For some years now like a lot of you I’ve been building LED matrix cubes, there’s something about the cube geometry that always interested me, and the hzeller library just made this so accessible. I’ve been through the iterations of different hardware, 3D printing frames, different boards, I2c integration, etc. Now I’m focused back on a project to try to create a premium cube lamp, like a Glorb, but am starting slow (by that I mean the audio reactive stuff would be parked for now).
The idea for now is to take the led-image-viewer base and extend it to communicate over bluetooth\wifi, to allow a mobile app to push GIFs from Giphy etc to the cube, have overlays for battery levels etc, build playlists, which is fairly straightforward.
The thing that always seemed to stump me is the cube mapping of a 2D image to the cube faces, which is odd as it’s obviously a solved problem, I think I was trying to do this on the Pi rather than pre-render a cube compatible image. One added benefit is could scale and carve up the image to a 3 chain 2 parallel matrix friendly canvas.
One thing I’m hoping to do which I haven’t done before is check the performance with a Pi Zero 2 as I’ve always used the Pi 3 and 4’s, but that would make it a little cheaper (those damn panels fluctuate massively in price).
Anyway I’ve worked through (as others have done!) the basic mapping which I’m hoping to push to the cube and render it as a compatible stream for performance. This is how far I’ve got (in the attached link below), just curious if anyone else is interested as I’m hoping to push this forwards similar to what other people have done.
