I would kinda love to do one of those "bootstrap" things where you start with machine code and build an interpreter which you then use to build a compiler and so on and so forth
@SuricrasiaOnline forth is really good for this since you can implement a good forth in about 2000 bytes
@SuricrasiaOnline you can also build an actual forth compiler out of a working forth interpreter rather easily and use that to start making binary utilities and stuff
@SuricrasiaOnline this is reminding me of a thought i had over the weekend:
retired: writing emulators for retro gaming consoles.
inspired: inventing a retro gaming console architecture then writing an emulator and games for it.
@SuricrasiaOnline i've got a sweet one that's been on the design table for years now, motorola 68000 processor, vector display, the works! basically, the best that the mid 80s had to offer ;)
i have no clue if this hardware would ever work, probably someone more talented than i could do it, but that's not the point! it only has to work in my bizarro machine emulation <insert mad scientist laugh> XD
@elmiko I put together a cpu architecture/emulator a while ago which also had a syscall instruction. the emulator also emulated a posix kernel when the syscall instruction was invoked. however I only implemented a very small handful of the system calls
@SuricrasiaOnline My dream project is to do this from an even lower level - design the machine code, build a processor and computer around it, then get started with programming it.
The Ben Eater CPU project's made this seem a lot more achievable.
ｃｙｂｒｅｓｐａｃｅ: the social hub of the information superhighway
jack in to the mastodon fediverse today and surf the dataflow through our cybrepunk, slightly glitchy web portal