[Simh] pdp11 and unix

[Gregg Levine]

Hello!
Interesting.

I was only reporting what I remember as to the history of the whole
example we call UNIX.

And last year at the Vintage Computer Festival East, (Yes Dave W, the
same one where we crossed paths.), I saw a PDP-11 system having
finished dumping his program output to a TTY setup. I commented then
that the instructions shown resembled an 6502 one, I was also thinking
of the original 6800, but did not say that, and then it wasn't until I
walked away that I thought of a 68000, but only because I was inspired
by something I had read regarding the history of what was used in the
first Mac or its ancestor. And then continuously until much later when
reason caused Apple to switch to the PowerPC. Let's not discuss the
decision to switch to Intel.

And yes, I did recall that UNIX (or BSD) was both written in C and in
the host's version of Assembler. But I didn't put that down.

But I believe we are going around in circles here.
-----
Gregg C Levine gregg.drwho8 at gmail.com
"This signature fought the Time Wars, time and again."


On Fri, Feb 26, 2016 at 8:49 AM, Clem Cole <clemc at ccc.com> wrote:
>
> On Thu, Feb 25, 2016 at 10:26 PM, Gregg Levine <gregg.drwho8 at gmail.com>
> wrote:
>>
>> I've seen Assembler output from a PDP-11 someplace. It's always
>> reminded me of a frustrated 6502 microprocessor or a 6800 series one.
>> But only just.
>
>
> Interesting- the 68000 should remind you of the PDP-11.   What would become
> the 68K (remember it was a skunk works project and not an official one), was
> a definite reaction to the 6800/6809 not being good enough (single
> accumulator system; not general registers).   As someone that cut his teeth
> with the 6502 (and 6800) at the same time as I learned the 11, I never
> considered them similar.
>
> That said, the 68K guys were all PDP-11 programmers (in fact the CAD system
> as it were, was an 11/70 running a V6 flavor).    Les Crudele (Lead designer
> of the 68000) once told me that they wanted to use the PDP-11 instructions
> set, but KO had just put Cal Data out business for using the same
> instruction set on a the same (Uni)bus.   So you can definitely see the
> Unibus influence in their "experiment" but part of why it has A and D
> registers was to make sure it was different enough from the 11 that no one
> could claim a rip off.
>
> I asked Les why they did not try to license it from DEC, and he said they
> were afraid the project was going to shut down, so they kept a low profile.
> They just wanted to prove that they could make a chip with a 16 bit barrel
> shifter (which is the major real estate issue when you look at the 68K - is
> the big array in the center of the chip), and a real bus.    BTW: one of the
> other really good things that the 68K inherited from DEC that it's brethren
> at Moto and MOS Tech lacked was the idea of an unimplemented op code
> generating an exception.   Before that time, none of the micros bothered
> (good story about how Apple polluted the instruction set because it - but
> that's a different email).
>