24 May
2016
24 May
'16
2:10 p.m.
On 22 May 2016 at 04:52, Guy Sotomayor Jr <ggs at shiresoft.com> wrote:
Because the 808x was a 16-bit processor with 1MB
physical addressing. I
would argue that for the time 808x was brilliant in that most other 16-bit
micros only allowed for 64KB physical.
Whether 8088 was an "8 bit" or "16 bit" processor depends heavily on
how
you define those.
Or, you could phrase it, that the 8 bit processors at the time handled
64KiB of RAM. The 808x still could see only 64KiB at a time, but let you
place that 64kiB almost anywhere that you wanted in a total RAM space of
1MiB, and let you set 4 "preset" locations (CS, DS, SS, ES). There were
some instructions, such as MOV, that could sometimes operate with 2 of
those presets.
Thus, they expanded a 64KiB RAM processor to 1MiB, with minimal internal
changes.
As opposed to starting over from scratch, ala Motorola.
Starting over from scratch made it possible to build an arguably better
processor, whereas doing minimal changes reduced (and sometimes even
eliminated) the need to rewrite existing code.
Which is more important - theoretically better processor, or availability
of existing software? Not everybody would choose the same.
It was claimed that in porting Wordstar to PC, Micropro took longer to
change the manual than to get the software working.
Accordingly, software availability for the PC was more "immediate" than
for the Mac. Both had extensive software availability by the end of the
1980s, but how about during the first 6 months after release of the
machine?
Since "Nobody programs in assembly any more, nor ever will again"
(-Clancy/Harvey), that isn't as big a deal as it once was.
(OB_Picky: Due to the overlap of segment and offset, on machines that had
21 address bits, real mode actually had a maximum of 1114096 (10FFF0h)
bytes, instead of 1048576 (100000h). That was accessed by HIMEM.SYS at a
time when memory space was scarce enough that another 64K could make a
big difference.)