There's a no-go with an implemented core using the actual FPGAs that could outperform a plain 020 with AGA.
I thought the FPGA arcade replay (whatever it's called) and the vampire 600 was doing better than plain 020 already & they are using relatively old and cheap FPGA's.
There's a no-go with real native hardware fully compatible with AGA, see the S-AGA on the Natami.
Natami is not a good example. Once you have AGA fully compatible then some extensions would probably not be too hard. You have to be pragmatic to ship something like that, which is something they weren't particularly good at. The same problem the AAA team at commodore had.
At the end nothing like a new Amiga portable hardware will be made.
FPGA's can't achieve (for now) what we want at an affordable price.
Design new native compatible chips and bake them is stratospherically expensive.
I can't see there being anywhere near enough demand that baking would be cost effective. It also takes away some of the benefits, as you can't ever make it do anything different.
I think a laptop/netbook version of the FPGA Arcade is much more likely to happen. Once the standard version has shipped then mikej should look into crowd funding that.
