[Coco] CoCo4! 50% done!

[Frank Swygert]

Others wrote:

I agree $200 is about the limit for such a device.  There is the same
chicken and egg problem that prevented any of the original "coco 4"
attempts from seeing success.  People don't write software for platforms
without users, and users don't buy platforms without software.  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".
---------
But are people prepared to pay that for a retro hobby computer when you can
buy a laptop with built in LCD and hard drive for less nowadays?

I think the magic price point is somewhere in the vicinity of $200. Anything
more than that and people will question what they really want it for when
there is no software that particularly uses the new features.
=======================

Herein lies the problem! For around $200 I can buy a brand new Atom powered machine (with Linux), or any number of dual core processor refurbished desktop computers (some with monitor -- 2.3GHz dual core, 2GB ram, 320GB HD, 17" LCD, and Win7 home premium -- $215+S&H @ Newegg). A new Android netbook can be had for $80-$160, or a full refurbed 14" LCD laptop (2GB RAM/120GB HD/2.4GHz dual core/Vista).

The problem with a standard PC is easy interfacing with the outside world, but there are many adapter boards readily available (see http://store.qkits.com/category.cfm/PC  for examples -- they have a parallel port relay control card for $36.50 that is interesting, and could be used with a CoCo with a parallel port card -- real parallel port, not sure a serial to parallel adapter would work). That adds $50 or more to the price though, and a learning curve to program them (most use C++ or Visual Basic). You could still come in around $200 with a used or refurbed computer though.

The CoCo was easy to program in BASIC and you could use the joystick ports and even the cassette relay for simple I/O projects, or build a relay card (or other card) for the cartridge connector. That took more to program, but it could be done. I'd think that easy interfacing and programming would be the main attraction for any new 8-bit computer, not application or gaming software -- that and general computer knowledge/learning. There are too many cheap video gaming consoles and surplus computers around for more "normal" computing. Most of us here don't even use the CoCo for "normal" work any more! I did as long as it was somewhat practical, but eventually had to stop because it just couldn't do what was needed any more. I read all my e-mail on it until 1999! When hi-res graphics became the norm even for e-mail I just had to stop, and the CoCo became a plaything. Not many will pay over $200 for a toy that will be severely limited compared to even a smart phone. Some may still want to learn how a computer really works, but not many. The only real viable market is an easy experimenter's machine that could even be used as the brains in a final product -- at least that's my opinion. Something that would compete with the Arduino boards but be a much easier to use finished product.

The abundance of cheap computers brings me back to suggesting that the features of the FPGA machine be incorporated into one of the CC3 emulators. At least then there would be a larger network of possible programmers and users for the extra features. I would still like to see an emulator set up with the OS in the background so that the computer booted up ready to go as a CoCo. Strip the main OS down so it boots faster, and only include a few native utilities that might be helpful. The hardware FPGA machine can dictate what the emulator can do, because if there is too much difference between the two there will be a rift in software development, and that wouldn't be good for either camp. The hardware device will be the limiting factor.