This is an organic growing Cura Engine repository forked from Cura 15.04 legacy branch. My short-term goal is to experiment the feature that optimizes path planning with asymmetry traveling salesman solver. The long-term goal is to make legacy Cura Engine great again (no political implication here). Your contribution is always welcome.
I'm a hobbyist/maker/programmer/rebel/pro software freedom advocate. I don't work for Ultimaker company who owns Cura engine. Neither do I own an Ultimaker brand 3D printer.
I feel upset with the recent change of Cura engine (not the UI, I love its new UI) that always messed up slicing with overly complicated settings and deliver sub par planned path in GCode.
I want to make 3D printing accessible to ordinary, regardless which brand of 3D printer you own.
I like the fact that a profit driven company supports an open source software project. However, I'm strongly against the idea that an open source software project should be dictated by one single interest party, whose sole interest is to promote their hardware sales. So it is a different matter between authoritarian ownership vs providing financial/human resource support. Because there are obvious ethical problem and interest conflicts between repository owner and outside contributor.
If you feel the same way as I did, you are welcome to join with me.
If you are a non-techie 3D printer users like the most, you are welcome to try on this Cura Engine. I will post a release under Mac and Linux for the latest OS.
If you are a programmer who wants to contribute your brain power, I'm open to take a look of your PR. But I want to be clear that I will do it in the following manner:
I will have two branch in public: stable and dev branch.
The stable branch is the master branch. If your PR is related to support new hardware setting like dual extruders or a critical bug fix or performance improvement, chances are you are in this category. I will merge it into stable branch ASAP.
The dev branch is for experiment purpose. All dev branch should start with prefix dev- and a brief description of your feature. If you require a new print settings in the profile, chances are you are in dev branch. Things like changing path planning for one particular STL model belongs to this category. I will accept your PR as a separate branch for experimental purpose. After an extended long period of testing and usage, I will consider merging it to the stable branch.
The motivation behind this branching philosophy is that I'm a Occam's razors principle believer. My working experiences makes me believe that the solution with fewest assumption is always the best. We should limit the number of settings of slicing engine at all cost.
One can argue that engine needs to be as much flexible as possible. Thus, you need tons of settings to tweak. Well, if you are not a programmer but as an end user, this might be a fair statement. But this is a disaster for engineering if you need to maintain the damn code. Also, when the complexity of settings reach a certain level, no one (not even one single programmer) can understand the hack and the consequence of the choice you make. Image you have 100 settings and each have 3 options, you got 3^100 combination to tweak. Good luck to solve that by you. Unless you have have a quantum computer with AI can solve this NPC problem, otherwise Occam's razors principle rules.
Things I have done so far:
- Shared my 15.04 source code notes and 15.04 class UML in this wiki.
- Wrote a developer guide.
- Enhanced logging with G3LOG.
- Enabled CMake build.
- Enabled CI in Travis.
Things I will do next:
- Improve path planning
- Promote this legacy repository to non UM brand hackers
- Make legacy Cura Engine great again!
Happy hacking!
Ricky