Thinking about giving this a go or maybe staying with just a DOS install
Recently played a lot of PS1 titles again as my PS1 died. Also using it to play with old computer operating systems.
- metal slug on neo geo
- various amiga games
- Shining Force on the Genesis
- checking out the Atari ST scene (the enemy, I had an amiga back in the day)
- going through the SNES library to see what I missed
- played with dragster on the 2600 core for some time
Eagerly awaiting the Saturn core… I want to play through shining force 3, I have a few Saturns in various states of disrepair.
I am half tempted to get a CRT when the Saturn Core goes stable for Virtua Cop
When you make something yourself, you don’t deal with pre-fab boards. You start with all the raw video formats and signals and work up to ADCs and converters and serializers. Mike Chi is doing a lot of that work but has no interest open sourcing his efforts. In fact, he’s censoring his efforts with a clown emoji because of the Chinese grey market.
That’s how my current employer do things (contractor in industry). We create a bunch of HVDC control products that require FPGAs to get the tight timing loops required for safety.
When we develop a new product we start with a development board, develop all the FPGA algorithms on that board, and in the meanwhile make our own PCB while sourcing components. Granted this is a multi-billion dollar company so we can afford to pay for both PCB development and bulk chip orders.
Best is to start with the algorithms, so look for an FPGA board with display out so you can validate the design. This is the hard part, HLS helps some but VHDL or Verilog knowledge is pretty much a necessity. If you know how to program an embedded system with FreeRTOS or similar, the step to FPGA programming is not very far away.
Hardware is the least sexy part, you have the PCIe bus (or USB C port) as input and a Displayport as output, a few extra things like clock and memory, trace it in a PCB program, boom done. Obviously I’m simplifying and there is still a ton of craftmanship going into the PCB, but it’s more an issue of connecting A to B than anything else, ideally in a way that can be cheaply and efficiently mass produced.
If you can do it in an FPGA over a PCIe bus you can then transfer that design to an ASIC. Again, a lot of non-sexy work required to get this to work, like creating the equivalent of a mold for serial production, but nothing that will require heavy R&D resources.
Soooo… Yeah, start with buying a dev board and work your way from there, is my advice.
Not to be mean, but maybe you should make a separate FPGA development thread.
This is specifically about the Mister project.