[Coco] Re: [Color Computer] Re: Hello All

KnudsenMJ at aol.com KnudsenMJ at aol.com
Mon Jun 7 17:29:25 EDT 2004


In a message dated 6/7/04 4:15:15 PM Eastern Daylight Time, 
kevdig at hypersurf.com writes:

>   My two cents on this is that with the advent of Linux, I have very
>  little interest in OS9 anymore.

That's pretty much my situation here.  My UltiMusE program on Linux has long 
outstripped the capability to back-haul its newer features into OS-9/68K on 
the MM/1, let alone on the Coco.  In fact, ISTR that the MM/1 version got some 
features that wouldn't work back to the Coco, even with all the fancy virtual 
memory stuff going on.  (BTW, I can still give out the VM code in C, if 
anyone's interested).


> But I am still interested in the Coco for its hardware hacking fun 
capabilities.

For sure!  To say nothing of the fun of writing tight assembly code, which I 
did lots of before I went into OS-9.  Though still did some assembler routines 
there, too.

> But I am tossing around a few ideas for the Coco bus:
>   CPU paks (a CPU (68k, 80186, tms9900, 6502, etc), some ram and probably
>       a timer int)

Do you know about The Rocket, Chris Burke's 68008 plug-in CPU replacement 
that *almost* went into production?

>   updated spectrum analyzer Pak

You mean more accurate than the present one, which is plenty fun but not 
exactly scientific in its indications?

>   updated music synth Pak

Might make more sense to work on speech synthesis than music.  A MIDI 
interface lets you play almost anything.

>   static ram Pak (to allow a deuce to run at 1.79 MHz)

I did build one of these way back when, for the same reason -- to fool a 
"half-fast poked" Coco 1-2 into thinking RAM was ROM, so it would run double speed 
all the time, and still refresh the dynamic RAM.  Nowadays you'd need only 
one RAM chip to cover the whole space.  Maybe add some DIP switches to the 
address decoding logic to determine how many 8K blocks it steals.
--Mike K.



More information about the Coco mailing list