Making AAA a reality through FPGA
 

..

Go for it!

Honestly? A GREAT idea but software does not exist. Unless you are taking "if they build it, they will come" metaphor.

..

Whats' stopping somebody from making 3AC a reality? ;)

..

IIRC the AAA chip set was never finished by Commodore, or in particular by Dave Haynie who was working on it at the time Commodore went under.

But there is really nothing stopping anyone from working on the project (other than lack of interest) that I'm aware of.

Don't think Haynie was involved in the development of the chipset, it was more like him building prototype boards using those chips so yea he for sure knew in detail what didn't work properly yet.

..

But you made some assumptions upfront and they will limit Amiga 2400 - for example all chips can be made as one and they do not need to be limited in similar way to current chip-set.
See no point in repeating AAA limitations either - some features can be used to enrich legacy HW but still - better do some AAA features list - what is most important from your perspective.
Single chip can be cost optimal and perhaps in distant future physically achievable...

..

And what would you want to do with it?


I'm not being facetious.

..

Are you planning to write your own low level code to do so?


There's no software for any of this, the OS never supported it, no applications will get 16Bit audio, High color support, etc.


The hardware isn't the most important part.

Is it known if it was designed in a hardware description language?
If it was, have the sources survived somewhere?

I think it is a wonderful idea and you should combine it with the 3AC architecture to make a new world-class computer. You could call it the Amiga 3200 or Amiga 5000 depending on which casing form you select.

If you ask Electronic Arts, I’m sure they’ll be willing to create Deluxe Paint VI for AAA screen modes, and Team 17 could make a new Superfrog game with 16-bit sound.

Ask me if you need any help with anything – I’m really good at coming up with good names for custom chips.

..

Doesn't the Vampire SAGA offer these sorts of things? I have a vamp and it's great but I can't think of any software which actually uses these "advanced" features, and SAGA has now been out for many years.

SAGA is probably waay more useful than any AAA would be, but this whole idea sounds more like filling in the gaps in they history that might have been, rather than delivering something useful (like the AA3000+ motherboard with the DSP).
I admire the effort that went into the AA3000+ and I like the idea of doing these "if only" projects, and yes, I'd be impressed if some ppl would manage to re-create AAA in any workable fashion, but it's usefulness would really be limited to a niche within a niche unless some other magic also happens along the way (software support etc).

I would also say forget about AAA.
Just make an AA chipset with enormous bandwidth so that the CPU can access it as fast as it would access fast ram, superfast blitter, built-in scan doubler / flickerfixer.
Oh, and give Paula some more DMA slots so that one can use >28KHz sample rate in all screen modes. And give it a proper DAC too so that 14bit audio is always super clean.

Current software would immediately benefit from this and no new drivers needed.