TJ-2, a very early word-processor-like program for the PDP-1

71 views
Skip to first unread message

Daniel P. B. Smith

unread,
Jan 31, 1997, 3:00:00 AM1/31/97
to

I have posted a reproduction of the documentation for a VERY early
word-processor-like program, which ran on a PDP-1 with 4K words (9K bytes)
of RAM in 1963. I think the program was written by Pete Samson, although
no names appear in the documentation. For your nostalgic pleasure:

http://world.std.com/~dpbsmith/tj2.html

Comments welcome.

--
Daniel P. B. Smith
dpbs...@world.std.com

ds...@cameonet.cameo.com.tw

unread,
Jan 31, 1997, 3:00:00 AM1/31/97
to

Daniel P. B. Smith (dpbs...@world.std.com) wrote --

> I have posted a reproduction of the documentation for a VERY early
> word-processor-like program, which ran on a PDP-1 with 4K words (9K bytes)
> of RAM in 1963. I think the program was written by Pete Samson, although
> no names appear in the documentation. For your nostalgic pleasure:
>
> http://world.std.com/~dpbsmith/tj2.html
>
> Comments welcome.

Thank you, thank you, thank you. That is fascinating.

I'd like to comment on one of your notes:

TJ-2 does not resemble modern word processors with "on-screen
formatting" or "WYSIWYG editing." It somewhat resembles earlier word
processors like RUNOFF, TROFF or WordStar in which the editor and
formatter are separate....

WordStar has had on-screen formatting since very early days -- probably
since it first appeared in '78 or '79. To my knowledge, it has always
been as WYSIWYG as you'd want to get on a character-cell display (one
*can* go too far in that direction). In '81 or so all you needed for
editing and printing was WS.COM, a dynamic code overlay, and a message
overlay (there were also dynamic mail-merge and spelling overlays). It
was an integrated editor and formatter, and it did letter-quality
printing in proportionally spaced fonts on "incremental" printers (in
those days, this generally meant daisywheels).

I've heard others compare WordStar to nroff, troff, et al, and I can't
for the life of me figure out where this notion comes from.

BTW, WordStar version 7.0 came out in '92 and is still sold. The
WordStar Users' Group is 16 or 17 years old and still going strong, and
there is a *very* active Bitnet-style mailing list.

A diamond is forever.

Dan Strychalski
ds...@cameonet.cameo.com.tw

Bob Manners

unread,
Jan 31, 1997, 3:00:00 AM1/31/97
to

> I have posted a reproduction of the documentation for a VERY early
> word-processor-like program, which ran on a PDP-1 with 4K words (9K bytes)
> of RAM in 1963. I think the program was written by Pete Samson, although
> no names appear in the documentation.

4Kwords == 9Kb. I take it that the PDP-1 had an 18 bit word length then?

--
------------------------------------------------------------------------------
Bob Manners 24 Temple Street, Oxford, OX4 1JS, Tel/FAX: 01865 245819
My REAL address is: r...@swift.eng.ox.ac.uk
------------------------------------------------------------------------------

Stephen Westin

unread,
Jan 31, 1997, 3:00:00 AM1/31/97
to

In article <RJM.97Ja...@europa.ox.ac.uk> r...@europa.ox.ac.uk (Bob Manners) writes:

> > I have posted a reproduction of the documentation for a VERY early
> > word-processor-like program, which ran on a PDP-1 with 4K words (9K bytes)
> > of RAM in 1963. I think the program was written by Pete Samson, although
> > no names appear in the documentation.
>
> 4Kwords == 9Kb. I take it that the PDP-1 had an 18 bit word length then?

You betcha. My impression of the PDP genealogy:

PDP 1,4,7,9,15: 18-bit machines
PDP 5,8,12: 12-bit
PDP 6,10,20: 36-bit
PDP 11: 16-bit
PDP 2: Never built
PDP 3: Designed, never built by DEC, but rumor has it a customer
bought DEC modules and assembled one. Don't know word length.
PDP 13: Never existed
PDP 14: Industrial automation controllers. Don't know word length
PDP 16: Not a machine, but a system of register-transfer modules
(memory, ALU, etc) from which a customer could build various
computer-like devices. I used them in a class at Michigan; I
believe DEC donated lots of 'em after they died in the market.
They came out just in time to be obsoleted by the increasing
level of integrated circuits of the time.

-Stephen H. Westin
swe...@ford.com
The information and opinions in this message are mine, not Ford's.
--
-Stephen H. Westin
swe...@ford.com
The information and opinions in this message are mine, not Ford's.

Douglas W. Jones,201H MLH,3193350740,3193382879

unread,
Jan 31, 1997, 3:00:00 AM1/31/97
to

From article <RJM.97Ja...@europa.ox.ac.uk>,
by r...@europa.ox.ac.uk (Bob Manners):

> 4Kwords == 9Kb. I take it that the PDP-1 had an 18 bit word length then?

Exactly. The PDP-1, PDP-4, PDP-7, PDP-9 and PDP-15 were related in that
regard. The later 18 bit machines were not compatable with the PDP-1,
though.

DEC Press has come out with a wonderful book, Digital At Work, that includes
photos and overviews of the specs for most of the PDP machines, as well as
predecessors like the MIT TX0.

You can find a short table of these attributes in the first part of the
alt.sys.pdp8 FAQ.

Doug Jones
jo...@cs.uiowa.edu

A.R. Duell

unread,
Jan 31, 1997, 3:00:00 AM1/31/97
to

jo...@pyrite.cs.uiowa.edu (Douglas W. Jones,201H MLH,3193350740,3193382879) writes:
>DEC Press has come out with a wonderful book, Digital At Work, that includes
>photos and overviews of the specs for most of the PDP machines, as well as
>predecessors like the MIT TX0.

Yes, that book is rather fun but there are a number of errors in it that
may be confusing at first. I seem to recall that one of the pictures of a
'PDP11 System' is actually a VAX 11/730 (or maybe the other way round),
etc. My favourite error is the fact that a very nice picture of a PDP8 of
some flavour has been printed left-right reversed, or at least, I think it
has (unless there was an ASR-33 like teletype with the punch on the right,
and a single DECtape drive that filled the _right_ hand half of the rack
panel).

None-the-less, there's a lot of good stuff in it, and it's certainly worth
buying if you are interested in the old DEC machines.

> Doug Jones

--
-tony
ar...@eng.cam.ac.uk
The gates in my computer are AND,OR and NOT, not Bill

Douglas W. Jones,201H MLH,3193350740,3193382879

unread,
Jan 31, 1997, 3:00:00 AM1/31/97
to

From article <5ct39s$j...@lyra.csx.cam.ac.uk>,
by ar...@eng.cam.ac.uk (A.R. Duell):

>>DEC Press has come out with a wonderful book, Digital At Work, ...


>
> Yes, that book is rather fun but there are a number of errors in it that
> may be confusing at first. I seem to recall that one of the pictures of a
> 'PDP11 System' is actually a VAX 11/730 (or maybe the other way round),
> etc. My favourite error is the fact that a very nice picture of a PDP8 of

> some flavour has been printed left-right reversed, ...

I hadn't noticed that one! Page 50, lower left. Not only is the tape
reader-punch on the ASR 33 on the wrong side, and the TU 55 drives with
their tape reels on the wrong side, but you can clearly read the unit
number on the upper drive -- a backwards 4.

I get the feeling that layout professionals really like people to face
to the right in profile views, and that that photo is as likely to have
been left-right reversed when it was originally run as an advertising
photo back in the 1970's as it is to have been reversed in for the book.
I've seen other left-right reversals in DEC ads from that era.

Doug Jones
jo...@cs.uiowa.edu

Johnny Billquist

unread,
Jan 31, 1997, 3:00:00 AM1/31/97
to

>In article <RJM.97Ja...@europa.ox.ac.uk> r...@europa.ox.ac.uk (Bob Manners) writes:
>
>> > I have posted a reproduction of the documentation for a VERY early
>> > word-processor-like program, which ran on a PDP-1 with 4K words (9K bytes)
>> > of RAM in 1963. I think the program was written by Pete Samson, although
>> > no names appear in the documentation.
>>

>> 4Kwords == 9Kb. I take it that the PDP-1 had an 18 bit word length then?
>

>You betcha. My impression of the PDP genealogy:
>
>PDP 1,4,7,9,15: 18-bit machines
>PDP 5,8,12: 12-bit
>PDP 6,10,20: 36-bit

*Sigh* There never was any such beast as the PDP-20.

>PDP 11: 16-bit
>PDP 2: Never built

True, but it was reserved for a 24-bitter according to MRC.

>PDP 3: Designed, never built by DEC, but rumor has it a customer
> bought DEC modules and assembled one. Don't know word length.

36-bits. One was built in 1960, according to MRC.

>PDP 13: Never existed
>PDP 14: Industrial automation controllers. Don't know word length

12-bit, but with a 1-bit accumulator.

>PDP 16: Not a machine, but a system of register-transfer modules
> (memory, ALU, etc) from which a customer could build various
> computer-like devices. I used them in a class at Michigan; I
> believe DEC donated lots of 'em after they died in the market.
> They came out just in time to be obsoleted by the increasing
> level of integrated circuits of the time.

And it was a 16-bitter according to MRC.

Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: b...@update.uu.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol

Roger Ivie (a known boogerhead)

unread,
Jan 31, 1997, 3:00:00 AM1/31/97
to

In article <5csjas$4...@reader.seed.net.tw>, ds...@cameonet.cameo.com.tw wrote:
>I've heard others compare WordStar to nroff, troff, et al, and I can't
>for the life of me figure out where this notion comes from.

From the dot commands, near as I can figure.

Roger "using WordStar for 16 years and loving it" Ivie
iv...@cc.usu.edu


Stephen Westin

unread,
Jan 31, 1997, 3:00:00 AM1/31/97
to

In article <5ctbdp$cjd$1...@Zeke.Update.UU.SE> b...@Zeke.Update.UU.SE (Johnny Billquist) writes:

<snip>

> >PDP 6,10,20: 36-bit
>
> *Sigh* There never was any such beast as the PDP-20.

OK, DECsystem-20. Or DECSYSTEM-20. Or decSystem20. Or whatever silly
combination of capitalization and punctuation the marketing people
came up with.

<snip>

> >PDP 16: Not a machine, but a system of register-transfer modules
> > (memory, ALU, etc) from which a customer could build various
> > computer-like devices. I used them in a class at Michigan; I
> > believe DEC donated lots of 'em after they died in the market.
> > They came out just in time to be obsoleted by the increasing
> > level of integrated circuits of the time.

> And it was a 16-bitter according to MRC.

I seem to recall that the ALU came in 4-bit chunks, so you could roll
your own. It was about two decades ago, so I might well be wrong. I
tried to implement an IBM-360 instruction set with the stuff, but
don't remember if I had a 32-bit ALU or not. Pretty sure I built
32-bit registers.

Chris Ward

unread,
Feb 1, 1997, 3:00:00 AM2/1/97
to

> >PDP 5,8,12: 12-bit
Along with the knowledge that the Data General Nova was decended from the
PDP-8
(Edson DeCastro worked on both) two other machines that I worked with were
very similar to
the PDP-8. They were both made by Nicolet instrument Corp, the 1080 and
the 1180. I do not
remember much about the 1080 except that I think it was 18 bits (it ran
FFT's for an NMR) but I did
program the 1180 which was a 20 bit machine. Mounted in a Nicolet 1099 (I
think) FFT-IR machine
I think it may have been designed around the resoultion of the digitizers
that they had available.
I was able to program it like a PDP-8 (it had the same kind of Page zero
addressing for
indirect references and stuff, and was a single AC machine, and it may have
had relative addressing)
and it had a disk operating system called Dexter-2. It might have been an
OEM machine, but I do
not remember anything that would indicate a different origination of the
beast.

Anyone else see one of these?

John Everett

unread,
Feb 1, 1997, 3:00:00 AM2/1/97
to

In article <01bc0fe8$97d6d740$0b4284a9@wardch>, c...@mail.idt.com says...

>
>Along with the knowledge that the Data General Nova was decended from the
>PDP-8
>(Edson DeCastro worked on both)

Ain't revisionist history wonderful! Actually the Nova was descended (as
opposed to decended) from the PDP-X (or X-project) at DEC. This was DEC's
first attempt at a 16 bit machine, rejected by the operations committee.
Eventually a new design emerged as the PDP-11.

--
jeve...@wwa.com (John V. Everett) http://www.wwa.com/~jeverett


Will Rose

unread,
Feb 1, 1997, 3:00:00 AM2/1/97
to

ds...@cameonet.cameo.com.tw wrote:
[...]
: WordStar has had on-screen formatting since very early days -- probably

: since it first appeared in '78 or '79. To my knowledge, it has always
: been as WYSIWYG as you'd want to get on a character-cell display (one
: *can* go too far in that direction). In '81 or so all you needed for
: editing and printing was WS.COM, a dynamic code overlay, and a message
: overlay (there were also dynamic mail-merge and spelling overlays). It
: was an integrated editor and formatter, and it did letter-quality
: printing in proportionally spaced fonts on "incremental" printers (in
: those days, this generally meant daisywheels).

Not sure you needed the MSG overlay - if it didn't exist (to save space)
I seem to recall WS gave a standard "urk" message of some sort. 'Course,
you had then to work out what the message might have been.

Will
c...@crash.cts.com

ds...@cameonet.cameo.com.tw

unread,
Feb 2, 1997, 3:00:00 AM2/2/97
to

Will Rose (c...@cts.com) wrote --

>ds...@cameonet.cameo.com.tw wrote:
>[...]
>: WordStar has had on-screen formatting since very early days -- probably
>: since it first appeared in '78 or '79. To my knowledge, it has always
>: been as WYSIWYG as you'd want to get on a character-cell display (one
>: *can* go too far in that direction). In '81 or so all you needed for
>: editing and printing was WS.COM, a dynamic code overlay, and a message

>: overlay.... [snipped by DS]


>
> Not sure you needed the MSG overlay - if it didn't exist (to save space)
> I seem to recall WS gave a standard "urk" message of some sort. 'Course,
> you had then to work out what the message might have been.

Yup, in 3.x at least, you could do without the message overlay
in a pinch. And working on an Apple II with one 128K disk drive,
you could get yourself into a pinch pretty easily. I have vague
memories of zapping WSMSGS.OVR once or twice... *very* vague....

Dan Strychalski -+=%#@#%=+- ds...@cameonet.cameo.com.tw

Chris Ward

unread,
Feb 2, 1997, 3:00:00 AM2/2/97
to

> >Along with the knowledge that the Data General Nova was decended from
the
> >PDP-8
> >(Edson DeCastro worked on both)
>
> Ain't revisionist history wonderful! Actually the Nova was descended (as
> opposed to decended) from the PDP-X (or X-project) at DEC. This was DEC's

> first attempt at a 16 bit machine, rejected by the operations committee.
> Eventually a new design emerged as the PDP-11.
>

Does not mean that the PDP-X machine could not have been a machine
in the middle. I am not trying to revise history, but instead find out
what
it was. From my readings and listenings Mr. DeCastro was heavily involved
in the PDP-5/8 world and as I recall the project being cancelled did result
in his
leaving DEC and founding Data General. And there were even rumors of a
court
decision over DEC claiming ownership of the design, and (for instance the
bit
arrangement in the instruction decoding) where the Supreme Court said that
making these rearrangements were sufficent in this case to create a new and
novel machine, but that in future cases it would not be, because the
technique
had become general knowledge.

Chris Ward

shen...@escape.widomaker.com

unread,
Feb 2, 1997, 3:00:00 AM2/2/97
to

ds...@cameonet.cameo.com.tw writes:

> TJ-2 does not resemble modern word processors with "on-screen
> formatting" or "WYSIWYG editing." It somewhat resembles earlier word
> processors like RUNOFF, TROFF or WordStar in which the editor and
> formatter are separate....

>WordStar has had on-screen formatting since very early days -- probably


>since it first appeared in '78 or '79. To my knowledge, it has always

No, it didn't. It was at best able to format given a document width
and show things like page breaks. It was never anywhere near WYSIWYG.
That is what the poster meant by on screen formatting. The display
of Wordstar was more to show you where you were in a document and some
basic characteristics, not what you would see on paper. Even my worst
ever printer looked better than that.

>I've heard others compare WordStar to nroff, troff, et al, and I can't
>for the life of me figure out where this notion comes from.

Then you must never have used the dot notation of Wordstar. Most of its
power was in the dot commands. nroff uses dot notation for its formatting
command set as well, and that is why Wordstar is considered similar.
The 8-bit Atari program TextPro was a lot like that as was the IBM PC
program PC-Write.

>BTW, WordStar version 7.0 came out in '92 and is still sold. The
>WordStar Users' Group is 16 or 17 years old and still going strong, and
>there is a *very* active Bitnet-style mailing list.

--
csh - shen...@widomaker.com - http://www.widomaker.com - Linux 2.0.25
----------------------------------------------------------------------
"The Heisenberg Uncertainly Principle has completely defeated the
Einsteinian Physisist" -- Huh? What? Who said that?

ds...@cameonet.cameo.com.tw

unread,
Feb 3, 1997, 3:00:00 AM2/3/97
to

shen...@escape.widomaker.com wrote --

> ds...@cameonet.cameo.com.tw writes:
>
>> WordStar has had on-screen formatting since very early days -- probably
>> since it first appeared in '78 or '79. To my knowledge, it has always
>
> No, it didn't. It was at best able to format given a document width
> and show things like page breaks. It was never anywhere near WYSIWYG.
> That is what the poster meant by on screen formatting. The display
> of Wordstar was more to show you where you were in a document and some
> basic characteristics, not what you would see on paper. Even my worst
> ever printer looked better than that.

I was careful to write "on-screen formatting" and not "WYSIWYG", because
there is now a big difference. Before the Mac came out, however, what the
WordStar editing screen provided was called WYSIWYG, and it is still
called on-screen formatting. Margins, indents, outdents, page breaks,
exactly what text will print on each line -- all the formatting that
*can* be shown on a character-cell display is shown. Dot commands and
non-printing control codes can be displayed or hidden with two
keystrokes. There's a big difference between that and what you get using,
say, vi and nroff or troff. I call it on-screen formatting.

>> I've heard others compare WordStar to nroff, troff, et al, and I can't
>> for the life of me figure out where this notion comes from.
>
> Then you must never have used the dot notation of Wordstar. Most of its
> power was in the dot commands. nroff uses dot notation for its formatting
> command set as well, and that is why Wordstar is considered similar.
> The 8-bit Atari program TextPro was a lot like that as was the IBM PC
> program PC-Write.

