Arciel Rekman

89 Followers
131 Following
295 Posts
Programmer @ Epic Games, relapsed demoscener, recovering Linux user, Z80 coder
PC intro must be worse than the Amiga one
Amiga did it better
DLSS 5 commentary by Eugenius Stratilatovas
This is a piece of history - a 3D modelling program for the ZX Spectrum 48K from 1982 (released the same year as the computer). I vaguely remember using it (much, much later of course) to create some crude logo, but I have never had the original casette for it. It even came with pretty detailed instructions, huh.

RE: https://mastodon.gamedev.place/@floooh/115900619627455204

Template metaprogramming is the bane of C++. Why did the author of that code go that route?

C++ compilers will be probably the first to introduce LLM-based error output.

I ported my fork of #z80ai to #ZXSpectrum. Now it indeed ran at 3.5MHz (CP/M version on #ZXSpectrumNext must have been using 28Mhz). This simple convo took takes 4.5 minutes :)

(Optimizations are surely possible. I also pessimized it a bit by adding border colors just to not be bored waiting for a reply)

Grab the source and .tap file here: https://github.com/RCL/z80ai/tree/main/examples/tinychat/build_tap

Speed example
It is of course very primitive. If you have #ZXSpectrumNext , you can run examples out of the box in the Next standard CP/M (you may need to build CHAT.COM but it's a breeze). At 3.5Mhz the token generation speed isn't great of course. #z80ai

Ok, so here is a working 40KB language model for Z80. It mght be a good occasion to put the tech under the microscope and understand it through and through.

https://github.com/HarryR/z80ai

GitHub - HarryR/z80ai: Z80-μLM is a 2-bit quantized language model small enough to run on an 8-bit Z80 processor. Train conversational models in Python, export them as CP/M .COM binaries, and chat with your vintage computer.

Z80-μLM is a 2-bit quantized language model small enough to run on an 8-bit Z80 processor. Train conversational models in Python, export them as CP/M .COM binaries, and chat with your vintage compu...

GitHub