Gigatron Hackers

As I've said elsewhere, an additional PC might be the best investment for extra performance, so long as there are specialized instructions for using it. Since it would be chained like the main PC, it is doing 16-bit additions itself. So the lack of an ADC instruction would less likely impact general...

I've thought more about the above. What would be nice would be if this could be a native instruction: Signal=0 ? PC = (RAM @ vPC++) * 16 : PC = Vector That improves the previously proposed native instruction that acts as vCPU Fetch/Dispatch. The idea is to put the native JMP based on RAM @ vPC as th...

Another Possible Strategy I've thought more about the special instruction one could add in their implementation. It would require a special counter-register for itself (vPC) that is used as a MAR set into RAM, which is used to do paragraph-aligned jumps and post-increment vPC. Put another way: PC =...

I've been churning this idea more in my head. Since vCPU treats things like a microcoded machine, it makes sense to throw more hardware at that. The biggest advancement could be to add another 16-bit program counter. Dedicate that one to the vCPU. Have an instruction that will use the new counter as...

I mentioned microcoded CPUs because the Gigatron's firmware ROM is essentially microcode as far as the vCPU goes. I asked myself that and mentioned that I asked myself about microcode because that gave clues about where to add more ICs to make tighter vCPU interpretation. Someone mentioned wanting a...

The video line is managed by the code in page 2 of the ROM. The row is tracked by zero page parameter videoY and frameCount is used as a simple system clock. There isn't a column value since the column is just 160 ora([Y,Xpp],OUT) instructions in a row with no time to do anything else. Thanks. I do...

I've been refining my thoughts. Hardware vCPU Support, Continued If you want to make a machine like a Gigatron closer to Von Neumann, one of the first steps would be to make the rest of the machine not rely on bit-banging. After designing for that, give a Giga-similar machine a ROM paragraph jump. I...

I haven't explored it. Making a 25 MHz VGA circuit on a PCB is pushing things for most people. You'd need tricks such as using 74ACT logic and likely pipelining. If the Gigatron is driving it, it would be simpler in that you wouldn't need a sync circuit, and the CPU has no problem with producing the...

Okay, I've come up with an idea in my mind for multiplication. It would take 8 cycles plus any overhead. The idea is 8/8/16 unsigned. I realized that you can do it the same way as you multiply by hand. It is just place, add, and shift. The way this would work is to use shift registers and adders, an...

Thank you Monosonite! I still haven't started as I'm trying to work it out in my mind. My focus is now more on what amenities can be added to the base hardware to make interpretation easier. Things like: 1. Proper flags and related instructions. If we had proper ADC/SBB instructions, chaining additi...