I used dot commands to good effect for many years*, and I know that nroff
uses similar commands. I don't see how that makes WordStar the same kind
of program as nroff.

Dan Strychalski ds...@cameonet.cameo.com.tw

*Actually I changed the "dot" command flag character to 29h [)].

ds...@cameonet.cameo.com.tw

unread,
Feb 3, 1997, 3:00:00 AM2/3/97
to

I wrote --

> I've heard others compare WordStar to nroff, troff, et al, and I can't
> for the life of me figure out where this notion comes from.

...and Liam Quin (l...@sq.com) replied (privately, but with cc's to the
newsgroups, so there's no telling if my [expletive deleted] news server
will ever carry them) --

> You always used to be able to put things like
> .PP
> in a Wordstar file (or just type them in) to start a new paragraph, etc.

Things *like* .pp, yes (.pp is not a WS dot command), but not to start a
new paragraph. Paragraphs are delimited by 0Dh 0Ah (cr+lf). Within a
paragraph, WS uses 8Dh 0Ah, that is, it sets the high bit on the
carriage-return code and treats the sequence as a soft line break.

> These commands are similar to nroff commands.
> You don't need to use them, and for all I know they aren't even documented
> any more... They were there in the CP/M versions of Wordstar at least.

You do need them for many things. They are documented: the 1992 DOS WS 7
manual's command index lists about 70 of them. In WS 7, you can type and
edit them manually or let the program enter them automatically in
response to your input in Macintosh-like dialog boxes. WS for W-----s
doesn't use them at all, though.

Dan Strychalski ds...@cameonet.cameo.com.tw

JMFBAH

unread,
Feb 3, 1997, 3:00:00 AM2/3/97
to

You've been talking about a program called nroff. Since I was the first
group leader of the RUNOFF group at DEC, would you mind enlightening me?
We had a NRUNOF (sp?) for a while when the commands were being modified so
that semi-colons could be a terminator character of a formatting command
and the commands were shortened to 1-2 characters.

/BAH


Rick Hawkins

unread,
Feb 3, 1997, 3:00:00 AM2/3/97
to

In article <shendrix.854927050@escape>, <shen...@escape.widomaker.com> wrote:
>ds...@cameonet.cameo.com.tw writes:

>No, it didn't. It was at best able to format given a document width
>and show things like page breaks. It was never anywhere near WYSIWYG.

but it beat the daylights out of word perfect :)

>>I've heard others compare WordStar to nroff, troff, et al, and I can't
>>for the life of me figure out where this notion comes from.

>Then you must never have used the dot notation of Wordstar. Most of its


>power was in the dot commands. nroff uses dot notation for its formatting
>command set as well, and that is why Wordstar is considered similar.

But you had to be careful in WS. .. was the comment command. It's been
a while, but in 82 as an undergraduate, a pair of ellipses (for omitted
text) deleted two lines from one of my English papers: after
reformating, both landed on the first lines . . .

I finally seem to have broken myself fo word star commands. Or at
least, I've remapped all of the ones i still use to word commands :)

>>BTW, WordStar version 7.0 came out in '92 and is still sold. The
>>WordStar Users' Group is 16 or 17 years old and still going strong, and
>>there is a *very* active Bitnet-style mailing list.

>"The Heisenberg Uncertainly Principle has completely defeated the


>Einsteinian Physisist" -- Huh? What? Who said that?

But it only exactly beats the not quite certan Einsteinian :)
--
R E HAWKINS
rhaw...@iastate.edu

These opinions will not be those of ISU until they pay my retainer.

Joe Morris

unread,
Feb 3, 1997, 3:00:00 AM2/3/97
to

ds...@cameonet.cameo.com.tw writes:

>Daniel P. B. Smith (dpbs...@world.std.com) wrote --

>> I have posted a reproduction of the documentation for a VERY early


>> word-processor-like program, which ran on a PDP-1 with 4K words (9K bytes)
>> of RAM in 1963. I think the program was written by Pete Samson, although
>> no names appear in the documentation.

My recollection agrees with yours; similarly, my copy of the documentation
(Memo PDP-9-1, dated 9 May 1963) also has no signature block (mildly curious,
since many of the memos of that period have both an author's name and
an approval (often by J. B. Davis) on the last page.


> TJ-2 does not resemble modern word processors with "on-screen
> formatting" or "WYSIWYG editing." It somewhat resembles earlier word
> processors like RUNOFF, TROFF or WordStar in which the editor and
> formatter are separate....

It did, however, use the CRT to display words that required hyphenation.
The user pointed to the places where a word could be hyphenated, then
hit the "save" or "don't save" spot to control whether the hyphenation
information was to be added to the in-core directory.

The words that requried operator intervention were displayed long before
the output paper tape punch reached their location; with a reasonably
prompt reaction to their appearance the paper tape output would be
generated at full punch speed.

Joe Morris / MITRE

Max ben-Aaron

unread,
Feb 3, 1997, 3:00:00 AM2/3/97
to

High their:

In the late 60's & early 70's I worked for a company (Medidata, later
Searle Medidata) which started life as a not-for-profit spin-off from
Lincoln Lab. (as I have heard), called American Science Institute. The
chief engineer, Ed Rawson was a friend of Dec's Olsen and he managed to
get hold of the modules used for the prototype PDP-2 which never
reached the market. ASI used them to build their own machine
(designed, I believe, by Chuck Corderman) which they called "Casino"
and was sometimes jocularly referred to as a PDP-2 1/2. Casino was
noteworthy for having, very early in trhe game, graphics capabilities.
It also had some special terminals which had labels that cannot be
repeated on this (family) newsgroup.

Mb-A

=================================================================

In <SWESTIN.97...@dsg145.nad.ford.com>


swe...@dsg145.nad.ford.com (Stephen Westin ) writes:
>
>In article <RJM.97Ja...@europa.ox.ac.uk> r...@europa.ox.ac.uk
(Bob Manners) writes:
>

>> > I have posted a reproduction of the documentation for a VERY
early
>> > word-processor-like program, which ran on a PDP-1 with 4K words
(9K bytes)
>> > of RAM in 1963. I think the program was written by Pete Samson,
although
>> > no names appear in the documentation.
>>

>> 4Kwords == 9Kb. I take it that the PDP-1 had an 18 bit word length
then?
>
>You betcha. My impression of the PDP genealogy:
>
>PDP 1,4,7,9,15: 18-bit machines
>PDP 5,8,12: 12-bit
>PDP 6,10,20: 36-bit

>PDP 11: 16-bit
>PDP 2: Never built

>PDP 3: Designed, never built by DEC, but rumor has it a
customer
> bought DEC modules and assembled one. Don't know word
length.

>PDP 13: Never existed
>PDP 14: Industrial automation controllers. Don't know word
length

>PDP 16: Not a machine, but a system of register-transfer
modules
> (memory, ALU, etc) from which a customer could build
various
> computer-like devices. I used them in a class at
Michigan; I
> believe DEC donated lots of 'em after they died in the
market.
> They came out just in time to be obsoleted by the
increasing
> level of integrated circuits of the time.
>

Max ben-Aaron

unread,
Feb 3, 1997, 3:00:00 AM2/3/97
to

In <5ctbdp$cjd$1...@Zeke.Update.UU.SE> b...@Zeke.Update.UU.SE (Johnny

Billquist) writes:
>
>In <SWESTIN.97...@dsg145.nad.ford.com>
swe...@dsg145.nad.ford.com (Stephen Westin ) writes:
>
>>In article <RJM.97Ja...@europa.ox.ac.uk> r...@europa.ox.ac.uk
(Bob Manners) writes:
>>
>>> > I have posted a reproduction of the documentation for a VERY
early
>>> > word-processor-like program, which ran on a PDP-1 with 4K words
(9K bytes)
>>> > of RAM in 1963. I think the program was written by Pete
Samson, although
>>> > no names appear in the documentation.
>>>
>>> 4Kwords == 9Kb. I take it that the PDP-1 had an 18 bit word length
then?
>>
>>You betcha. My impression of the PDP genealogy:
>>
>>PDP 1,4,7,9,15: 18-bit machines
>>PDP 5,8,12: 12-bit
>>PDP 6,10,20: 36-bit
>
>*Sigh* There never was any such beast as the PDP-20.
>
>>PDP 11: 16-bit
>>PDP 2: Never built
>
>True, but it was reserved for a 24-bitter according to MRC.
>
>>PDP 3: Designed, never built by DEC, but rumor has it a
customer
>> bought DEC modules and assembled one. Don't know word
length.
>
>36-bits. One was built in 1960, according to MRC.
>
>>PDP 13: Never existed
>>PDP 14: Industrial automation controllers. Don't know word
length
>
>12-bit, but with a 1-bit accumulator.
>
>>PDP 16: Not a machine, but a system of register-transfer
modules
>> (memory, ALU, etc) from which a customer could build
various
>> computer-like devices. I used them in a class at
Michigan; I
>> believe DEC donated lots of 'em after they died in
the market.
>> They came out just in time to be obsoleted by the
increasing
>> level of integrated circuits of the time.
>
>And it was a 16-bitter according to MRC.
>
> Johnny
>--
>Johnny Billquist || "I'm on a bus
> || on a psychedelic trip
>email: b...@update.uu.se || Reading murder books
>pdp is alive! || tryin' to stay hip" - B. Idol

High their:

Was the PDP-14 "LINC", a commercial version of a control computer built
at MIT'S Lincoln Lab?

