23 Apr
1998
23 Apr
'98
1:29 p.m.
[PDP11 risc or cisc]
Pete Turnbull:
I know that was directed at Allison, but I'd say
that key features of RISC
architectures include large numbers of general registers,
one-instruction-per-cycle, and hardware decode rather than microcode, not
just the obvious minimised instruction set.
The PDP-11 architecture has only 7 GP registers (since you can't really use
the PC for just anything) but that's good for the times, and they really
are interchangable, so I'd be willing to argue that it wins on that.
I'm glad somebody agrees with me on that! IMHO the concept of a GP
register is a RISC sort of thing. And, Allison, if you think RISC
should be register-rich, I claim the PDP11 was for its date, and
certainly was compared to micros of the 1970s.
It loses on the one-instruction-per-cycle, though.
Instructions take vastly
different amounts of time to execute, depending on what they are, and
they're all several cycles long. Just think about the FP instructions, or
Yeeeeesss...
I don't like the "one instruction per cycle" definition of RISC - for a
start,
what is a cycle? I prefer to think of RISC as an "every cycle is sacred"
philosophy - you don't waste cycles. I'd try to get _memory cycles_ as often
as the hardware permits them - on the 6502, for example, one per cycle (and it
almost manages it!), on 8080/Z80/PDP one every two or three cycles - but I
wouldn't make them all instruction fetches!
the Commercial Instruction Set. That's not the
most CISC thing you've ever
seen? :-) At a more mundane level, the additions of instructions like ASH
Despite having a 11/44, I have never seen a Commercial instruction Set :-)
is pretty CISC -- in fact the whole idea of extending
the instruction set by
altering or adding to microcode is the essence of CISC, and the antithesis
of a Reduced Instruction Set Computer.
Agreed. Later PDPs were more CISC, and this reached its maximum in the
Vax. But the basic architecture is IMHO a risc one - very simple and
very powerful.
And of course it loses on the microcode vs hardware
decode.
Except the early ones. Allison, are you sure it was the 11/05? I claim
it was the 11/15 (I have an 05). However I will concede that 05 may
have at one time been a name for an 11/20 variant.
Similar, but in many ways quite different. I just had
this argument (from
a somewhat different point of view) on another mailing list. The 68K is
much more like a PDP-11 than anything else, but it has a lot of clutter
added.
Fair enough.
That's my third of a tanner.
:-)
Philip.
PS I shall try and refrain from further comment on this issue - I don't
want to be the one who started a RISC versus CISC flame war!