We’ve often thought that it must be harder than ever to learn about computers. Every year, there’s more to learn, so instead of making the gentle slope from college mainframe, to Commodore 64, to IBM PC, to NVidia supercomputer, you have to start at the end. But, really, you don’t. You can always emulate computers from simpler times, and even if you don’t need to, it can be a lot of fun.
That’s the idea behind the MonTana mini-computer. It combines “…ideas from the PDP-11, MIPS, Scott CPU, Game Boy, and JVM to make a relatively simple 16-bit computer…”
The computer runs on Java, so you can try it nearly anywhere. The console is accessed through a web browser and displays views of memory, registers, and even something that resembles a Game Boy screen. You’ll need to use assembly language until you write your own high-level language (we’d suggest Forth). There is, however, a simple operating system, MTOS.
This is clearly made for use in a classroom, and we’d love to teach a class around a computer like this. The whole thing reminds us of a 16-bit computer like the PDP-11 where everything is a two-byte word. There are only 4K bytes of memory (so 2K words). However, you can accomplish a great deal in that limited space. Thanks to the MTOS API, you don’t have to worry about writing text to the screen and other trivia.
It looks like fun. Let us know what you’ll use it for. If you want to go down a level, try CARDIAC. Or skip ahead a little, and teach kids QBasic.
WASM-4 is another good option.
There are a ton of fantasy consoles, the best ones offer an interesting twist.
For me, I like having cycle times attached to each instruction. So I kind of avoided stuff like PICO-8. And ended up settling on GB Studio, which gives me some of the convenience of a good fantasy console but it can target actual hardware too.
On MonTana, I love the idea. I’ve often wondered if some of the mini-computer architectures could have made it in a different timeline. Actually TI TMS9900 and GI CP1600 were 16-bit architectures from the 1970’s that managed to make it into game consoles and toys in the 1980’s.
In a way, I would have preferred TMS9900 in micro controllers over Atmel AVR, but TI never really understood the hobbyists or small volume market. TI was very good at making high volumes for big contractors, they were always rather weak on end-user marketing or even direct-to-engineer marketing.
Intel famously out maneuvered other silicon manufacturers by simply handing out a summary datasheets at conventions and during sales meetings that had the meat of what the chip does on it. (I think the 4 page 8086/8088 datasheet was the first example of that?)
I do love simply doing GBA dev. there’s tonc, emulators with GDB, toolchains…
and with a proper flashcard you can even try on real hardware…
Credibility drops instantly to 0%