Mb-A.

Stephen Westin

unread,
Feb 4, 1997, 3:00:00 AM2/4/97
to

In article <5d5qms$b...@sjx-ixn7.ix.netcom.com> xen...@ix.netcom.com (Max ben-Aaron) writes:

<snip>

> High their:
>
> Was the PDP-14 "LINC", a commercial version of a control computer built
> at MIT'S Lincoln Lab?

Nope. The LINC-8 was a PDP-8 (I?) sharing a box (and, I think, memory)
with a LINC. It was a commercial product from DEC, and worked because
both the '8 and the LINC were 12-bit machines. The LINC had a bunch of
analog I/O and was useful for monitoring and controlling lab
experiments. The PDP-12 was the next generation of the LINC-8, and the
LINCtape evolved into DECtape.

I seem to recall that Cooley Lab at the University of Michigan wrote a
simple operating system for the LINC-8 which went on to become OS/8.

Douglas W. Jones,201H MLH,3193350740,3193382879

unread,
Feb 4, 1997, 3:00:00 AM2/4/97
to

In article <5d5qms$b...@sjx-ixn7.ix.netcom.com>
xen...@ix.netcom.com (Max ben-Aaron) writes:

> Was the PDP-14 "LINC", a commercial version of a control computer built
> at MIT'S Lincoln Lab?

The first LINC was built in 1962 at Lincoln Labs using DEC logic modules.
By the end of 1963, an additional 20 or so machines had been built by
their end-users. At first, users simply purchased parts from DEC, but
then DEC began to sell the parts as a kit -- was this the first computer
sold in kit form? -- and as I understand it, DEC even allowed some
customers to assemble and test the machine at Maynard and then haul it
home ready to run.

In late 1964, DEC began to manufacutre and sell ready-to-run LINC
systems, made to order. All this happened before the 1965 introduction
of the PDP-8 and well before the 1966 introduction of the LINC-8 (a machine
that could run both the PDP-8 and LINC instruction sets. Eventually,
DEC built and sold 21 LINC systems, while customers built 29. In
addition, DEC sold 142 LINC-8 systems, and something like 3500 PDP-12
systems were sold to follow up on that success. Others also came out
with LINC clones, for example, the Spear MicroLINC.

Doug Jones
jo...@cs.uiowa.edu

David M. Razler

unread,
Feb 4, 1997, 3:00:00 AM2/4/97
to

swe...@dsg145.nad.ford.com (Stephen Westin ) wrote:

| In article <5d5qms$b...@sjx-ixn7.ix.netcom.com> xen...@ix.netcom.com (Max ben-Aaron) writes:
|

| <snip>
|
| > High their:


| >
| > Was the PDP-14 "LINC", a commercial version of a control computer built
| > at MIT'S Lincoln Lab?
|

| Nope. The LINC-8 was a PDP-8 (I?) sharing a box (and, I think, memory)
| with a LINC.

PDP-8. the PDP-8/I-LINC machine was the PDP-12

| It was a commercial product from DEC, and worked because
| both the '8 and the LINC were 12-bit machines. The LINC had a bunch of
| analog I/O and was useful for monitoring and controlling lab
| experiments. The PDP-12 was the next generation of the LINC-8, and the
| LINCtape evolved into DECtape.
|

The PDP-14 was a semi-custom machine controller.

David M. Razler
david....@worldnet.att.net

Mark H. Wood

unread,
Feb 4, 1997, 3:00:00 AM2/4/97
to

> In article <19970203141...@ladder01.news.aol.com>,

> JMFBAH <jmf...@aol.com> wrote:
>>You've been talking about a program called nroff. Since I was the first
>>group leader of the RUNOFF group at DEC, would you mind enlightening me?

Hey, kudos to you and all the RUNOFF group. I used it on various systems to
write a modest pile of documentation and at least one term paper, and I credit
that experience with my ability to make decent looking documents without having
to corral and direct a hundred thousand fancy WYSIWYG gizmos. When I see what
the next generation is composing, without the benefit of that experience, I
don't know whether to laugh or cry.
--
Mark H. Wood, Lead Systems Programmer +1 317 274 0749 [@disclaimer@]
MW...@INDYVAX.IUPUI.EDU Finger for more information.
I am endeavoring to construct a mnemonic circuit using stone knives and
bearskins. -- Spock

Eric Werme

unread,
Feb 4, 1997, 3:00:00 AM2/4/97
to

jmf...@aol.com (JMFBAH) writes:

>You've been talking about a program called nroff. Since I was the first
>group leader of the RUNOFF group at DEC, would you mind enlightening me?

nroff is an old time Unix program that replaced roff, and older and, as far
as I know, lost program. (And probably no loss.) troff is a nroff
variant designed to talk to the Graphic Systems phototypesetter, but
various paths exist to talk to PostScript an other current printing
systems.

Nroff files generally include macro libraries, typcically for man pages
or one of two for documents.

Random nroff sample:

.IP "\fBgateway\fP \fIhost\fP
True if the packet used \fIhost\fP as a gateway. I.e., the ethernet
source or destination address was \fIhost\fP but neither the IP source
nor the IP destination was \fIhost\fP. \fIHost\fP must be a name and
must be found in both /etc/hosts and /etc/ethers. (An equivalent
expression is
.in +.5i
.nf
\fBether host \fIehost \fBand not host \fIhost\fR
.fi
.in -.5i
which can be used with either names or numbers for \fIhost / ehost\fP.)
.IP "\fBdst net \fInet\fR"
True if the IP destination address of the packet has a network
number of \fInet\fP, which may be either an address or a name.

Upper case commands are usually macros:

.IP - Indent Paragraph
.in - Indent (inches, points, etc)
.nf - no fill
.fi - fill

Embedded commands can change fonts - \fB bold, \fI italics. I forget
what \fR and \fP are, but you get the idea.

I assume PDP-10 RUNOFF and nroff have a common ancestor, but have never
been curious to look into it. The macro language in nroff has variables and
string processing functions that let the clever author do some
remarkable things.
--
<> Eric (Ric) Werme <> This space under reconstruction <>
<> <we...@zk3.dec.com> <> <>

Doug Humphrey

unread,
Feb 4, 1997, 3:00:00 AM2/4/97
to

Does anyone remember the Imlacs? Fortran language only
as far as I know, but I only goofed on them a little bit.
Pretty early devices...

Not dec machines, I know, but the dec-heads are the people
who I trust to understand the old machines ;-)

Doug
(PDP10s forever!)

Barry Shein

unread,
Feb 5, 1997, 3:00:00 AM2/5/97
to

There was R which ran under ITS and used similar dot commands as I
remember, and I thought R started life on CTSS. Can't say that with
any authority, just vague memories. The point being that there was
some relationship between "R" and "RUNOFF", the letter R wasn't
coincidental.


--
-Barry Shein

Software Tool & Die | b...@world.std.com | http://www.std.com
Purveyors to the Trade | Voice: 617-739-0202 | Login: 617-739-WRLD

Bob Manners

unread,
Feb 5, 1997, 3:00:00 AM2/5/97
to

> Hey, kudos to you and all the RUNOFF group. I used it on various systems to
> write a modest pile of documentation and at least one term paper, and I credit
> that experience with my ability to make decent looking documents without having
> to corral and direct a hundred thousand fancy WYSIWYG gizmos. When I see what
> the next generation is composing, without the benefit of that experience, I
> don't know whether to laugh or cry.

Don't lump all the 'next generation' together. I used TeX/LaTeX for
everything except UNIX man pages, for which I use groff.


Bob
--
------------------------------------------------------------------------------
Bob Manners 24 Temple Street, Oxford, OX4 1JS, Tel/FAX: 01865 245819
My REAL address is: r...@swift.eng.ox.ac.uk
------------------------------------------------------------------------------

JMFBAH

unread,
Feb 5, 1997, 3:00:00 AM2/5/97
to

mw...@indyvax.iupui.edu (Mark H. Wood) wrote:

>> In article <19970203141...@ladder01.news.aol.com>,
<> JMFBAH <jmf...@aol.com> wrote:

>>You've been talking about a program called nroff. Since I was the first
>>group leader of the RUNOFF group at DEC, would you mind enlightening me?

>Hey, kudos to you and all the RUNOFF group.

<snip>
Thank you. Maybe it's time to document more of the ways we used to work
when Digital was DEC.

As the programming staffs grew, the work load at Tape Prep grew. The jobs
would come in, be assessed for length of time the job would take, and get
put into a priority list. It got so complex that nobody (writers,
reproducers, nor programmers) were getting their jobs done on time. So
Tape Prep was reorganized into three groups: 1) the RUNOFF group which
would handle all work that needed to be done in RUNOFF, 2) the
quick-turnaround group which took in jobs that could be done in less than
an hour or so, and 3) all the rest of the jobs.

At that time, all documentation that was published and shipped by DEC was
done on typewriters by a group of (about) 20 women. None of that stuff
was on-line. A few exceptions were the -10 functional specs and (I think)
the processor manual. After getting broken in, I decided that everything
should start to be done on-line. I began a campaign to convince writers to
type their stuff on-line; there were some who preferred to write on paper
but not many. It took some convincing that a complete retyping of a
chapter wasn't needed just because they added one text line; this was the
case when the documents were done by typewriters.

