[Coco] A new CoCo? - was [Re: Coco Digest, Vol 12, Issue 31]

[Andrew]

Frank,

Part of me wants to say PC hardware isn't difficult to interface with, 
part of me wants to say it is. When ISA (and EISA) was the standard, it 
was fairly easy to interface hardware to the PC. With PCI, it becomes 
more difficult, simply because of the speed issues. USB isn't hard to 
interface with (there are many manufacturers of USB interface chips, and 
plenty of info on how to use them), nor is the parallel port (though 
this port is slowly dropping away, too).

You are right about the compact size for controller applications, but if 
this is what you want to do, then you are better off as the designer 
sticking with a real microcontroller for the task (since it would be 
much smaller and less power hungry). If you need monitoring or graphical 
output, interface the uC with a real computer via USB or a serial link.

The problem with building a "new coco" is the amount of money you would 
outlay, that you would never be able to recoup because the user base 
would be so small. Is it a computer, or a controller? Many designers 
would ask if it was a controller, why all the extra (video, sound) 
hardware? Why pay for that?

Sticking with a software only solution, based on GPL'd open source, 
would ensure the legacy of the CoCo through the generations, as the base 
system would get ported to new hardware platforms (hmm, it would be 
interesting to see a CoCo based on a Dreamcast, for instance - or on a 
cluster!). Plus, the software is free to download, develop on, and 
develop for. In a way, such a system could (in theory) bring easy to use 
computing back to the masses - rather than the mishmash a lot of people 
put up with. As their skills advance, the computer could automatically 
grow with them.

Emulation is only an option where backwards compatibilty is desired. 
Yes, there is a ton of great software out there from the ages (ie, 
Dungeons of Daggorath, Gates of Delerium, CoCo Max, etc) - but why not 
instead write new software to take advantage of the new hardware? I 
think some level of compatibility should be kept for normal BASIC code 
(that is, code that doesn't use ML routines and such). Emulate the rest 
- basing the new box off of Linux and X (or SVGAlib) - MESS would work 
perfectly for this task, if that is desired.

In fact, that would be one way to "rebuild" the CoCo for modern 
hardware: a Knoppix-like live-CD that boots into MESS - of course, you 
run into the ROM distribution issue (which is another strike against 
emulation - not everyone can legally have the ROMs).

Andrew Ayers