[Coco] CC-Five (was Re: Pseudo CoCo4???) (LONG)
jcewy at swbell.net
Mon Jan 22 10:22:19 EST 2007
Mark McDougall wrote:
> Joel Ewy wrote:
>> WARNING: Extended musings about hypothetical enhanced CoCo compatible
>> follow. This is not a short note.
> Are you familiar with the C-One project Joel?
> It's pretty much *exactly* what you envisage for the Coco.
IIRC that's essentially a C-64 implemented in FPGA. I think it uses a
separate (non-FPGA) CPU in a plug-in board. The FPGA mainly emulates
the rest of the Commodore hardware. I think the CPU is one of the
later, more powerful incarnations of the 6502.
> Sadly, it's pretty much a failure, though it could be argued that it is
> largely due to the original architect's failure to deliver. I have a
> theory or two of my own too..
I'm aware of it and have looked at the web sites a few times, but
haven't really been following along closely. In what sense is it a
failure? Technically? Commercially? I think we would have to come at
it with modest expectations. The CoCo is probably never again going to
be a mass-marketed home computer. On the other hand, those of us
already loony enough to still be using and talking about the Color
Computer 16 years after Rad Shack quit caring about it can see the value
in a computer that does some things very differently from what has
become the norm in the PC world.
> If such a project was ever seriously considered by the Coco community,
> they would do well to learn a few lessons from the C-one before setting
> anything in concrete.
That's a good point. What little I've read about it suggests that the
thing works, so it's not a technical issue. Maybe the problem, as you
imply, stems from an individual's failure to deliver. Probably nobody's
going to get rich off a CC-Five (or whatever). It isn't going to
support any large, full-time businesses. But this is why I think that
the whole idea should be decoupled from a particular hardware product.
I think it should exist as a specification, which can be realized in
emulation, and in an FPGA design -- preferably one that uses as much
off-the-shelf hardware as possible. If the hardware design is a flop,
it could still be implemented in emulation. In fact, the emulation
would be simple. It would probably be done long before the hardware.
So in a sense, this would be a way of not setting anything in concrete,
just making some forms one could use for casting concrete later. :)
> Having said all that - bring it on! First step would be getting a
> working Coco 1 FPGA implementation though...
I agree. The System09 on opencores.org would likely be a good start,
though it isn't particularly CoCo-like yet. It runs FLEX. Why not
start by porting NitrOS-9 to it? I think there is an effort afoot in
Australia to make a CoCo 1 in FPGA.
More information about the Coco