The RUNOFF group gradually transformed itself into a real typesetting
group (long after I left). RUNOFF was not very good at manipulating text
for publishing but it was very useful. For instance, double columns that
were automatically left/right justified and space-filled were impossible
to do, but we managed [smiling emoticon here]. I believe that the only
job I ever turned down was submitted by a one of the five V.P.s (for the
life of me I can't remember the name). My boss was insistent that I do
the job for fear that the VP would get unhappy. I pointed out that he
would be even more unhappy when we produced a sloppy job that took 20
hours/month to do. What he really wanted was a VISICALC piece of
software. This was in 1972; there was no such thing. I turned down the
job, explaining to him that it was more efficient for someone with good
secretarial skills to do the job since the software didn't exist.

I was always sorry that all those women (of the typing pool) were
eventually put out of a job but I was not sorry to finally get that
documentation into bits. Interestingly enough, the person who made a
transition to typing on-line from the pool was a woman who was considered
to be backward; but she could type 120 w.p.m. When she was being trained
to work on a computer, the decision was being made that she wouldn't be
able to learn RUNOFF. The plan was that she would WYSIWYG the
documentation into a file, and someone, like me, would insert RUNOFF
commands to do the formatting. That was unacceptable and I insisted that
she be taught RUNOFF commands, monitor-level commands and everything in
between. She learned and she just loved it.

/BAH


John Everett

unread,
Feb 5, 1997, 3:00:00 AM2/5/97
to

In article <1997Feb4.1...@indyvax.iupui.edu>, mw...@indyvax.iupui.edu
says...

>
>> In article <19970203141...@ladder01.news.aol.com>,
>> JMFBAH <jmf...@aol.com> wrote:
>>>You've been talking about a program called nroff. Since I was the first
>>>group leader of the RUNOFF group at DEC, would you mind enlightening me?
>
>Hey, kudos to you and all the RUNOFF group.

I think you perhaps misunderstand, Barb (JMFBAH) was not responsible for the
creation of RUNOFF, but for its DEC-wide adoption as a documentation
standard.

The story of RUNOFF's creation is another piece of DEC folklore. Some person
(who's name is lost to history) had created a "text-processing" program that
ran on the PDP-10. He wanted to sell it to DEC. He came to Maynard one Friday
and demonstrated it to a group of us -10 types. He was careful to remove the
program from System 2 before he departed, but left behind some documentation.
On Monday, Bob Clements showed us how he had spent his weekend. He had
written a program that basically conformed to the guy's documentation. Bob
called his weekend project RUNOFF.

tli

unread,
Feb 6, 1997, 3:00:00 AM2/6/97
to

Doug Humphrey <do...@ss1.digex.net> wrote:
Does anyone remember the Imlacs? Fortran language only
as far as I know, but I only goofed on them a little bit.
Pretty early devices...

Yeah, but then again, I'm a Mudder. Mudd had an Imlac running as late as
1982.

Tony
--
Good Happens -- just a whole lot less frequently

Tom Van Vleck

unread,
Feb 6, 1997, 3:00:00 AM2/6/97
to

Peter da Silva wrote:
>Early in the history of UNIX, back before Version 7 and the great UNIX
>explosion, there was a fairly simple text processor used for formatting
>the manual pages. It was inspired by "RUNOFF" and called "ROFF" (when you
>work on ASR33 teletypes, you like making command names short!).

Jerry Saltzer wrote the RUNOFF command for CTSS on the MIT 7094.
I remember using it for documentation in 1965, so it was a public
command by then. Saltzer acknowledged inspiration from TJ-2 and
expensive typewriter.

I believe the R formatter of ITS came after RUNOFF.

Doug McIlroy, Rudd Canaday, and Dennis Ritchie
wrote a version of runoff for 645 Multics in BCPL, so that the Multics
group could use its own system for documentation. Both of these
runoffs took input lines beginning with dot as command lines.

Bob Morris and Dennis Ritchie moved the BCPL runoff to BTL's
GE-635 and called it roff. McIlroy then wrote a roff from scratch
in BCPL, expanding and improving on runoff; Ken Thompson and Dennis
Ritchie then wrote a machine-language version of roff for their new
system, UNIX, about 1970, to justify the purchase of the PDP-11.

DEC's RUNOFF command might have been inspired by CTSS's, since
the DEC engineers who went to MIT in the 60s would have seen it.
Or it might have been an independent invention, with a coincidence
in names.

Paul Wexelblat

unread,
Feb 6, 1997, 3:00:00 AM2/6/97
to

Yeah, We had an Imlac <something or other> at BBN in the Early 70's. I
don't remember much about it excepth that its assembly language was
virtually identical to a DEC system (maybe even the PDP-1) but it had
another processor in it that ran some graphics stuff (a "display list).
It was sitting back near the MTIMP (The Mag Tape IMP - yes, back then
there was a serious attempt to do realtime mag tape transfers over the
ARPAnet; this was before TCP/IP, though :-> )
...wex

Jay R. Ashworth

unread,
Feb 7, 1997, 3:00:00 AM2/7/97
to

ds...@cameonet.cameo.com.tw wrote:
: I've heard others compare WordStar to nroff, troff, et al, and I can't

: for the life of me figure out where this notion comes from.

The dot command formatting engine?

Cheers,
-- jr '.PP' a
--
Jay R. Ashworth j...@scfn.thpl.lib.fl.us
Member of the Technical Staff Unsolicited Commercial Emailers Sued
The Suncoast Freenet Pedantry: It's not just a job, it's an adventure.
Tampa Bay, Florida +1 813 790 7592

Tom Knight

unread,
Feb 7, 1997, 3:00:00 AM2/7/97
to

In article <thvv-06029...@thvv.vip.best.com> th...@best.com (Tom Van Vleck) writes:

> I believe the R formatter of ITS came after RUNOFF.

R was well after runoff -- done in the Dynamic Modelling Group (home
of Zork) and well after the more popular TJ-6 formatter used for ITS
documentation.


Alan Frisbie

unread,
Feb 7, 1997, 3:00:00 AM2/7/97
to

In article <5d8rb7$6...@ss1.digex.net>,
do...@ss1.digex.net (Doug Humphrey) writes:

> Does anyone remember the Imlacs?

Remember them? I have an Imlac PDS-1 here in my office,
along with the manuals and print sets. The last time I
turned it on, it still worked.

-- Alan E. Frisbie Fri...@Flying-Disk.Com
-- Flying Disk Systems, Inc.
-- 4759 Round Top Drive (213) 256-2575 (voice)
-- Los Angeles, CA 90065 (213) 258-3585 (FAX)

John Everett

unread,
Feb 8, 1997, 3:00:00 AM2/8/97
to

In article <thvv-06029...@thvv.vip.best.com>, th...@best.com says...

>
>DEC's RUNOFF command might have been inspired by CTSS's, since
>the DEC engineers who went to MIT in the 60s would have seen it.
>Or it might have been an independent invention, with a coincidence
>in names.

The following is from Bob Clements, the author of DEC's PDP-10 RUNOFF. The
reason I'm posting (as opposed to Bob) is self-explanatory:

"Hi, John,

I haven't been posting lately due to getting ticked off at the resulting
junk email. Haven't had time to fix the environment to hide my address
properly. Feel free to post this if you omit the cleartext form
of my email address.

In article <5daa2f$lfc$1...@kirin.wwa.com> you write:

>The story of RUNOFF's creation is another piece of DEC folklore. Some person
>(who's name is lost to history) had created a "text-processing" program that
>ran on the PDP-10.

[...]


>On Monday, Bob Clements showed us how he had spent his weekend. He had
>written a program that basically conformed to the guy's documentation. Bob
>called his weekend project RUNOFF.

Well, this isn't how I remember it, and it gives me too much credit.

I don't remember anyone coming to offer such a program to DEC, but it
might have happened.

The actual origin of RUNOFF on the -10 was indeed me, but I didn't
write it from scratch or from specs. I did a fairly direct translation
of the source code for SDS-940 RUNOFF, which was written at UCB.
I had to redo the I/O and system call stuff, of course, but the
text formating logic was all straight out of the Berkley code.

If you can find a copy of the source code of RUNOFF, you'll see big
sections with registers named A, B and X (I think that's right), which
were the main registers of the 940. It didn't have general registers
like the DEC machines. That code was just plain copied.

Today, I suppose you'd get in trouble for that. In those days, it
wasn't that big a deal. Mea culpa."

Paul Wexelblat

unread,
Feb 8, 1997, 3:00:00 AM2/8/97
to

Jay R. Ashworth wrote:
>
> ds...@cameonet.cameo.com.tw wrote:
> : I've heard others compare WordStar to nroff, troff, et al, and I can't
> : for the life of me figure out where this notion comes from.
>
> The dot command formatting engine?

<snip>

I could guess, but I expect that it's not common knowledge that WortStar
_was_
written by a PDP-1 programmer. Wordsatr (as orifginally written for the
Imasi (I think))
predates RUNOFF (predicessor to nroff).
I don't know TJ-2, but I did write a text justification program for the
PDP-1 that was
implemented in Logo. (I also wrote some of Logo, that was written for
the PDP-1 in Assembler
- Midas (predicessor to MACRO) but that's a different story.

