2d input, just like the original unit thickness drawing was hinting. After all the purpose in life of linear_extrude is to convert from 2d shapes to 3d solids. Therefore, accepting that transformed 2d shapes can exist in 3d space with coordinates z!=0, you might as well have expected the first extrusion to give a result equivalent to this: module extrude3() { color("blue") rotate([0,-45]) linear_extrude(height=10) translate([15,0]) circle(r=5); } extrude3(); http://arnholm.org/tmp/openscad/extrusion_confusion.png If you understand these inconsistencies, it is manageable. However, the behaviour of extrude1() is really unreasonable in my mind. I think it would have been more consistent and logical if extrude1() gave the same result as extrude3(), i.e. the extrusion was in the direction of the original drawn unit thickness of the transformed 2d circle, starting from the placement of the circle in 3d space. If you understand these inconsistencies it isn't a big problem, but I can imagine new users being very much confused by them. Carsten Arnholm

square(); produces a square below the XY plane with F6, so I think there is a bug drawing 2D objects on top of all the schizophrenia. On 23 January 2016 at 10:37, Carsten Arnholm <arnholm@arnholm.org> wrote: > On 23. jan. 2016 03:17, MichaelAtOz wrote: > >> Top view after render >> <http://forum.openscad.org/file/n15824/2D_-Z_top.jpg> >> Side view - note negative Z >> <http://forum.openscad.org/file/n15824/2D_-Z_Right.jpg> >> Angled views, make it look like it is not aligned to the axis >> <http://forum.openscad.org/file/n15824/2D_-Z_angle_same.jpg> >> <http://forum.openscad.org/file/n15824/2D_-Z_angle_opposite.jpg> >> >> Is this fixable? >> > > This seems paradoxical, or perhaps not? The problem relates to the value > of the non-existent z coordinate in a 2d [x,y] coordinate system? > > I think 2d coordinates on OpenSCAD are slightly (just slightly) > scizophrenic. 2d shapes are sometimes drawn/interpreted/used for a value of > z != 0, showing that it really has a z value as in your example. > > When you rotate a 2d shape around for example the y-axis to give the > coordinates z != 0, strange things start to happen. First, it makes a lot > of sense: > > rotate([0,-45]) translate([15,0])circle(5); > > http://arnholm.org/tmp/openscad/circle2d_rotated.png > > Now you have a 2d circle placed in 3d space where the points on the > circumference of the circle have different z-coordinates in the global > system. It is also drawn correctly with a small unit thickness in the > direction perpendicular to the transformed local 2d coordinate system. It > is the local coordinate system of the circle that is 2d, not its placement > or orientation in 3d space. The display this in OpenSCAD appears to support > this understanding. > > But try this "extrude1": > > module extrude1() > { > color("green") > linear_extrude(height=10) > rotate([0,-45]) > translate([15,0])circle(5); > } > extrude1(); > > For some reason, the (unstated) projection of the rotated circle down to > the global coordinate system x-y plane is now taken as the starting point > of an extrusion in global z direction. This is where the thing is > (slightly) scizophrenic, the circle has and it hasn't values for its > z-coordinates, and the direction of its thickness/direction of extrusion > has changed. > > One could perhaps argue that a more consistent behaviour would be for the > above to cause the same result as this "extrude2": > > module circle3d(r=1,dx=1,ry=1) > { > rotate([0,ry]) translate([15,0]) > linear_extrude(height=0.001) circle(r); > } > > module extrude2() > { > color("red") > hull(){ > circle3d(r=5,dx=15,ry=-45); > translate([0,0,10]) > circle3d(r=5,dx=15,ry=-45); > } > } > extrude2(); > > That is, it is more consistent if you insist that linear_extrude operates > only in z of the global 3d coordinate system. However, it is equally > logical to expect that linear_extrude extrudes in the perpendicular > direction of its 2d input, just like the original unit thickness drawing > was hinting. After all the purpose in life of linear_extrude is to convert > from 2d shapes to 3d solids. Therefore, accepting that transformed 2d > shapes can exist in 3d space with coordinates z!=0, you might as well have > expected the first extrusion to give a result equivalent to this: > > module extrude3() > { > color("blue") rotate([0,-45]) > linear_extrude(height=10) > translate([15,0]) circle(r=5); > } > extrude3(); > > http://arnholm.org/tmp/openscad/extrusion_confusion.png > > If you understand these inconsistencies, it is manageable. However, the > behaviour of extrude1() is really unreasonable in my mind. I think it > would have been more consistent and logical if extrude1() gave the same > result as extrude3(), i.e. the extrusion was in the direction of the > original drawn unit thickness of the transformed 2d circle, starting from > the placement of the circle in 3d space. > > If you understand these inconsistencies it isn't a big problem, but I can > imagine new users being very much confused by them. > > Carsten Arnholm > > > > _______________________________________________ > OpenSCAD mailing list > Discuss@lists.openscad.org > http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org >

On 01/23/2016 12:06 PM, nop head wrote: > square(); produces a square below the XY plane with F6, so I think > there is a bug drawing 2D objects on top of all the schizophrenia. > Yes, that one: https://github.com/openscad/openscad/issues/1202 ciao, Torsten.

> On Jan 23, 2016, at 06:18 AM, Torsten Paul <Torsten.Paul@gmx.de> wrote: > > On 01/23/2016 12:06 PM, nop head wrote: >> square(); produces a square below the XY plane with F6, so I think >> there is a bug drawing 2D objects on top of all the schizophrenia. >> > Yes, that one: > https://github.com/openscad/openscad/issues/1202 > Right, let’s just change it :) -Marius