[Coco] Today, I have seen the CoCo 4...

Mark McDougall msmcdoug at iinet.net.au
Tue Jun 8 21:28:18 EDT 2010


Frank Swygert wrote:

>  But either way none of the original CoCo hardware fits without further 
> adapter boards for connectors and hardware. So in the end it's more or 
> less six of one, half dozen of the other. 

Whilst your argument is true of off-the-shelf FPGA development kits, the 
really exciting potential is there to produce an FPGA-based board 
*specifically* for Coco emulation, that would incorporate both modern 
interfaces (like SD/MMC or USB storage adapters, wireless networking, VGA 
etc) and legacy Coco interfaces (cartridge, joystick, composite etc).

One recent example of such a product is the so-called One-Chip MSX (OCM) of 
which one production run was produced for the Japanese market.
<http://en.wikipedia.org/wiki/1chipMSX>

Then there's always the C-One (C64), Minimig (Amiga A500) and Suska (Atari 
ST) purpose-built hardware which included legacy interfaces to varying degrees.

I've always invisaged having purpose-built TRS-80 (Model I/III/4) and/or 
Coco FPGA-based hardware that can be used stand-alone or fit inside a real 
Tandy case; capable of emulating any flavour of TRS-80 and all manner of 
both modern and legacy peripherals. People are starting down this path to 
some degree now, perhaps not FPGA-based ATM but incorporating a lot of 
peripheral functionality in add-on boards that connect to original hardware.

Truth-be-told, a few of us had already started on the design of such a beast 
for the Coco. Unfortunately the project was abandoned late in the design 
stage due to rather sad circumstances.

The major sticking point I see now is the size (or lack thereof) of the 
TRS-80/Coco enthusiast market (which dwarfs in comparison to the Apple II, 
C64 or even Atari market) and the expense of doing a small run of what are 
moderately complex boards. If anyone doubts the cost/risk involved I need 
only point you to the C-One.

We did consider doing a "core emulation board" that would mate with a 
target-specific peripheral board, but that only blows out the design time 
considerably, and increases the unit costs considerably. We concluded that 
it would be quicker and cheaper to re-spin the same core design for 
different targets/form factors.

If I ever win lotto, you can count on a Coco4 board... ;)

Regards,

-- 
|              Mark McDougall                | "Electrical Engineers do it
|  <http://members.iinet.net.au/~msmcdoug>   |   with less resistance!"



More information about the Coco mailing list