I did try the algorithm on soemthing that's closer to a flat surface and it seems to do better. I also found the paper that I referenced so the coefficients are theoretically OKish now. The facet count goes up by a factor of 4 with every generation, so there's a limit to how much OPENscad can handle. On Mon, Oct 14, 2024 at 03:30:06AM -0400, discuss-request@lists.openscad.org wrote: > Date: Sun, 13 Oct 2024 08:47:09 +0100 > From: nop head <nop.head@gmail.com> > Subject: [OpenSCAD] Re: Tool idea for matching complex curves > To: OpenSCAD general discussion Mailing-list > <discuss@lists.openscad.org> > Message-ID: > <CAEEmnrPX8jniFra2o--h8CYNHpUs8cqFJkj00Xc8MmnAW3aA0w@mail.gmail.com> > Content-Type: multipart/related; > boundary="0000000000001c979b062456eef1" > > Interesting but I think it isn't quite right. It seems to generate a > surface texture, possibly where each subdivision meets the next one. > > [image: image.png] > > On Sun, 13 Oct 2024 at 01:17, Nathan Hellweg via Discuss < > discuss@lists.openscad.org> wrote: > > > Not sure if this will show up in a thread or not... but I did a proof > > of concept level interpolating subdivision. I'll try sending it as an > > attachment and see what happens. > > > > -- > > > > _______________________________________________ > > OpenSCAD mailing list > > To unsubscribe send an email to discuss-leave@lists.openscad.org > >