I spent many months redesigning, improving and rebuilding a prusa clone, not because I couldn't afford anything else, but precisely because I could afford to "waste" time and money learning and having fun.

Once I felt like the printer was in a useful state, I spent some more time getting it to a point where it could print nice ABS parts for a Voron Trident. Of course I couldn't just build a Trident. That would be too easy. Before I had even finished assembling it, it already had a number of bespoke modifications that I had designed.

And that's just the tip of the iceberg. Have you ever _looked_ at the klipper source code? It's like a fractal of weirdness. I mean it works, and clearly the person behind it is very knowledgeable about many things. But it makes for such an enormous and fun playground for improvements and redesigns. I've redesigned the entire build system for the firmware component. I've made the host component an actual (almost) normal python package. I changed a bunch of core aspects so that it could be packages for a linux distro. I am working on making the native helper a normal python native extension library too. And I am also writing some proper test rig for it.

And while I was at it, I started writing my own display software which doesn't use Wayland or X. It is going quite well actually. (Writing it in Rust)

This hobby (and, really, any hobby) has as much depth and obscurity as you are willing to look for.