25 May
2007
25 May
'07
5:45 a.m.
On 25 May 2007 at 5:11, Ensor wrote:
> True, but couldn't 6800 family peripheral
chips be made to work
with the 68k
fairly easily? If you're going to have to re-write
the code anyway, driving
different peripheral chips shouldn't be too difficult (unless you're using
Z80 peripherals with vectored interrupts, DMA etc).
Yes, but at the time, the range of 808x peripheral chips was far
broader than that available for 6800. I could never figure out if
Moto really intended the 6800 as a serioius stand-alone
microprocessor or not. They certainly didn't seem to take much
notice when the 6502 walked away with what should have been 6800
business.
IKWYM. I don't recall much about the Z8000 now,
but I did read the data
sheets for it somewhere around 1984 (1983?).
Cipher used the Z8002 extensively in their tape drives and Onyx+IMI
had a Z8000 Unix box, as did a few other vendors. Olivetti used it
in their M20 desktop computer, but it made precious few inroads into
the PC market otherwise. AMD was a second-source for it and Siemens
partnered with them to form AMC, whose purpose was to promote the
Z8000 with software and systems. That one fizzled after a mercifully
short period of boondoggling. (Hey Al K., I've got some AMC Z8000
product manuals; are you interested in them for Bitsavers?)
Quite, when I built my first PC in 1990, I ran the
MS-Dos version of
Wordstar until I finally gave in and switched to Win95 aroud the end of '96.
In fact, ISTR using it regularly on my ST in the late 80's using a PC
emulator.
You'd be surprised at how convenient the MS-DOS "put the function
code in CL and CALL 0005h" came in handy with translated code. That,
and the FCB, which, in important fields matched the CP/M FCB.
Actually, I remember reading an interview in an issue
of Byte some years
ago, where the leader of the PC design project was quoted as saying that his
original choice for the processor was the 68000.
It was mine for Durango's own business systems. We'd actually built
a wirewrapped prototype and were developing the basic software
(diagnositcs, IPL code, etc.) for it when Bill Davidow, who sat on
Durango's board, declared that it'd be a cold day in hell when we'd
use a non-Intel CPU.
As a result, we wasted precious time (months and months) sturggling
with buggy steppings of the 80186 and vague promises from both Intel
and Microsoft that 80286 Xenix would be ready Real Soon Now. So
Durango put out their 80186 CPU with rebadged Beehive terminal as a
PC "almost compatible". Customers didn't know what the heck to make
of it and deserted us in droves. I like to think that if we'd
stayed the course with a 68K design, we might have survived a bit
longer.
Although....if we'd had 68k/UNIX based boxes on
our desktops instead of
8086/DOS who's to say that there wouldn't have been a similar explosion of
software?
To me, one thing that saved the PC from becoming yet another niche
product was its video display--and the products developed to take
advantage of it, Windows being one of them. Heck, I still have the
source code somewher for Mewel, if anyone remembers that.
Cheers,
Chuck