CuraEngine VS Cura

Here (at the bottom of the page) you can find an overview of the calculation and the git repository also holds the entire source code of the Cura Engine (which is the part of Cura that actually does the slicing and G-code creation) where the calculations happen. The repository is in C++, but the calculations are actually universal.

There have been some improvements but I would not say it's fixed yet. 😕 It is my favorite 5.x seam for sure! You can read about the work done here: https://github.com/Ultimaker/CuraEngine/pull/1762

If you can figure git and out how to compile Cura and CuraEngine from source, you can test it by using the "vms" branch on both my CuraEngine and Cura github forks.

If you want to contribute by testing it you can find the necessary commits for Cura here and here for CuraEngine.

There's also a pull request on the main Cura repository, seems like it's nearly ready to be merged in: https://github.com/Ultimaker/CuraEngine/pull/1282

I am currently developing a pipeline for creating STL files, slice them and create a print job based on it. My setup at the moment is an Ultimaker S5, which offers a simple REST-API, for example starting a job with a G-Code or UFP file. I am using the cadQuery library for creating parametrized 3d shapes and export them as a STL file. I want to use the CuraEngine CLI interface (Backend for Cura) for slicing or their library libArcus (python bindings) but there is literally no documentation or any kind of examples, except the source code. There is also the prucaSlicer, which also offers a CLI interface for creating G-Code, but no support for the Ultimaker S5.

Bridges are hard to detect because of the workflow of the slicing engine. To put it simply: a bridge cannot be detected until the perimeter is layed down. But since both in the Cura Engine (which btw is now also at the core of the perimeters in PS) and in the Slic3r the Area generator operates simultaneously on periters and skin, it's impossible to inject in the pipeline and change the skin orientation. For two reasons: -First, they cannot detect a bridge because the start and end point are arbitrary and not static. It does make sense for us, but not for the algorithm. -Second, they cannot define an orientation. It cannot see that is a rectangle, and that therefore it should be best to lay the skin following a path parallel to the longest side.

You might be able to use Curaengine to provide weight estimates.

If you can figure git and out how to compile Cura and CuraEngine from source, you can test it by using the "vms" branch on both my CuraEngine and Cura github forks.