This is probably more an FYI than anything, but when I started working on the builtin_time function, I ran some timing tests. I had several nurb_length runs that were running in roughly half a second, and the same code run in v2025.02.10 is now taking 10x as long. I'll probably start off with a previous version and see if I can figure out when this happened. If anyone has an idea of what might be up, please let me know. EBo --

We did upgrade the 2D subsystem to Clipper2 December 19th: https://github.com/openscad/openscad/pull/5501. It’s possible that there are performance regressions related to 2D operations. If you could test with immediately before and immediately after that, it could give us an idea. If you find a regression, please share a small but representative example. Opening a GitHub issue is a good steÎp forward in that case. -Marius > On Feb 10, 2025, at 17:31, John David via Discuss <discuss@lists.openscad.org> wrote: > > This is probably more an FYI than anything, but when I started working on the builtin_time function, I ran some timing tests. I had several nurb_length runs that were running in roughly half a second, and the same code run in v2025.02.10 is now taking 10x as long. I'll probably start off with a previous version and see if I can figure out when this happened. If anyone has an idea of what might be up, please let me know. > > EBo -- > _______________________________________________ > OpenSCAD mailing list > To unsubscribe send an email to discuss-leave@lists.openscad.org

Marius, I'll give this a quick try... <crumbunny> somehow I hosed one of my libraries bouncing around and I hosed git. I'll have to drop playing with this until I get my system back in running order. On Mon, Feb 10, 2025 at 9:00 PM Marius Kintel <marius@kintel.net> wrote: > We did upgrade the 2D subsystem to Clipper2 December 19th: > https://github.com/openscad/openscad/pull/5501. It’s possible that there > are performance regressions related to 2D operations. If you could test > with immediately before and immediately after that, it could give us an > idea. > If you find a regression, please share a small but representative example. > Opening a GitHub issue is a good steÎp forward in that case. > > -Marius > > > On Feb 10, 2025, at 17:31, John David via Discuss < > discuss@lists.openscad.org> wrote: > > > > This is probably more an FYI than anything, but when I started working > on the builtin_time function, I ran some timing tests. I had several > nurb_length runs that were running in roughly half a second, and the same > code run in v2025.02.10 is now taking 10x as long. I'll probably start off > with a previous version and see if I can figure out when this happened. If > anyone has an idea of what might be up, please let me know. > > > > EBo -- > > _______________________________________________ > > OpenSCAD mailing list > > To unsubscribe send an email to discuss-leave@lists.openscad.org > >