Scaling along the z-axis should also be prohibited. E.g. emit a Warning message and don't scale along that axis for a 2D object. -- View this message in context: http://forum.openscad.org/text-2D-or-3D-tp12217p20351.html Sent from the OpenSCAD mailing list archive at Nabble.com.

As we can't render a mix of 2D and 3D objects, why they have to have volume? They are not 2-manifolds.

I got some obscure warning about scaling z being ignored for 2D when I wasn't scaling but rotating in 3D to bring them back to the Z plane and do a 2D difference. I had to recode and do the difference in 3D, even though I wanted a 2D result. So there are some warnings but they didn't make a sense to me. On 6 February 2017 at 23:02, adrian <adrianh.bsc@gmail.com> wrote: > Scaling along the z-axis should also be prohibited. E.g. emit a Warning > message and don't scale along that axis for a 2D object. > > > > -- > View this message in context: http://forum.openscad.org/ > text-2D-or-3D-tp12217p20351.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 agree changing to 0.01 would be an improvement, but I'd still like to see a warning (in addition). I was playing with a model for 3D printing and saw the (now familiar) Z-axis flickering where my text was exactly 1 unit thick and inset exactly 1 unit from the surface. As I decreased the thickness of my text below 1, nothing seemed to be happening. Above 1 things worked as expected, but below 1 (eventually I tried 0.000001) nothing seemed to be changing. Eventually I Googled the problem and found this thread with the explanation. So even if it gets changed to 0.01, somebody is going to get confused when they change it to a number below that and it doesn't behave as expected. OpenSCAD should issue a warning. -- View this message in context: http://forum.openscad.org/text-2D-or-3D-tp12217p20360.html Sent from the OpenSCAD mailing list archive at Nabble.com.

nophead wrote > I got some obscure warning about scaling z being ignored for 2D when I > wasn't scaling but rotating in 3D to bring them back to the Z plane and do > a 2D difference. I had to recode and do the difference in 3D, even though > I > wanted a 2D result. > On 6 February 2017 at 23:02, adrian &lt; > adrianh.bsc@ > &gt; wrote: >> Scaling along the z-axis should also be prohibited. E.g. emit a Warning >> message and don't scale along that axis for a 2D object. Very weird things happens when you rotate 2D designs about any axis other then Z. For instance: > module obj() > intersection(){ > rotate([30,0,0]) square(20); > rotate([25,0,0]) square(20); > } should be void if squares had 0 thickness for there is no 1D models in OpenSCAD. However, a preview of obj() is a wedge as if the squares were cubes. A 2D wedge, though! A render of obj() will show a 2D rectangle 20x17.4 or so in the XY plane. I can't understand neither its preview nor its render. Besides, linear_extrude() obj() will show a block whose base is the rectangle render() obj(). I don't think there is any reason to allow scales in Z or rotations about any axis different from Z for 2D objects. Or even translations outside the XY plane. 2D objects should live in the XY plane. Thickness of 2D objects isn't misleading only the user; it is misleading the OpenSCAD code. -- View this message in context: http://forum.openscad.org/text-2D-or-3D-tp12217p20363.html Sent from the OpenSCAD mailing list archive at Nabble.com.

See https://github.com/openscad/openscad/pull/1933 We can try to iteratively reach a decent solution from there. I started out just making the objects thinner. -Marius