[Coco] Coco ProtoType Board

[jdaggett at gate.net]

On 2 Sep 2006 at 12:11, Joel Ewy wrote:

> Mike Pepe wrote:
> > Hi James,
> >
> >> 2) some form of flexible address decoding included along with
> >> buffering. It is still unclear as to whether just the Coco 3 I/O
> >> range be flexible or also the range from $C000 to $FFDF also be
> >> included. 
> >
> > I think the PLD would be nice, but overkill.
> >
> If it's a 100-pin fine pitch SMT that only uses a few outputs and leaves
> the rest unconnected, has 150 registered macrocells that are permanently
> wasted, and takes the price right out of the impoverished CoCo hobbyist
> range, then I would concur with Mike that a few 74xx chips would be a
> better option.
> 
*****************

in a 100 pin TQFP, Xilinx has 72, 108 and 144 macro cell devices. That is 1600,  
2400, and 3200 gate equivalent devices. A XC9572 can do a simple 256 color 
VGA video adaptor and use less 70% of the available resources. 

Also to do all the data buss buffering and address decodeing, one will need about  
50 I/O pins plus the JTAG pins. 

> On the other hand, if the PLD can be re-programmed in situ, (I think at
> least some CPLDs can be??) then "overkill" here means "extra potential
> features".  Certainly you don't need more than a few TTL chips just to
> decode some addresses.  But if you can get more logic in the same space,
> and something approaching the same price, and if that logic can be
> appropriated by the user, then I would see it as worthwhile.  If the thing
> could be made to mimic an MPI, for instance, or could be programmed with
> internal registers or counters, then it would be a significant improvement
> over a dedicated address decoder circuit.  If the extra I/O lines were
> brought out to .1" headers, it would seem like useful potential, not
> overkill.
> 
******************

All CPLDs made today are in circuit programmable and are flash technology. 
They boast 20 year data retention. 

> A reasonable compromise, in the event that the PLD in question is not
> reprogrammable in circuit, would be to try to use up as much logic as
> possible on potentially useful functions, and make it available through
> jumpers or DIP switches, and MPI-like internal selection.
> 
> Of course the main purpose of such a board would be for hobbyists to
> attach their own gadgets.  But a simple expansion kit could turn it into
> an enhanced Multipak replacement for the less technically inclined.  I
> have a CoCo 3 in a mini-tower case and have used a ribbon- Y-cable to hook
> up a Disto SC-II/4-in-one and an RS-232 pak.  I'm not going to tear apart
> my one MPI to try to wedge the board in the mini-tower.  I don't think it
> would fit anyway.  But if I had an inexpensive project board that had MPI
> circuitry built-in, that would be another story.  I would also burn
> several of my favorite cartridge games on a 27512 and stick that in one of
> the "slots".  Wouldn't it be nice if the logic to switch banks in a larger
> EPROM would be built into such a device?  I'm confident I could add that
> myself.  I have the hardware and software (and I think, the knowledge) to
> program PALs and GALs.  But if the capability is already present in a
> board I could purchase for, say, <$50, I would rather save myself the time
> and effort.  And most CoCo hobbyists probably don't have a universal
> device programmer.
> 
************

I have looked into a revamped MPI and expanded capabilites. There are 
interesting potentials here. 

james