[Coco] CoCo4! 50% done!

[Joel Ewy]

On 02/05/2014 01:34 PM, Aaron Wolfe wrote:
> [snip]
> I agree $200 is about the limit for such a device.  [...]  We also
> have the same issue of competing with the well established CoCo 3 standard
> for mindshare... if the platform isn't significantly better then why
> upgrade at all, yet if the platform is too different then its "not a coco".
>
> Personally I think most of the upgrades seen in CoCo3FPGA are the low
> hanging fruit that makes the most sense.  A 25Mhz NitrOS9 system is really
> nice and doesn't require new software.  Everybody likes having the same
> thing only faster :)  Same for vga output, ps/2 keyboard, high speed
> drivewire, built in sound, etc.  [...]  A great aspect of the FPGA platform is that you don't have to
> commit to a single definition.  For example if CoCo3FPGA decided to
> implement a new MMU, nobody with a DE1 now has to buy new hardware, its
> just reprogram the FPGA to be the new MMU, and the party has started.
> That flexibility eliminates many difficult choices you would have to make
> if one were to design a coco 4 using traditional components, and I think
> that provides the best real hope we have of seeing a standard platform.

I agree very much about the price point, and the flexibility of the 
FPGA.  I think that some people may not quite grasp the fact that an 
FPGA-based computer can easily be re-configured, so the feature set 
needn't remain fixed.  In principle, one could define a Next-Gen CoCo 
that retains a high degree of legacy compatibility, OR a bunch of new 
features, depending on which ~firmware (not the right term, but I can't 
think of it ATM) file you load up in it.  So someone might buy the 
platform to play old CoCo games (if that cycle-accurate CPU comes along) 
or replace original hardware that has bit the dust or been eaten by the 
attic monsters, and then incidentally have a machine capable of running 
new software.

I agree with Bill L. that the potential market is likely far too small 
to attract the Shack, or make any other mass-production attempt 
feasible.  (Though I could just imagine a scenario in which RS could 
make a go of it:  Ship it with all the games they have the rights to 
(maybe none by now, especially since they sold their computer business 
off), put a Geek Port on it, and some easy software tools to poke at the 
outside world, bundle it with a cool book about the history of the CoCo, 
and put it on the shelf next to the Arduinos, Propellers, and BASIC 
Stamps.  If they could do that for $100, I think they'd sell quite a 
few, but I don't see them having the vision for that kind of 
market-segment-crossing, and there may well be Intellectual Property 
issues at this point.)

But I think there is a sweet spot in a compromise between a wholly DIY 
CoCo3FPGA and a polished consumer-like product.  I think the development 
boards are really pretty affordable, though they don't have all the I/O 
we would ideally like, and they don't have a nice case.  I think the 
opportunity is in selling interface boards that will plug into the 
development boards and add some necessary I/O. Then add a laser-cut 
acrylic mounting plate in two different styles:  vintage CoCo case 
adapter, and mATX mounting.  (Maybe the same plate could even work for 
both.)  Some internal hook-up cables and a DriveWire cable complete the 
kit, or you can just buy the I/O boards for a cheaper, more DIY option.  
A bonus enhancement would be a module that could store the bit files (or 
whatever they're called) for different Next-Gen CoCo configurations in a 
flash memory and be capable of loading them over the JTAG interface 
without having to hook up to an external PC.  With something like that, 
you could reconfigure between Classic and Super CoCo with ease.

I would also add that it's probably been a few years since anybody made 
any significant money selling software for the CoCo.  I'm not an Open 
Source fundamentalist, but really, the only reason anybody cares about 
the CoCo at this point is to have fun.  You can have fun playing old 
games and indulging in nostalgia, you can have fun hooking up gadgets to 
a computer that has a readily accessible bus interface, you can have fun 
collecting vintage goodies and retro-futuristic remakes, and you can 
have fun writing bare-metal programs and teaching an old dog new 
tricks.  So I'm not convinced that nobody is going to write software 
that takes advantage of new features of a Next-Gen CoCo, but I think 
it's more likely to be motivated by playing around with it and seeing 
what you can make it do than by looking for compensation or 
recognition.  NitrOS-9 has been a collaborative effort.  Why do people 
continue to work on it? It's a worthwhile project in its own right, 
whether there are a lot of users or not.  But the fact that it's out 
there, and continues to be improved benefits the whole community, and 
increases the chances that somebody like Bill Pierce will write more 
software for it.

JCE

> --
> Coco mailing list
> Coco at maltedmedia.com
> http://five.pairlist.net/mailman/listinfo/coco
>