Hacker News Reader

Comment by jerf | Hacker News Reader

OK, yes, I get it, every major programming language can implement an emulator for DCPU-16 in a few dozen lines. Can someone hurry up and post the [consults the Trendy Language Calendar] Node implementation so we can call it a day on the DCPU-16 implementations?

(Incidentally, I'm not saying DCPU-16 is uninteresting, and if somebody's got something more interesting than a straight-up implementation I'm still all ears. But... Church-Turing, you know?)

View on HN

I agree.

I would be a lot more interested into high level languages for DCPU-16.

Also, most DCPU-16 implementations I have seen so far are bytecode interpreters. I would love to see an actual JIT translating emulator that generates x86 or even better llvm code. The performance difference should be considerable.

Someone has started: https://github.com/baconization/d16x

And:

https://github.com/Rotten194/Tenc

https://github.com/Blecki/DCPUC

I was going to write one using DynASM (it would be very short, simple, and fast) but it was unclear to me that DCPU-16 had any substantial programs written for it where you could show off a speed difference.

I have been looking for an excuse to write an article about how to use DynASM though.

I'd like to see an implementation in OrgASM, but it's hard to type with just one hand.

JIT compiling DCPU-16 bytecode using LLVM is something I plan to do later if I get time. I imagine by then someone else will already have done it, but if not, then its something I plan to do.

Come on, isn't it nice to see a Rosetta Stone for something that isn't math-oriented?

I think one of Notch's goals with DCPU-16 was bringing back the feeling of programming for computers with very limited specs and little complexity. I think this is proof that his plan is working.

Apparently the Stack Overflow questions are already rolling in: http://stackoverflow.com/questions/tagged/dcpu-16

I imagine this must be slightly entertaining to witness for Notch.

Reply on news.ycombinator.com