Using, learning, programming and modding the Gigatron and anything related.
- Posts: 5
- Joined: 21 Jan 2020, 19:48
What I love about the Gigatron is the software tools. The ability to write real code (basic or whatever) and see it run. This is so much better than the labor intensive method of entering machine instructions into RAM or ROM.
But this all assumes the fixed instruction set of today. Has anyone thought about making the control unit completely pluggable? It is a bit hard with the existing board layout but the idea is to replace U11, U12, U13, U14, U15, U16, and diodes with a custom design. The first one could of course be 100% compatible with Gigatron. You might even want to include the RAM and I/O enable circuitry from the expansion bus idea.
The inputs are:
The outputs are:
- EL, EH, PL PH, IX, XL, YL, OL, LD
- AL, AR0-3
- /WE (actually /WE_RAM and /WE_PORT)
- /OE (actually /OE_RAM and OE_PORT]
- A15_RAM and A16_RAM
This is a total of 10 inputs and 20 outputs. For prototyping new instruction sets, you could build a board that uses a couple of programmable ATF22V10C chips (which run at up to 7.5ns) or even a ATF2500C PLD. The PLD (or an extra ATF22V10C) gives the ability to have more than 20 outputs for new control signals. Once happy with the design this can be then converted to a TTL logic and a diode array.
- Posts: 455
- Joined: 13 May 2018, 08:26
That's a cool idea! HG has proposed
a modified instruction set for his chess program. A cartridge would be neat.