Please address specific follows-up to alt.folklore.computers as I dont
subscribe to all the
crossposted groups mentioned.
...wex

Douglas W. Jones,201H MLH,3193350740,3193382879

unread,
Feb 9, 1997, 3:00:00 AM2/9/97
to

From article <32FD3B...@concentric.net>,
by Paul Wexelblat <w...@concentric.net>:

>
> I could guess, but I expect that it's not common knowledge that WortStar
> _was_ written by a PDP-1 programmer. Wordsatr (as orifginally written for the
> Imasi (I think)) predates RUNOFF (predicessor to nroff).

????? RUNOFF (on MULTICS) dates from the late 1960's. That begat RUNOFF
on the Bell Labs GCOS system (I have the manuals for it that I got when
I worked at Bell in 1973), and that begat NROFF and TROFF in around 1973.
(The CAT Phototypesetter was still a hot new toy at Murray Hill when I
was there).

Nothing developed for the IMSAI predates any of this!!!!

A far more interesting lineage is the one that links TJ-2 to HTML. Anyone
want to lay it out in full? (Hint: HTML grew in the context of SGML,
which originally grew in the context of a type formatter on the IBM-360
family, which was based on ideas developed at IBM Cambridge Labs, which
had a connection with work done on CTSS at MIT -- the common ancestor
from which RUNOFF grew, by the way -- which was, no doubt, inspired by
TJ-2.)
Doug Jones
jo...@cs.uiowa.edu

JMFBAH

unread,
Feb 10, 1997, 3:00:00 AM2/10/97
to

jeve...@wwa.DEFEAT.UCE.BOTS.com (John Everett) wrote a memory correction:

<snip>


<I think you perhaps misunderstand, Barb (JMFBAH) was not responsible for
the
<creation of RUNOFF, but for its DEC-wide adoption as a documentation
<standard.

<snip documentation about the real author of RUNOFF>

Yup. I did misunderstand this person. I also did a great big blunder
when I wrote that Pete Conklin wrote RUNOFF; I did not remember that Bob
wrote it. Please forward my humblest apologies to Bob Clements [emoticon
here wincing from the application of 2 lashes of a wet noodle].

Now, I'm befuddled about what Pete did write that helped me put bread on
my table. FACTPR maybe?

/BAH

Alan Bowler

unread,
Feb 10, 1997, 3:00:00 AM2/10/97
to

In article <5cvtfc$22j$3...@kirin.wwa.com> jeve...@wwa.DEFEAT.UCE.BOTS.com (John Everett) writes:
>In article <01bc0fe8$97d6d740$0b4284a9@wardch>, c...@mail.idt.com says...
>>
>>Along with the knowledge that the Data General Nova was decended from the
>>PDP-8
>>(Edson DeCastro worked on both)
>
>Ain't revisionist history wonderful! Actually the Nova was descended (as
>opposed to decended) from the PDP-X (or X-project) at DEC. This was DEC's
>first attempt at a 16 bit machine, rejected by the operations committee.
>Eventually a new design emerged as the PDP-11.

It is not all that revisionist. While the Nova was a 16 bit system,
it had a very pdp-8 flavour to it. The fact that internal to DEC
there was a (rejected) intermediate design between the PDP-8 and the Nova
doesn't much change the fact that it is a "spiritual" decendant of the
PDP-8.

Mike Albaugh

unread,
Feb 11, 1997, 3:00:00 AM2/11/97
to

John Everett (jeve...@wwa.DEFEAT.UCE.BOTS.com) wrote:

: The following is from Bob Clements, the author of DEC's PDP-10 RUNOFF.

: The actual origin of RUNOFF on the -10 was indeed me, but I didn't


: write it from scratch or from specs. I did a fairly direct translation
: of the source code for SDS-940 RUNOFF, which was written at UCB.

Hmmm, I thought it looked familiar, but the "Runoff"-like
thingy on the [SX]DS-9[34]0 :-) was, AFAIK, called "Autosecretary".
I may even have a manual buried in those boxes of books in the attic.
I never actually had an account on the '940, being an undergrad
at the time, but I _did_ pull cable for its terminals and once even
repair one of the boards in the terminal interface. For you young'uns,
this was a rack of DEC "System Modules" (Preceded Flip Chips, and
were also used on the PDP-6 we had) where each line interface (what
we'd call a UART today) used two or three cards. I know the tranmitter
and receiver were separate, I just don't recall whether there was
also a separate "control" card. Perhaps I can stir some memories
out there. These little wonders were so well designed that the
later UARTs from Western Digital and General Instruments pretty
much copied them verbatim (Perhaps they were designed to DEC spec),
and I always get a little peeved when I run into a serial interface
that is _less_ well designed but done later (8530, 8250, Rockwell VIA...)

Anyway, the odd thing I remember about Autosecretary was
that it output a tape that had to be flipped over and fed to a
Flexowriter. Apparently Friden and Teletype had differing ideas
about where the sprocket holes should go :-)

: If you can find a copy of the source code of RUNOFF, you'll see big


: sections with registers named A, B and X (I think that's right),

Matches my recollection, although as I said before I did not get
_that_ close to the machine. Amazing what a machine with 16K of
24-bit words per process (about 64 or 128K total) could do. OTOH,
the limited per-process space made "fork" the tool of choice
for task decomposition, and I'm still living with _that_ headache
today on my company-issue Unix machine, _years_ after per-process
address space made single-address-space tasks possible...
(I also blame them for making "delete" the default "interrupt"
key. Made using the paper-tape reader a real thrill :-)

: Today, I suppose you'd get in trouble for that. In those days, it


: wasn't that big a deal. Mea culpa."

As I recall, Project Genie (the 930->940 conversion and
software development) was ARPA-funded, and the source code was
freely distributable.

Mike
| alb...@agames.com, speaking only for myself

John Bayko

unread,
Feb 11, 1997, 3:00:00 AM2/11/97
to

In article <E5EGE...@thinkage.on.ca>,

Here's what the Great Microprocessors List has to say about the
various descendants of the PDP-8 - the complete descriptions are
available in the list from my home page,or the CPU Info Center at:

http://infopad.eecs.berkeley.edu/CIC/

Great Microprocessors of the Past and Present (V 9.2.2)
[...]
Section One: Before the Great Dark Cloud.
[...]
Part V: The 6809, extending the 680x (1977)

Like the 6502, the 6809 was based on the Motorola 6800 (1974), which
was initially based on the design of the big endian DEC PDP-8, though
the 6809 expanded the design significantly. [...]

Section Two: Forgotten/Innovative Designs before the Great Dark Cloud
[...]
Part VI: Intersil 6100, old design in a new package

The IMS 6100 was a single chip design of the PDP-8 minicomputer
(1965) from DEC (low cost successor to the PDP-5 (1963)).
The old PDP-8 design was very strange, and if it hadn't been
so popular, an awkward CPU like the 6100 would have never had a reason
to exist.
[...]

Part VII: NOVA, another popular adaptation

Like the PDP-8, the Data General Nova was also copied, not just in
one, but two implementations - the Data General MN601 (MicroNova),
and Fairchild 9440. However, the NOVA was a more mature design (by PDP-8
designer Edson DeCastro, who came to Data General from DEC).
[...]
Another CPU, the PACE, was based on the NOVA design, but featured
16 bit addressing, more addressing modes, and a 10 level stack (like
the 8008).
The 32 bit ECLIPSE (pre 1983) was Data General's successor to the 16
bit Nova. [...]

Part VIII: Signetics 2650, enhanced accumulator based (1978?)

Superficially similar to the PDP-8 (and IMS 6100), the Signetics
2650 was based around a set of 8 bit registers with R0 used as an
accumulator, [...]

--
John Bayko (Tau).
ba...@cs.uregina.ca
http://www.cs.uregina.ca/~bayko

Ross Alexander

unread,
Feb 12, 1997, 3:00:00 AM2/12/97
to

I've edited heavily since my comment is a brief one.

ba...@borealis.cs.uregina.ca (John Bayko) writes:
> Here's what the Great Microprocessors List has to say about the
>various descendants of the PDP-8 - the complete descriptions are

[...]


> Like the 6502, the 6809 was based on the Motorola 6800 (1974), which
>was initially based on the design of the big endian DEC PDP-8, though
>the 6809 expanded the design significantly. [...]

It's not at all clear to me what the line of descent between a machine
(6800) with

two accumulators (mostly orthogonal)
PC relative displacement branches
condition codes
8-bit organization
hardware stack
index register
vectored interrupts (admittedly primitive)

and a earlier machine (pdp8) with

one accumulator and a totally non-orthogonal quotient register
page relative branches
no condition codes
12 bit organization
no stack
no index register (I don't count mem locs 0010 .. 0017 as index registers)
flat interrupts

might be beyond that fact that they're both small 2s complement von
Neumann machines. I've hacked on both and they have totally different
`feels'.

In short, the 6800 doesn't seem to share a lineage with the pdp8.
The intersil 6100, definitely.

regards,
Ross
--
Ross Alexander, ve6pdq -- (403) 675 6311 -- r...@cs.athabascau.ca

Douglas W. Jones,201H MLH,3193350740,3193382879

unread,
Feb 12, 1997, 3:00:00 AM2/12/97
to

ba...@borealis.cs.uregina.ca (John Bayko) writes:
> Here's what the Great Microprocessors List has to say about the
>various descendants of the PDP-8 - the complete descriptions are

> Like the 6502, the 6809 was based on the Motorola 6800 (1974), which


>was initially based on the design of the big endian DEC PDP-8, though
>the 6809 expanded the design significantly. [...]

