On Fri, 7 Nov 1997, Tim Shoppa wrote:
As regards
historical interest, well, it has an IOBYTE at address 3,
divided into 4 2-bit fields that define the console, punch, reader and
list device - long befroe CP/M.
I don't think Gary Kildall or anyone else DRI ever pretended that
the IOBYTE was their idea :-). My CP/M 1.3 Alteration Guide, in fact, says
The definition of the IOBYTE function
corresponds to the Intel standard as follows: a sinqle location in memory
(currently location 0003H) is maintained, called IOBYTE, which defines the
logical to physical device mapping wbich is in effect at a particular time.
Not surprising, especially since Gary Kildall tried to get Intel to
buy and market CP/M before he was told to go do it on his own...
Absolutely right Tim! As I remember Gary was working at Intel with a
MDS-800 (just like mine 8-) with ISIS II loaded (reason for the 4k offset
- ISIS owned that space) with PL/M (just like mine 8-) and ASM (just like
mine 8-). At least that is what the comments in my CP/M 1.4 source code
says 8-). And, gee, I can compile it and RUN it on my little MDS-800 just
fine!
I knew I would NEED to have these BIG HONKING BLUE BOXS one day 8-)
BC