[Coco] Coco4 thoughts

Fri Jan 2 17:27:20 EST 2009

I have to agree with Paul and Carlos. They are pretty much on the same track -- a nice learning/experimenters computer, cheap enough to be used not only to develop something, but to power it as well. 

Cost wise, the "CoCo4.exe" will probably be the most effective. PC hardware, especially a couple generations old, is cheap and plentiful, and already has everything built in. Heck, it would be nice to kick DOS, Windows, or Linux completely out of the picture -- the "CoCo4" would in effect be as much an OS as anything else. That would be ideal for DECB anyway, but might hinder running Nitros-9. Of course porting Nitros to the x-86 architecture has always been a possibility for that arena.

The cost factor is why Tandy went to the GIME. And that's our biggest hurdle now. Tandy skimped on it - made it "just enough" as far as speed and power, to save money. Can't blame them in a way -- the CoCo was supposed to be a cheap computer and the price of mainstream PCs were spiraling downward, cutting deeply into profit margins. I suppose we really should be grateful that there even was a CoCo3. 

Any video producing machine will have to have a reverse engineered GIME to be compatible. That has been a major hurdle. With the cost of FPGAs coming down, and I think someone has actually done the ground work, that's not so far fetched. Basing the machine on an existing evaluation board will be expensive though. The price needs to be kept under $300, though I'd really like to see one closer to the vicinity of $200, but I just don't think it's possible. Carlos, I'd love it if you proved me wrong!

So what would a new machine have to have as a MINIMUM -- not wants, but needs. That should be the starting point, then add the wants if/as possible. 

1. Minimum of CoCo3 software compatibility. I don't think not supporting the older, seldom used graphics modes of the CC1/2 will be a big loss, especially if losing them will save $$$ in the price tag. 

2. Capable of using modern peripherals. This looks to be a harder part, but I think it's easier than it sounds. There are USB floppy drives (though only 3.5"), and there are drive emulator boards for USB devices. That might be the easiest way to go. Instead of a hard drive a USB device would be good, or an SD card device. There are affordable 8GB USB thumb drives! Just supporting USB drive emulation might be easier and cheaper than supporting both, and USB is pretty universal. May need a special transfer program to transfer files to/from a PC if any special setup of the device is needed for the CoCo (program for the PC to read/write to the CoCo formatted USB drive).

3. Printers are a different story. There are few "smart" printers around anymore. The only one I know of are a few dot matrix printers still made for multi part forms still used by businesses. Epson still sells a 9 pin 300LX+II with serial, parallel, and USB ports, available for around $200. Would be nice to have ink jet support, but that would require some overhead. Someone wrote about an adapter that would translate ESC/2 codes into something a standard "dumb" inkjet (or dare I say laser?) printer could print, but I'm thinking a micro controller programmed with a printer driver would be the best solution. Of course then it would have to support a limited number of printers -- maybe the HP family? I'm assuming the drivers are all similar though. Probably not -- so any adapter or other solution would be limited to a few specific models. Still, it would probably be best to just leave the serial port (a real hardware port AND a bit-banger port would be nice!) and make a MC converter. A PC can always be used as a printer processor, I guess. I doubt many would be trying to use a new CoCo-like machine for word processing, though a printer for code listings would be nice. If it's as simple as storing the listing on a USB drive and sticking it in a PC (and possibly running a transfer program) to transfer and print, I could live with that. Or buy a $200 (or used) dot-matrix printer. 

4. Video. Must be CC3 compatible, and I'd really like to see video on board. If on board it must be compatible with off the shelf monitors. If the new machine's main purpose is for experimenters/learning basics, however, it could rely just on a TV for a monitor. I don't like the idea of having to rely on a PC, but in reality it could be a hardware board that could work alone, but require a PC for video and programming input. Most would really like to see higher resolution -- at least 640x480. 640x192 is the highest the CC3 will produce. More than 640x480 would probably take costs way up due to memory and hardware requirements. 

5. There are experimenter computers out there now, so the further we go in that direction instead of a stand-alone programmable learning/experimenting/game machine similar to the original CoCo, the less applicable it becomes. I mean it would just be of fleeting interest to CoCo people if it's only partially CoCo3 compatible and requires a PC to work. The only advantage such a machine would have is that the board could be programmed then disconnected from the PC to "do its thing". So do PIC controllers, BASIC Stamps, and several others though. If you're building a retro computer do it all the way! 

6. Then there's this little beastie: http://www.multilabs.net/Retro.html. It's very similar to a CoCo1. It has 64K (32K for ROM and graphics, but the graphics memory isn't accessible, and 32K program memory), low res graphics (CoCo1 had some better modes), etc. If nothing else, the interfacing devices offered by this company may help. Their ezVGA module will only produce 320x240 graphics though. They have PS/2 to serial keyboard and mouse adapters, and even a serial VGA adapter. This might be all most experimenters need, and the Retro is only $99. I'm thinking that a CoCo3-like computer similar to this is what people would really like to see. 

-----------------
Date: Wed, 31 Dec 2008 16:42:24 -0500
From: "Paul E. Jones" <paulej at arid.us>

My son has a Lego Mindstorm NXT robot.  While I am quite interested in it --
pretty cool with the Bluetooth interface and programmable microcontroller --
he is far less interested.  The reason is that if he creates something, he
can't share his creations.  He gets more enjoyment out of scripting on video
games sites and updating his own web site.

Date: Wed, 31 Dec 2008 21:07:06 -0600
From: Carlos Santiago <carlossantiago at austin.rr.com>

My view of the Coco family of computers has always been as a hobbyist.  

Anyone that is interested in a Coco as their primary machine, will  
demand many new features and applications that already exist in a PC  
or Mac. These users will have a hard time because they will try to  
mould the Coco into something that it is not.

As for a Coco4.exe Steve Bjork is already working on this along with a  
new version of Basic. These will satisfy the need for something that  
runs on the PC, but it will not allow you to develop any new Coco  
specific hardware.

The Coco has alway been a computer for those interested in writing  
programs and adding new and interesting hardware features.  A hobbyist  
computer.

My interest is to rekindle that spirit.

A new Coco4 will allow those who had great ideas about Coco software  
and hardware to implement them. 

I would like to start a list of people that are interested in the  
idea of a Coco4 and also those of you that are capable of  
contributing. This would be a project for the Coco community to  
provided those of you with the type of machine you dream about. Some  
of the feedback I received was that it would cost too much. Or that  
the design would be too complex. I have been a hardware engineer since  
1979 and have spent much time developing products that are low cost  
and feature rich.

On thing that is clear is that the Coco4 may materialize in the near  
future as software or hardware. If you would like a specific feature  
or capability, the only way to obtain it is by being involved. I mean  
in the real sense, not just to request a crazy feature to see if it  
can be done. The start of all this will be a development document that  
will define the boundaries of the project. Once this is complete and  
agreed upon, the design work can start.

-- 
Frank Swygert
Publisher, "American Motors Cars" 
Magazine (AMC)
For all AMC enthusiasts
http://farna.home.att.net/AMC.html
(free download available!)