Noel Chiappa:
I'd lay good money that the vast majority of PDP-11's never ran Unix. And
UNIX might have happened on some other machine - it's not crucially tied to
the PDP-11 - in fact, the ease with which it could be used on other machines
was a huge part of its eventual success.
=======
I have to disagree in part: the PDP-11 is a big part of
what made UNIX so widespread, especially in university
departments, in the latter part of the 1970s.
That wasn't due so much to the PDP-11's technical details
as to its pricing. The PDP-11 was a big sales success
because it was such a powerful machine, with a price that
individual departments could afford. Without a platform
like that, I don't think UNIX would have spread nearly the
way it did, even before it began to appear in a significant
way on other architectures. Save for the VAX, which was
really a PDP-11 in a gorilla suit, that didn't really happen
until the early 1980s anyway, and I'm not convinced it
would have happened had UNIX not already spread so much
on the PDP-11.
It worked both ways, of course. I too suspect that a
majority (though I'm not so sure about `vast') of PDP-11s
never ran UNIX. But I also suspect that a vast majority
of those that did might not have been purchased without
UNIX as a magnet. I don't think those who weren't
around in the latter 1970s and early 1980s can appreciate
the ways in which UNIX captured many programmers and
sysadmins (the two were not so distinct back then!) as
no other competing system could. It felt enormously
more efficient and more pleasant to work on and with
UNIX than with any of the competition, whether from DEC
or elsewhere. At the very least, none of the other
system vendors had anything to match UNIX; and by the
same token, had UNIX not been there, other hardware
vendors' systems would have had better sales.
Sometime around 1981, the university department I worked
at, which already had a VAX-11/780 and a PDP-11/45 running
UNIX, wanted to get another system. Data General tried
very hard to convince us to buy their VAX-competitor.
I remember our visiting their local office to run some
FORTRAN benchmarks. The code needed some tweaking to
work under their OS, which DG claimed was better than
UNIX. Us UNIX people had trouble restraining our chuckles
as we watched the DG guys, who I truly believe were experts
in their own OS, taking 15 or 20 minutes to do things that
would have taken two or three with a few shell loops and
ed commands.
DG did not get the sale. We bought a second-hand VAX.
Blame UNIX.
Norman Wilson
Toronto ON
On 2016-01-25 02:11, John Cowan<cowan(a)mercury.ccil.org> wrote:
> Ronald Natalie scripsit:
>
>> >There were the Dec Professional 325 and 350 desktops which had the
>> >F-11 and the 380 had the J-11 (which should make a pretty snazzy little
>> >retro UNIX system)
> As well as the 310, which was not a desk*top* but a whole desk with a
> PDP/8-A built into it. The first regular job I ever had was with a
> company that sold these along with their accounting software.
The 310 was not called a Professional, though. It was the EDUsystem if I
remember right. There was also PDP-11 based EDUsystems, called 350. Not
the same as the desktop thingy...
Isn't it wonderful how DEC reused different designations sometimes.
There was also a DECstation 88, if I remember right, which was a PDP-8
based thing.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt(a)softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
> From: Clem Cole
> to help debug the kernel, we even put adb into the core resident port of
> V7 which was tricky - Noel I seem to remember we .. stole that from you
> guys at MIT
Well, I certainly don't remember doing such a thing - but I should point out
that the Unix 'community' at MIT was not at all in good touch with each
other. So perhaps someone else at MIT did it? Or perhaps it was done after
I left for Proteon?
Also, the group I was in - CSR - was, during my time with them, not well
connected to other Unix users outside MIT. So even the things we _did_ do seem
not to have made it to many (any?) people. I'm not sure why this was:
probably, since we were working exclusively on early TCP/IP stuff, we were
mostly in touch with other networking people.
The disconnect to the rest of MIT may have been because, in our case, the
technical community at Tech Square didn't have good contacts with the rest of
campus; we were kind of self-sufficient. The AI Lab people had some contacts
with the Plasma fusion group, and later the EE department on campus, but CSR
(and maybe all of LCS - I'm not sure, the groups in LCS were pretty isolated
from each other) didn't.
Also, Tech Sq was mostly about PDP-10's - initially running ITS, later TWENEX
- and only a couple of smaller groups ran Unix. The DSSR group had an 11/70,
and we were quite close to them, but AFAIK we were the only two groups in Tech
Sq running Unix. I don't think anyone else at MIT had a PDP-10, until the EE
department on campus got an TWENEX machine, so there wasn't really anyone on
campus for most of Tech Sq to interact with.
Noel
On 2016-01-25 02:11, jnc(a)mercury.lcs.mit.edu (Noel Chiappa) wrote:
>
> > The later M9301 (see disassembly of the contents here:
> >http://ana-3.lcs.mit.edu/~jnc/tech/pdp11/M9301-YA.mac
> > of one variant) didn't clear memory either
>
> OK, so_my_ memory is failing! That code does in fact test the memory.
