This family which was manufactured by Rockwell and others was very popular
because it was much cheaper than the similar 6800 series by Motorola,
just like the Z80 was much cheaper than the 8080/8085, which explains why
most of the home computers of around 1980 had either a 6502 or Z80.
Before the PC with it's 8088 became the standard, Sinclair tried to use
the 68008 as an alternative and Tandy/Radio Shack used the 6809 in it's
Coco (Color Computer).
The 6502 was like the 6800 derived from the very beautiful and orthogonal
PDP-11 instruction set, but the concept of orthogonality was even less
understood by the designers of the 6502 than it was by the Motorola folks
that made the 6800. (They later even went on to create the unorthogonality
monster known as the 68000...)
Well since the 6502 hadn't really a concept there never came a serious
successor although Apple used the 65816 processor (probably for compatibility
reasons) for a while to artificially prolong the life of the Apple ][
which was in the serious part of the Apple market already succeeded by
the Lisa/Mac.
I always disliked the 6502 instruction set with only it's program counter
being 16 bits and the rest of the registers (a, x, y, s) all being 8 bits
and it having very strange and uncomprehensible addressing modes which
were on top of it different for the x and y register... Well the first
serious compiler for the Apple ][ required the instalment of a Z80 board
in one of the IO* slots which would take over the control of the complete
or most of the computer... As far as I could judge most Apple users had
such a card after a while mocking the existence of the original 6502 in
the computer.
Not that the Z80 has well-balanced instruction set, but at least it has
registers with the proper size for it's (8-bit) generation. The 6800 also
has. Probably the best 8-bit processor architectures were the 6803/11
and the 6809 from an instruction set point of view, although the 6809
(and now the 6812) are better suited to the outdated mode of programming
in assembler and the 6811 is very well suited for compiler generated
code.
Oops but wasn't this supposed to become a 6502 fan page? ;-)
Here the link that prompted me to finally start this page:
atarihq.com/danb/6502page.htm
