CubingB is a timer/analyzer for speedsolving Rubik's cubes (and related puzzles). It focuses on supporting "smart cubes" (i.e. bluetooth cubes) for recording the exact moves of a solve in real time.
CubingB is at a very early stage, and only supports macOS, and only the MoYu Weilong AI for smart cube functionality. It's written in Python, and uses PyQt and PyOpenGL for all sorts of graphical goodness, and pyobjc for binding to macOS' native CoreBluetooth APIs for communicating with smart cubes. It uses a SQLite database, managed with SQLAlchemy and Alembic.
This is just a hobby project written for my own use, so don't expect an easy installation experience or much technical support. If you're not a programmer that knows how to install Python packages, don't bug me with issues yet, please.
- Normal cubing timer features
- Basic scramble generation, single/aoX calculations, etc.
- Full random state scramble support
- Support for Square-1 and FTO (face-turning octahedron) scrambles
- Sweet tooltip diagrams that pop up in the middle of scrambles in case you lose your place
- Graphs: view graphs of a statistic (like "ao12" or "best ao100" or "solves per day") over time for one or more sessions. Also, see a list of the top solves for one or more sessions
- Session management: renaming, drag-n-drop to reorder, and merge sessions
- Algorithm database: see a big list of all algorithms, mark ones you know
- Smart cube support
- Full incremental turn and gyroscope data recorded and timestamped throughout solves on a smart cube. This can be viewed later like a video, with play/pause/scrubbing. The data is stored in a compact binary format, but can still accumulate rather quickly (about 1K of data for 2-3 seconds of solving)
- Algorithm training: choose an algorithm set, then solve the shown case as fast as you can. Then do it again.
- Algorithm drilling: do the same algorithm over and over, and try to improve how fast you can do it
- Click-and-drag to rotate, scroll to zoom on 3D cube viewer
- Smart cube hotkeys: r to reset cube state (i.e. make the virtual cube solved), c to calibrate gyroscope (takes the current gyroscope position and assumes its the standard white top, green front orientation)
- Misc:
- All solve data stored in a SQLite database for easy analytics (if you're nerdy enough to know SQL)
- CSTimer importing (no interface, just a Python script for now)
- No limits, no internet required, and no giving up your solve data. All your data is yours and yours alone.
- Free and open source, yo
In the near future, the priority will be in analyzing solves (i.e. making sane reconstructions incorporating the turn and gyroscope data, tracking algorithm performance, etc.), as well as creating training exercises. That's the whole reason I wanted a smart cube in the first place. This other stuff was just the basics to get a decent timer that's good enough for typical usage.
- Cross platform support: PyQt can theoretically run on most platforms. Qt supports bluetooth, but apparently it doesn't support the BTLE advertisement/scanning, which I think is needed for the MoYu cube (at least I couldn't get it to work).
- Other smart cube support: support for the GAN i3 should be coming soon-ish, I got one but need to do some reverse engineering (much trickier than the MoYu). Other than that, we'll see.
- Minor weirdnesses with rotations/gyroscope: no idea! I'm out of my depth on the math here at the moment. If you know how quaternions work, I'd love some help!
CubingB doesn't work! That sucks!
What does CubingB mean? Well, it could potentially stand for CubingBuddy, CubingBenchmarker, or CubingBehemoth, but really it's just a dumb variation on a dumb joke