>
> (Although, looking at it, I can't understand how it works; after writing the
> contents of R3 into the memory section it it asked to test, it complements the
> test value in R3, before comparing it with the memory it just wrote with R3,
> to make sure they are the same. Maybe there's an error in the dis-assembly?)
Read the code again, you missed it. :-)
The code first writes one value into memory (R3), then complements R3,
and for each location checks that the memory is *not* equal to R3, and
then writes R3 and checks that it now matches. Essentially checking that
it can be changed into a wanted value in time. And it does it two times.
First zeroing, and then writing ones, and then back to zeroes again, so
yes, the memory will be left containing all zeros, except for what
memory isn't tested.
> Anyway, it should have left the memory mostly containing all 0's.
Indeed.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt(a)softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
> From: Norman Wilson
> I have to disagree in part
You make a number of good points. A few comments:
> the PDP-11 is a big part of what made UNIX so widespread, especially in
> university departments
That last part was really a big factor, one not to be understated. That
penetration led to production of a whole generation of people who i) were
familiar with Unix, and ii) liked it, and were not about to put up with the
OS's being turned out by various vendors.
> I too suspect that a majority (though I'm not so sure about `vast') of
> PDP-11s never ran UNIX.
'Embedded systems'. The number of PDP-11's running timesharing was a small
share of the total number, I expect.
> I don't think those who weren't around in the latter 1970s and early
> 1980s can appreciate the ways in which UNIX captured many programmers
> ... as no other competing system could.
Very true. My jaw basically hit the floor when I first saw (ca. '75) what Unix
was like. People who didn't live through that transition can't _really_ grok
it, any more than my kids can really fully grok a world without mobile
phones. It wasn't as powerful as Multics, but I was completely blown away that
anyone could get that much capability into a PDP-11 OS.
Noel
>> It seems though that there should have been a PDP-11 based desktop
> Because DEC were a bunch of losers.
OK, that was kind of harsh. (Trying to send email too fast...) DEC had a lot
of really brilliant people, and they produced some awesome machines.
But when it comes to desktops, I think there is a certain amount of
bottom-line truth in that assessment: there was a huge market there
(obviously), and DEC should have been in a pretty good place to capture it,
but it completely failed to do so.
Why not? I put it down to corporate cultural intertia - ironically, the same
thing that allowed DEC to eat so much of IBM's lunch.
Just as IBM took way too long to understand that there was a very large
ecological niche for smaller machines _with customers who didn't want or need
the whole IBM hand-holding routine_, DEC never (or, at least, until way too
late to catch the wave) could change their mentality from producing really,
really well built computers for people who were all technical, to commodity
computers which needed to be made as absolutely cheaply as possible, and for
people who were non-technical.
The company as a whole just couldn't change its mindset that radically, that
quickly. (And a lot of the blame for that has to go to Ken Olsen, of course.
He just didn't grok how the world was changing.)
> There's some DEC history book which talks about DEC's multiple failures
> (on a variety of platforms, not just PDP-11 based ones) to get into the
> desktop market, if the title comes to me, I'll post it.
The best one on this topis, probably, is "Ultimate Entrepreneur", by Glenn
Rifkin and George Harrar, which gives a lot of detail on DEC's attempts to
build personal computers; also good is "DEC is Dead, Long Live DEC", by Edgar
Schein.
Noel
Clem Cole:
Also by the time DEC did try to build a workstation (after Masscomp,
Apollo, Sun et al had taken many of their engineers) it was too little too
late. The ship had sailed and they never recovered that market.
======
There was a window in the early 1990s when I think they could
have recovered. DEC had some pretty good MIPS-based workstations,
and Alpha was just coming out and was even better. Ultrix was
a good, solid system, and DEC OSF/1 (later Digital UNIX) was
getting there.
In 1994 or so, the group I worked in needed a new workgroup-sized
central server. Our existing stuff was mostly DECstations running
Ultrix (with a few SGI IRIX systems for specialized graphics).
We looked at the price and performance of various options:
everything SGI had was too pricey; Sun's was well behind in
performance (this was before UltraSPARC), and their OS was
primitive and required a lot of retrofitting to be usable
(this was also before Solaris 2 even came out, let alone
became stable; also before Sun grew up enough to ship a
decent X11 as part of the system).
So we bought a third-party system with an Alpha motherboard
in a PC-style case. In burn-in testing I discovered a bug in
the motherboard; the vendor were happy to fix it once they
could reproduce it in their lab (which took some doing, but
that was another story).
We were quite happy with that system, and would have bought
more had our entire department not been shut down in a
mostly-political fuss a couple of years later (that too is
another story).
DEC's desktop MIPS systems were quite good, and the Alpha
followons even better. Had the company's upper management
not by then lost all sense of how to run a company or to
sell anything ... but that was not to be.
Old-fart footnote: when our department shut down, I bought
some of our DECstations cheap from the university. I still
have them on a shelf downstairs; I've never done much with
them.
Norman Wilson
Toronto ON
