puzzler
When I use F5, I see what I expect: a grid. When I use F6, I see a
totally filled in rectangle, with the interior spaces gone. If I only
use F6 on the cube at the end, it is fine; if I only use F6 on the
linear_extrude of the Grid, it is fine.
What am I missing?
inches = 25.4;
vt = 1; // vertical thickness
th = 0.5; // grid thickness
ht = 1 * inches; // 6
wd = 1 * inches; // 9
module Grid() {
* for (y = [0:3:ht])
translate([0, y, 0])
square([wd, th]);
for (x = [-ht:4:2ht])
translate([x, 0, 0])
rotate([0, 0, 45])
square([wd, th]);
for (x = [-ht:4:3ht])
translate([x, 0, 0])
rotate([0, 0, 135])
square([wd, th]);
}
intersection() {
cube([wd, ht, vt]);
linear_extrude(vt)
Grid();
}
ERROR: CGAL error in CGALUtils::applyBinaryOperator intersection: CGAL
ERROR: assertion violation! Expr: itl != it->second.end() File:
/mxe/usr/x86_64-w64-mingw32.static.posix/include/CGAL/Nef_3/SNC_external_structure.h
Line: 1152
This works though, and is much faster.
linear_extrude(vt)
intersection() {
square([wd, ht]);
Grid();
}
On Tue, 15 Sep 2020 at 15:28, jon jon@jonbondy.com wrote:
When I use F5, I see what I expect: a grid. When I use F6, I see a
totally filled in rectangle, with the interior spaces gone. If I only
use F6 on the cube at the end, it is fine; if I only use F6 on the
linear_extrude of the Grid, it is fine.
What am I missing?
inches = 25.4;
vt = 1; // vertical thickness
th = 0.5; // grid thickness
ht = 1 * inches; // 6
wd = 1 * inches; // 9
module Grid() {
- for (y = [0:3:ht])
translate([0, y, 0])
square([wd, th]);
for (x = [-ht:4:2ht])
translate([x, 0, 0])
rotate([0, 0, 45])
square([wd, th]);
for (x = [-ht:4:3ht])
translate([x, 0, 0])
rotate([0, 0, 135])
square([wd, th]);
}
intersection() {
cube([wd, ht, vt]);
linear_extrude(vt)
Grid();
}
OpenSCAD mailing list
Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
Thanks! in retrospect, your optimization seems obvious.
Is the CGAL error a bug?
:)
On 9/15/2020 1:42 PM, nop head wrote:
ERROR: CGAL error in CGALUtils::applyBinaryOperator intersection: CGAL
ERROR: assertion violation! Expr: itl != it->second.end() File:
/mxe/usr/x86_64-w64-mingw32.static.posix/include/CGAL/Nef_3/SNC_external_structure.h
Line: 1152
This works though, and is much faster.
linear_extrude(vt) intersection() { square([wd, ht]);
Grid(); }
On Tue, 15 Sep 2020 at 15:28, jon <jon@jonbondy.com
mailto:jon@jonbondy.com> wrote:
When I use F5, I see what I expect: a grid. When I use F6, I see a
totally filled in rectangle, with the interior spaces gone. If I only
use F6 on the cube at the end, it is fine; if I only use F6 on the
linear_extrude of the Grid, it is fine.
What am I missing?
inches = 25.4;
vt = 1; // vertical thickness
th = 0.5; // grid thickness
ht = 1 * inches; // 6
wd = 1 * inches; // 9
module Grid() {
* for (y = [0:3:ht])
translate([0, y, 0])
square([wd, th]);
for (x = [-ht:4:2*ht])
translate([x, 0, 0])
rotate([0, 0, 45])
square([wd, th]);
for (x = [-ht:4:3*ht])
translate([x, 0, 0])
rotate([0, 0, 135])
square([wd, th]);
}
intersection() {
cube([wd, ht, vt]);
linear_extrude(vt)
Grid();
}
_______________________________________________
OpenSCAD mailing list
Discuss@lists.openscad.org <mailto:Discuss@lists.openscad.org>
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
OpenSCAD mailing list
Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
On 9/15/2020 7:27 AM, jon wrote:
When I use F5, I see what I expect: a grid. When I use F6, I see a
totally filled in rectangle, with the interior spaces gone. If I only
use F6 on the cube at the end, it is fine; if I only use F6 on the
linear_extrude of the Grid, it is fine.
What am I missing?
Not that it directly helps you, but it looks like you didn't notice this
error in the console pane:
ERROR: CGAL error in CGALUtils::applyBinaryOperator intersection:
CGAL ERROR: assertion violation! Expr: itl != it->second.end() File:
/mxe/usr/x86_64-w64-mingw32.static.posix/include/CGAL/Nef_3/SNC_external_structure.h
Line: 1152
As I try to simplify it, I also get:
ERROR: CGAL error in CGALUtils::applyBinaryOperator intersection:
CGAL ERROR: assertion violation! Expr: cet->get_index() ==
ce->twin()->get_index() File:
/mxe/usr/x86_64-w64-mingw32.static.posix/include/CGAL/Nef_3/SNC_external_structure.h
Line: 1219
And:
ERROR: CGAL error in CGAL_Nef_polyhedron3(): CGAL ERROR: assertion
violation! Expr: e->incident_sface() != SFace_const_handle() File:
/mxe/usr/x86_64-w64-mingw32.static.posix/include/CGAL/Nef_S2/SM_const_decorator.h
Line: 329
I've also seen variations that just crash the program, and that yield
UI-WARNING: Object may not be a valid 2-manifold and may need repair!
Here's the simplest example that I've been able to come up with:
union() {
translate([-10,-10]) cube(1);
linear_extrude(1) {
translate([1.1, 0]) {
rotate([0, 0, 45])
square([25, 1.1]);
}
translate([1.1, 0]) {
rotate([0, 0, 135])
square([25.4, 1]);
}
}
}
Messing with the values - even the apparently irrelevant values like the
X dimensions of the squares - changes the behavior.
I think CGAL is upset that the two squares just barely touch each
other. (Or, possibly, that after the various rounding and quantization
effects they touch each other along only one edge.)
A workaround is to nudge one set of squares just a little so that they
no longer perfectly align with the other, e.g.:
inches = 25.4;
vt = 1; // vertical thickness
th = 0.5; // grid thickness
ht = 1 * inches; // 6
wd = 1 * inches; // 9
module Grid() {
* for (y = [0:3:ht])
translate([0, y, 0])
square([wd, th]);
for (x = [-ht:4:2*ht])
translate([x, 0.001, 0]) // <<<< nudge
rotate([0, 0, 45])
square([wd, th]);
for (x = [-ht:4:3*ht])
translate([x, 0, 0])
rotate([0, 0, 135])
square([wd, th]);
}
intersection() {
cube([wd, ht, vt]);
linear_extrude(vt)
Grid();
}