[Coco] Any news on the so called CoCo4 or Next CoCo

[Frank Swygert]

Mark, I think that between us we've pretty much summed up both sides of 
the fence! Your "counter points" are the most sensible I've heard yet. 
Writing something native to run on an Intel processor would indeed be 
the fastest, but there is enough power in a P4 laptop (refurbs around 
$200 - http://www.hcditrading.com/Refurbished-Laptops/c115076/) to run 
an OPTIMIZED emulator/OS at about the same speed as the DE-1 board or a 
similar design. The most important aspect of the soft side is that 
development cost is really just time and it could be done quickly. A 
card with a legacy connector could be easily added later if desired.

On the FPGA side -- would it be possible to modify one of those 
proprietary cards (in software/FPGA) and tack onto an existing order to 
reduce costs, assuming there is one that provides suitable video? An 
FPGA board with everything needed and a connector configured for a 
cartridge port adapter for around $200 would be nice. I would prefer a 
compact hardware solution as well, it's just not been truly feasible. 
Total cost with a $200 board would still be around $300 (minus monitor), 
and then you can't do anything else with it.

At least with the soft solution there are alternate uses for the 
hardware, even with an optimized emulator. A dual-boot system comes to 
mind -- then no need for dedicated hardware. A fast multi-core computer 
should be able to run a standard emulator under another OS as fast as 
the hardware solution though, so back to square one!

--------------------------------
Date: Fri, 22 Oct 2010 12:33:27 +1000
From: Mark McDougall <msmcdoug at iinet.net.au>


On 22/10/2010 10:57 AM, Frank Swygert wrote:

> > The FPGA solution ISN'T cheaper than an inexpensive generic P4 computer.
>
You'd be hard-pressed to argue that an FPGA solution is going to be more
cost effective - particularly if you're talking about a custom design done
specifically to emulate the Coco. Sure, the base hardware can be quite
cheap, but then you need to add all the peripherals, and the screen, to
bring it in line - functionally - with a software emulation on a cheap 
notebook.

My argument is specifically about the "experience". It's my personal 
opinion
that a software emulator running on a notebook - regardless of how
seamlessly it is wrapped under auto-boot scripts - just doesn't cut the
mustard. At the end of the day - to me - it seems ludicrous that I'd
"cripple" my PC by running a soft-core CPU to emulate a mythical machine.

[An aside: it makes much more sense to me to simply write a DECB compiler
for the PC, and/or perhaps a graphics emulation layer, and bypass the need
for a 6809 core at all, if what you ultimately want to develop for is a
"Coco4". Even then, this is not really anything that can't be done
relatively easily in any modern graphics library using any modern language
on the PC (or browser for that matter)]

But I get that it makes no difference to others. And that's fine.

So no point debating it with me or anyone else. I don't think any of us on
either side of the fence are going to be swayed at all.

A significant factor for me, personally, is the interfacing to legacy
hardware, and that includes the Coco keyboard, floppy drives, the original
joysticks, cartridges and even the composite monitors. At the same time,
some - or even all - of those can be augmented/replaced with modern
equivalents; PS/2 keyboard, SD/MMC card, GameCube/PSX joysticks, flash and
VGA/DVI monitors. I like the idea of being able to plug my DoD cartridge
into the "Coco4", but also have it running on a DVI monitor with Gamecube
joysticks. Sure, I don't *need* to use a DoD cartridge, but it's pretty 
darn
cool that I can! Ditto for using the original Coco keyboard.

-- 
Frank Swygert
Publisher, "American Motors Cars"
Magazine (AMC)
For all AMC enthusiasts
http://www.amc-mag.com
(free download available!)