No! As near as I can tell, the 6800 had nothing to do with the PDP-8.
Single accumulator machines were the rule, not the exception, from 1946
when Berks, Goldstein and Von Neumann first proposed the fetch-execute
cycle as we know it, and the late 1950's. In the 1960's, virtually all
of the significant minicomputers were single-accumulator machines, and
strong shawows of that world-view held on in the first generation of
microprocessors. The only major micrpocessor today with strong vestiges
of this worldview is, of course, the Intel family of 80x86 machines.

Anyway, someone needs to get their hands on the documentation for
Motorola's first 8-bit processor, the MDP-1000, released in 1968. It
wasn't a microprocessor, it was a minicomputer, made with TTL, I think,
but I'll lay odds that the 6800 architecture draws on the experience of
the MDP-1000 far more than it draws on the PDP-8.

Also, John Bayko's list completely missed the HP 21XX series of
minicomputers. Both the HP 21XX and DG NOVA series can be described
as 16 bit versions of the PDP-8, but in widening the word, the two
teams came to different conclusions about what to do with the extra
bits (HP used 2 accumulators, DG used 4, and so on). DG was a DEC
spinoff, while HP (or rather, an HP subsidiary) was a major OEM customer
of DEC PDP-8 hardware in the mid 1960's -- so major that, according to
an interview published by CHAC, HP seriously considered buying DEC
before they decided to make their own machines.

Doug Jones
jo...@cs.uiowa.edu

John Bayko

unread,
Feb 13, 1997, 3:00:00 AM2/13/97
to

In article <5dt21u$f...@aurora.cs.athabascau.ca>,

Ross Alexander <r...@cs.athabascau.ca> wrote:
>> Like the 6502, the 6809 was based on the Motorola 6800 (1974), which
>>was initially based on the design of the big endian DEC PDP-8, though
>>the 6809 expanded the design significantly. [...]
>
>It's not at all clear to me what the line of descent between a machine
>(6800) with
>
> two accumulators (mostly orthogonal)
> PC relative displacement branches
> condition codes
> 8-bit organization
> hardware stack
> index register
> vectored interrupts (admittedly primitive)
>
>and a earlier machine (pdp8) with
>
> one accumulator and a totally non-orthogonal quotient register
> page relative branches
> no condition codes
> 12 bit organization
> no stack
> no index register (I don't count mem locs 0010 .. 0017 as index registers)
> flat interrupts
>
>might be beyond that fact that they're both small 2s complement von
>Neumann machines. I've hacked on both and they have totally different
>`feels'.
>
>In short, the 6800 doesn't seem to share a lineage with the pdp8.
>The intersil 6100, definitely.

Pretty indirect, but the basic idea was followed, at least
initially - the 6800 was the only accumulator-based microprocessor at
the time, compared to the register-based 8080. I think the differences
can be mostly attributed to learning from past mistakes.
I read in the same article that Motorola intended the 6809 to be
more like the PDP-11, while other designers wanted to continue
refining the 6800, and this is part of what eventually caused one
group to split off and form MOS Technologies to design the 650x. On
the other hand, apart from adding interesting addressing modes, I don't
know exactly how the 6809 was very PDP-11-ish...
Maybe 'inspired' would be a better description of the 6800's
relation to the PDP-8.

Bob Manners

unread,
Feb 13, 1997, 3:00:00 AM2/13/97
to

>the 6800 was the only accumulator-based microprocessor at
>the time, compared to the register-based 8080.

Pardon my ignorance here, but how would one define 'accumulator-based'
and 'register-based'?

As far as I'm aware of the (relatively recent) use of the word
'accumulator', it seems to refer to a register in a processor which is
sufficiently non-orthogonal that maths operations may not be done in
the other registers.

Does that make any sense?

Carl R. Friend

unread,
Feb 14, 1997, 3:00:00 AM2/14/97
to

Bob Manners wrote in article Nr. <RJM.97Fe...@europa.ox.ac.uk>:

>
> Pardon my ignorance here, but how would one define 'accumulator-based'
> and 'register-based'?
>
> As far as I'm aware of the (relatively recent) use of the word
> 'accumulator', it seems to refer to a register in a processor which is
> sufficiently non-orthogonal that maths operations may not be done in
> the other registers.

If I'm remembering things correctly here (feel free to correct me!
Dr. Jones?), the original "accumulators" were essentially "self-adders"
which both held a value, and, when a new item was submitted to them,
they added the submitted datum to their current contents, thereby
"accumulating" incoming data. The term is (in a computer sense) a
very old one.

In "modern" machines, the terms "accumulator" and "register" are
often used interchangably, although "accumulator" seems to be falling
out of favour.

Note that the term "register" didn't always denote a processor-
internal scratchpad; in the LINC machine designed in the early '60s,
the term "register" was used to refer to what we'd now call a memory
location. Hence, in the LINC, you had 1024 "registers" in both the data
and instruction fields, and a single 12-bit "accumulator" (which was,
in fact, a simple 12-bit latch with input gating that could engage an
adder.

The PDP-8 follows the same analogy; the main "register" is called an
"accumulator", possibly for no other reason than that's what the
designers were familiar with. The PDP-10 architecture sported sixteen
"ACs", which were also known as "Fast Memory" (in the KA, FM was an
option - without it the 16 ACs dwelt in the first 16 words of core).

--
______________________________________________________________________
| | |
| Carl Richard Friend (UNIX Sysadmin) | West Boylston |
| Minicomputer Collector / Enthusiast | Massachusetts, USA |
| mailto:carl....@stoneweb.com | |
| http://www.ultranet.com/~engelbrt/carl/museum | ICBM: N42:22 W71:47 |
|________________________________________________|_____________________|

gag...@clark.net

unread,
Feb 15, 1997, 3:00:00 AM2/15/97
to

In article <5djhge$s...@flood.weeg.uiowa.edu>,

Douglas W. Jones,201H MLH,3193350740,3193382879 <jo...@pyrite.cs.uiowa.edu> wrote:
>From article <32FD3B...@concentric.net>,
>by Paul Wexelblat <w...@concentric.net>:
>
<snip>

>(The CAT Phototypesetter was still a hot new toy at Murray Hill when I
>was there).
>
I'm probably wrong on the 'who was first' here, but the Typeset-10
group at DEC (Tim Stein and myself, and others) wrote a driver for
the CAT pretty early after its introduction. It was a neat little
typesetting machine, and we were glad to get one because it was
the first 'real' typesetting machine we had in house. I recall the
date as 1973-74 or so, but could be wrong. Others who worked on
Typeset-10 included Dan Bricklin. Typeset-10 was roughly modeled
after Typeset-8, which was a wildly popular newspaper production
system. The -10 version (in MACRO-10) had a number of innovative
features, including a neat hyphenation algorithm that used a cute
hashing algorithm derived from the -10's ability to do infinite
level indirection and clever bit operations. It ran circles around
anything else that we knew of.

The Typeset-10 language was, in fact, modeled after the way that
MIDAS (the -10 assembler at MIT) handled macros. You could
do integer arithmetic in it, using the algorithms from
Peter Sampson's wonderful STRING language, which allowed for
infinite precision. Also, if you wanted,
you could ouput the results directly was word strings in english,
or as roman numerals.

Originally, Typset-10 had been planned to use RUNOFF syntax, but
it was clear that such a syntax was too cumbersome. Some vestiges
remained, which was unfortunate or we would have invented TEX.

>
>A far more interesting lineage is the one that links TJ-2 to HTML. Anyone
>want to lay it out in full? (Hint: HTML grew in the context of SGML,
>which originally grew in the context of a type formatter on the IBM-360
>family, which was based on ideas developed at IBM Cambridge Labs, which
>had a connection with work done on CTSS at MIT -- the common ancestor
>from which RUNOFF grew, by the way -- which was, no doubt, inspired by
>TJ-2.)

I recall using RUNOFF or something very like it on the 7094
CTSS system at MIT, which would have been in the mid/late sixties,
and of course MIT-AI later had a much more powerful
version for both the PDP-6 and later for the PDP-10. So I'd put
the evolutionary chain as CTSS --> MULTICS/MIT-AI and lots of
branches after that. None of which is to speak of Electric Typewriter,
and the many other efforts that were going on at the same time,
including a truly bizzare text processor in use on the Phoenix
machine at MITRE, which Joe Morris probably remembers much better
than I do.


Robert Houk - SMCC Bos Desktop Hardware

unread,
Feb 16, 1997, 3:00:00 AM2/16/97
to

In article <5e30i3$5...@clarknet.clark.net> gag...@clark.net () writes:

>(The CAT Phototypesetter was still a hot new toy at Murray Hill when I
>was there).

I'm probably wrong on the 'who was first' here, but the Typeset-10
group at DEC (Tim Stein and myself, and others) wrote a driver for
the CAT pretty early after its introduction. It was a neat little
typesetting machine, and we were glad to get one because it was
the first 'real' typesetting machine we had in house. I recall the
date as 1973-74 or so, but could be wrong. Others who worked on

Probably pretty accurate, as the CAT (wasn't it "CATT" or "C/A/T/T",
or something like that?) was "well established" (and later nicknamed
by me as the "grind grind stink stink" machine) by the end of 1975
when I got there and inherited ITPS-10 (aka Typeset-10 in drag).

