🚀 go-pugleaf

On Sun, 28 Nov 2010 20:24:12 -0800, "Robert Kent"
<rk64@NOSPAMyahoo.com> wrote:

|?
|
|"james"  wrote in message news:6ugle6ddirs8tat6gg87sjshnb6uhdrbr1@4ax.com...
|
|On Fri, 19 Nov 2010 13:25:46 -0800, "Robert Kent"
|<rk64@NOSPAMyahoo.com> wrote:
|
||?About a decade or so ago, there were a bunch of proposals to create a Coco
||4. At least one person had already started building boards for it.
||
||What happened?
||
||
|
|
|I really don't know about boards.
|
|I have proposed several different approaches over the past 6 yrs.
|Never felt there was much acceptance so nothing done more than a
|feasability study.
|
|james
|
|---
|
|I think any Coco 4 should be in the form of an emulator. There are several 
|open source Coco 1,2 & 3 emulators available. Start with them and add bits 
|and pieces of other emulated systems. The code is already written, all you 
|need to do is piece it together.
|
|Want a Coco 4 with 4096 colors? Cannibalize the code written for an Amiga 
|emulator. Want a better/faster 6809? The possibilities are endless.
|

Why stop at 4096 colors? 16 bit color is doable by just a few changes
in hardware and roll out a new GIME chip in a FPGA. The GIME chip has
16 pallete registers so set up an array of 16 bit color that is 256
wide. Therefore you have any 16 outof 256 16 bit colrs. Very doable in
a FPGA.

Speed is not all that it is cracked up to be. There comes a point in
hardware and software where data throughput becomes a bottle neck. In
hardware that comes about at 25MHz. The FPGA version of the COCO3 is
pretty much limited to about 25 MHz clock speed. In a FPGA you can do
things with the opcodes to execute faster than the discrete version of
the CPU. In that case you do create some timing issues for code that
was written years ago. 

The real issues and concerns of the COCO community is that the older
hardware and chips are repidly becoming scarce. It now has been 20 yrs
since Motorola announced the discontinuation of the 6809. Even thought
the mask sets have been turned over to Rochester Semiconducter, that
does not mean they are cheap and readily available. The last
production run of 6809 chips by Motorola was around week 46 of 1999.
The last run of GIME chips was somewhere in the middle of 1987. Other
discrete chips are no longer available. They are canablized from dead
cocos adn used to repair others. Seen many 5.25 360K floppy drives on
the market lately?  Disks are available but some are not realiable. 

The real concern is that teh existign hardware is fastly dieing and
there is not much to repalce it with. In time the COCO1/2/3 will ease
into obscurity and a footnote in history. 

just my thoughts

james