It seems that OpenCSG PNG generation is also non-deterministic. Odd pixels
change at the edges on consecutive runs on the same computer. Perhaps it is
related to the STL problem. I.e. perhaps the tessellation is
non-deterministic and that leads to slightly different jaggies on diagonal
lines when the triangles are drawn.
--
Sent from: http://forum.openscad.org/
AI is taking over OpenSCAD? lol. I bet it's just a rounding error.
Sometimes in INKscape I set exact number say to 65 mm and later after
saving file it's 64.99998 or something like that. Maybe OpenSCAD
Preview cam has same type of error.
On 6/10/19, nophead nop.head@gmail.com wrote:
It seems that OpenCSG PNG generation is also non-deterministic. Odd pixels
change at the edges on consecutive runs on the same computer. Perhaps it is
related to the STL problem. I.e. perhaps the tessellation is
non-deterministic and that leads to slightly different jaggies on diagonal
lines when the triangles are drawn.
--
Sent from: http://forum.openscad.org/
OpenSCAD mailing list
Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
This is being run from the command line by a Python script so all the
numbers should be exactly the same each run, including the camera setting.
On Mon, 10 Jun 2019 at 18:29, Serge moneyforidea@gmail.com wrote:
AI is taking over OpenSCAD? lol. I bet it's just a rounding error.
Sometimes in INKscape I set exact number say to 65 mm and later after
saving file it's 64.99998 or something like that. Maybe OpenSCAD
Preview cam has same type of error.
On 6/10/19, nophead nop.head@gmail.com wrote:
It seems that OpenCSG PNG generation is also non-deterministic. Odd
pixels
change at the edges on consecutive runs on the same computer. Perhaps it
is
related to the STL problem. I.e. perhaps the tessellation is
non-deterministic and that leads to slightly different jaggies on
diagonal
lines when the triangles are drawn.
--
Sent from: http://forum.openscad.org/
OpenSCAD mailing list
Discuss@lists.openscad.org
http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org