> > I guess with Big Brother Intel the best processor designs
> > like the 6809 and the PDP-11 are the odd ones.
> Intel's had their share of odd processors as well, such
> as the 8035 and 8051. The 8089 as well, though it's not a
> general-purpose processor.
Who ever played around with low level programming on a
mainframe (/370ish) will recognize the structure Intel
had in mind when they designed the 8086/87/89 family.
On a 3/70 you usualy have at least one CPU (=86) and
one IOC (=89), while the FPU (=87) was a (never needed)
option. The high performance of classic mainframes was
based on this concept, and if you look at modern PCs,
somehow, 20 years later it creeps back into PCs with
programmable graphic processors and quite complicated
disk protocolls.
Having a seperate processor to handle all the low level
(and fast reaction) tasks of I/O handling also supports
well structured and simple oterating systems - and even
more, device drivers can be writen 100% independant from
the operating syswtem. If IBM had folowed Intel in using
the 8089 the problem of geting a device driver for unit
X and operating system Y wouldn't be the problem we have
today.
Intel was around 1976-1980 a quite innovative place,
the fact is widely ignored by common media. Just
remember the 432, or the 186, still one of the best
all in one CPUs around. Everybody just sees and
points out the PC CPUs - and a lot of them are
repeating the dogma that the 86 is a bad design.
I'm defenitly not an Intel fellow, just the other
way, I started out in the /370 world, and prefered
68xx / 65xx CPUs and I never liked the 8080/85 (or
Z80), but Intel-basching without diferentiation gets
me upset.
Gruss
H.
--
VCF Europa 4.0 am 03./04. Mai 2003 in Muenchen
http://www.vcfe.org/
Received on Fri Nov 22 2002 - 07:03:01 GMT