Typeset-10 included Dan Bricklin. Typeset-10 was roughly modeled
after Typeset-8, which was a wildly popular newspaper production
system. The -10 version (in MACRO-10) had a number of innovative
features, including a neat hyphenation algorithm that used a cute
hashing algorithm derived from the -10's ability to do infinite
level indirection and clever bit operations. It ran circles around
anything else that we knew of.

As I very vaguely recall, I timed it at around 2000 lines/second on
a KI-10, which I thought was pretty amazing, all things considered.

Originally, Typset-10 had been planned to use RUNOFF syntax, but
it was clear that such a syntax was too cumbersome. Some vestiges
remained, which was unfortunate or we would have invented TEX.

I wrote a RUNOFF macro for it, and managed to hornswoggle a friend
into completing Conklin's (?) abortive RUNOFF/TYPESET interface, so
we could take standard RUNOFF input and "typeset it". People were
pretty amazed to see hyphenated RUNOFF output come out on the line-
printers, let-alone proportionally-spaced stuff on the Diablo!
(Don't think we ever quite mastered footnotes, though...)

Ah, the good ole days . . . I *still* don't like WYSIWSYG formatters
like Word et al (although they do make tables a *lot* easier)

-RDH

jmfb...@ma.ultranet.com

unread,
Feb 17, 1997, 3:00:00 AM2/17/97
to

In article <c3b914ojsrp.fsf@urq>,
>like Word et al (although they do make tables a *lot* easier).

'ey Bob!

I remember using that software to create great galleys. I also remember
using your terminal, GT-40?maybe?. I seem to remember that it sounded
like a small airplane. I would type in RUNOFF format, run the .RNO file
through some kind of pre-formatter and then (somehow) get it printed on
the machine that was on MRO-1. It produced great copy; a great leap
forward from the Diablo.

By the way, this is my first test of entering an item using this ISP.
(AOL became a swear when it started to tell me to get off the system!).

/BAH

Brian Harvey

unread,
Feb 17, 1997, 3:00:00 AM2/17/97
to

People in this thread have been comparing RUNOFF and friends with TeX.
You might be interested to know that the PDP-10 was the host for an early
TeX precursor, the professional typesetting system that I worked on (along
with Lowell Hawkinson and some others) at Composition Technology
circa 1970, I believe the first program that handled high-quality
typesetting of mathematical formulas from user input in math terms
rather than exact positional terms.

Mark H. Wood

unread,
Feb 20, 1997, 3:00:00 AM2/20/97
to

Yup -- the "shadow ACs" area was underneath them, and the Monitor could use it
somehow (but it's been too long for me).

The IBM 360/44 used core for all kinds of stuff that you might not expect,
including registers. Again real registers were an extra-cost option I believe.
(It's also been many moons since I worked on one of these.) But you couldn't
load inner loops into the registers and jump to them like TECO reputedly did on
the PDP-10.
--
Mark H. Wood, Lead Systems Programmer +1 317 274 0749 [@disclaimer@]
MW...@INDYVAX.IUPUI.EDU Finger for more information.
I am endeavoring to construct a mnemonic circuit using stone knives and
bearskins. -- Spock

Chris Zach

unread,
Feb 21, 1997, 3:00:00 AM2/21/97
to mw...@indyvax.iupui.edu

> The IBM 360/44 used core for all kinds of stuff that you might not expect,
> including registers. Again real registers were an extra-cost option I
believe.
> (It's also been many moons since I worked on one of these.) But you
couldn't
> load inner loops into the registers and jump to them like TECO reputedly
did on
> the PDP-10.

Of course, the untimate hell was the PDP-8/S. With a 1 bit accumulator (ok, a
flip flop) I think it sequenced the instructions and data through the flip
flop one bit at a time. A true serial computer....

CZ
--
Time to take time
For Spring will turn to Fall
In just no time at all...


Smith and O'Halloran

unread,
Feb 21, 1997, 3:00:00 AM2/21/97
to

In article <1997Feb20.1...@indyvax.iupui.edu>,

Mark H. Wood <mw...@indyvax.iupui.edu> wrote:
>> The PDP-10 architecture sported sixteen
>> "ACs", which were also known as "Fast Memory" (in the KA, FM was an
>> option - without it the 16 ACs dwelt in the first 16 words of core).
>
>Yup -- the "shadow ACs" area was underneath them, and the Monitor could use it
>somehow (but it's been too long for me).

Not quite. Whether the AC references went to Fast Memory or to absolute
core locations 00-17 on the KA-10 depends on the setting of the "FM ENB"
switch on the console front panel.

Normally, FM ENB was left ON, enabling Fast Memory. But if you had
previously installed the 16-word bootstrap program into core while
FM ENB was OFF, the program would still be there inspite of intervening
power failures.

The true magic of the PDP-10 instruction set was in the RIM-10B loader.
The program is 14 instructions long and uses 2 accumlators for data.
It reads in 36-bit words from an 8-bit paper tape reader, deposits the
data in memory, and verifies the checksums of the program it is loading.
-Joe
--
INWAP.COM is Joe and Sally Smith, John and Chris O'Halloran and our cats
See http://www.inwap.com/ for "ReBoot", PDP-10, and Clan MacLeod.

John Everett

unread,
Feb 21, 1997, 3:00:00 AM2/21/97
to

In article <5ej2j4$3...@student.computer.org>, c...@alembic.crystel.com says...

>
>> The IBM 360/44 used core for all kinds of stuff that you might not expect,
>> including registers. Again real registers were an extra-cost option I
>believe.
>> (It's also been many moons since I worked on one of these.) But you
>couldn't
>> load inner loops into the registers and jump to them like TECO reputedly
>did on
>> the PDP-10.
>
>Of course, the untimate hell was the PDP-8/S. With a 1 bit accumulator (ok,
a
>flip flop) I think it sequenced the instructions and data through the flip
>flop one bit at a time. A true serial computer....

And performance to match. As a charter member of the PDP-8 programming group
(when the "group" was Roger Pyle and me) I had to test everything I wrote for
PDP-8/S compatibility. Some of the most boring hours I've ever spent.

Smith and O'Halloran

unread,
Feb 22, 1997, 3:00:00 AM2/22/97
to

In article <5ekohr$b...@shell3.ba.best.com>,

Smith and O'Halloran <in...@shell3.ba.best.com> wrote:
>The true magic of the PDP-10 instruction set was in the RIM-10B loader.
>The program is 14 instructions long and uses 2 accumlators for data.
>It reads in 36-bit words from an 8-bit paper tape reader, deposits the
>data in memory, and verifies the checksums of the program it is loading.

Someone asked me to post the RIM10B loader. Here it is, with an analysis.

Somewhere I read the story of how it was created; a really bright hacker
was tricked into creating it when his colleagues kept saying that it
couldn't be done. Anyone have the details?
-Joe

lynx -dump http://www.inwap.com/pdp10/rim10b.html

RIM10B LOADER

RIM10B ; Causes RIM10B loader to be punched
00/ 777762,,0 XWD -16,0
01/ 710600,,60 ST: CONO PTR,60
02/ 541400,,4 ST1: HRRI A,RD+1
03/ 710740,,10 RD: CONSO PTR,10
04/ 254000,,3 JRST .-1
05/ 710470,,7 DATAI PTR,@TBL1-RD+1(A)
06/ 256010,,7 XCT TBL1-RD+1(A)
07/ 256010,,12 XCT TBL2-RD+1(A)
10/ 364400,,0 A: SOJA A, ; Magic occurs here ****
11/ 312740,,16 TBL1: CAME CKSM,ADR
12/ 270756,,1 ADD CKSM,1(ADR)
13/ 331740,,16 SKIPL CKSM,ADR
14/ 254200,,1 TBL2: JRST 4,ST
15/ 253700,,3 AOBJN ADR,RD
16/ 254000,,2 ADR: JRST ST1
17/ CKSM=ADR+1

Here is an example of a two-word program as output by RIM10B

17/ 777776,,777 LOC 1000 ; Set starting address
20/ 201740,,3777 START: MOVEI 17,4000-1
21/ 505740,,777600 HLRI 17,-200
22/ 707677,,4576 ; Sum of previous 3 words
23/ 254000,,1000 END START

ANALYSIS

RIM
When the Read-In Mode (RIM) switch is pressed on the console of
a KA or KI, it sends a reset pulse down the I/O bus, sets the
PC flags to zero, and executes "DATAI D,0" (where D is the
device code selected by a set of 7 switches, the paper tape
reader is device 104). The DATAI reads in an IOWD, which has
the negative word count in the left half and starting address
minus one in the right half. The CPU then repeatedly executes
"BLKI D,0" until the left half of location 0 reaches zero.
("BLKI D,X" increments both halves of location X, reads in a
word from device D, and stores it the address that the right
half of location X now points to.)

00/ XWD -16,0
Transfer 16 octal (14 decimal) words, starting at location 1.

01/ST: CONO PTR,60
Start paper tape reader in binary mode

02/ST1: HRRI A,RD+1
Reset finite-state machine to looking for IOWD

State RD+1 = Looking for IOWD or JRST

03/RD: CONSO PTR,10
Read paper tape reader status, skip if "DONE" bit is set

04/ JRST .-1
Not set, keep looping until the bit does get set

05/ DATAI PTR,@TBL1-RD+1(A)
Index register A has RD+1, indexing TBL1-RD+1+RD+1 is TBL1+2,
which is the SKIPL CKSM,ADR instruction, therefore the
effective address is ADR. