Sign of life of Apollo FPGA project

[psxphill]

I understand the A600 was targeted as it was the machine with fewest upgrade options.

I thought it was because that was just what he had when he started. IMO it doesn't make much sense to pick the least popular Amiga just because it didn't have upgrade options. He could of course just be a masochist.

I'm unsure how much difference there is between the A600 and A1200/3000/4000 bus

Conceptually it's similar because it's a parallel cpu bus, but there are plenty of differences between them. A600 is only 16 bit, the rest are 32bit. A3000/A4000 cpu bus is asynchronous, while A600 is synced to the chipset.

but I think, due to the simplicity of the board design (thanks to the FPGA) it shouldn't be too difficult to design cards for other machines.

If it were simple then why has nobody else done it? There are off the shelf FPGA cores.

Once the core is stable and available I think (hope!) things may snowball...

Once the A600 version is stable they can start again and design new hardware and write new code for the next machine. Of course there will likely be a chunk of code that can be reused, but it's still a whole lot of work.

[psxphill]

Well there was the Vampire 1 last year, I think the hardware is not the problem, it is the development of the fpga cpu.

One person in the world has developed two versions so far and you think hardware isn't the problem? There are far more people developing cores for minimig/mist/fpgaarcade and there are fpga cpu cores you can take off the shelf. It's easier and cheaper to get into HDL side, prototyping and designing boards is a whole different game.

It shouldn't be too hard to translate the A600 Vampire 2 into an A1000/A500 board, so I'd expect to see that next as it might sell a few. Maybe an A2000 board if someone is interested, but as it's going to be closed source then you will need to get someone in the inner sanctum interested as it will probably require a few tweaks. Probably won't sell many but the effort should be low.

An A3000/4000 card is going to be a lot of work as it's a much faster and more complex cpu slot, different buster revisions to deal with etc. The high cost of buying a few of them for R&D and potentially damaging them, compared to the low number of people who would buy one is likely to put them off.

A1200 is probably in between complexity wise and is probably the card that will sell the most. However the different revisions of motherboards are likely to create stability issues and they might want to avoid that altogether.

I am not sure what "Just to mention that this board will never enter serial production. It will be used only for development purpose. " means, but it might be that you need to wait for V3 to be able to buy one.