Is this the problem? (not related to C= USA at all!)

[commodorejohn]

I don't see why it has to be only "keep up with the Joneses" or "stick to original equipment and give up on improvement." I think there's definitely room for improvement, and probably ways to make a boosted Amiga feasible for small-scale production, but I think that setting our sights on keeping up with the leading edge (or even the stable mid-range) of commodity PC hardware is pretty much unachievable at this point, unless one of us turns out to have a quirky billionaire uncle who suddenly dies and leaves us his entire estate. It'd be better to set sights on something less impressive but more achievable.

[commodorejohn]

Kinda depends on what you mean by a "true successor," then. Are you talking about a 68k-Amiga compatible machine? Is the idea to improve the existing Amiga experience, or recreate it from scratch on more powerful hardware? What is your goal?

[commodorejohn]

As long as everybody's laying out their ideas now, I might as well join in.

First off, I reject the idea that the infeasibility of keeping up with modern PC hardware means that we should just give up and either move to PCs or stick to vanilla Commodore hardware plus whatever expansions can be scrounged off eBay from an ever-shrinking pool of unclaimed equipment. That's silly.

I similarily reject the idea that a new Amiga should be designed to "take back the market," as if that's something you can even do on purpose. Let the Amiga be the Amiga, and if you want to revolutionize the industry with a radical new concept, build your own machine.

Also rejected is the idea that developing new hardware on an old standard is impossibly costly and doomed to fail. (Here's looking at you, Wolftothemoon!) Not only is that a bizarrely blanket statement, it's completely unsupported by the evidence. Plenty of retrocomputing communities have built and supported significant hardware projects, not least the Amiga community. The idea that building a whole computer is somehow vastly more difficult or expensive than building expansions that replace nearly every part of it is patently absurd. It might not be cheap, but it's not going to cost an arm and a leg and the body of your closest kin.

So what should it be? I say 68k-based, mostly because I find the 68k to be just awesome. I don't care whether it's a real 68k chip, a reimplementation, or if we somehow get ColdFire to work, it should be 68k. (With room left in the design for all you PPC freaks to add an accelerator, of course )

It should also be based on the ECS or AGA hardware, because again, totally awesome. An Amiga you can't root around in and twiddle registers for crazy effects is not truly an Amiga, I say. It should be expanded, though, at the very least in the sound department, and adding a chunky video mode couldn't hurt.

Finally, it should include built-in hardware for some or all of the obligatory expansion cards (IDE controller, Ethernet, USB, fast RAM above the Zorro space, etc.) No sense having people need to hunt that stuff down on eBay.

As for software? Whatever you can run on it, baby!