[Coco] [Color Computer] Is this likley to work ?

[RJRTTY at aol.com]

In a message dated 9/6/05 4:17:32 PM Eastern Daylight Time, 
dragon at aurigae.demon.co.uk writes:

> 
>  With the CPU and RAM databus isolated, it should be possible to build 
>  some decode logic to give the CPU an address space of greater than 64K,
>  you would still need to be able to page in the SAM controled RAM so that 
>  you could put data in it for the VDG to display, but withe the right 
>  logic you could page in just say an 8K page (as on the CoCo3).
>  
>  Is my logic sound, or am I just going slightly mad :)
>  
>  Both of course can be true.....
>  
>  Cheers.
>  
>  Phill.

    In 1986, I built a 1 meg expansion memory out ot thirty two
41256 memory chips and an MMU made out of discrete logic.
It was composed of
three boards layered on top of each other and it plugged into the
multipak interface via flat cable.   It was a fearsome looking beast and 
worked
the same as expansion memory cards for PC's around that time.
It was about the same size too.     The only use I put it to was as
a ram disk on my coco1.    I used it to do assemblies of a 3-D
rendering and rotation graphic program.     Assembly on a disk
drive would take
about 25 minutes while on the ram disk it took about three.  :)
I was really hard core coco in those days.
       I brought it to show off at the last fest if anybody remembers
tho I didn't demonstrate it.   Maybe at the next fest?

Roy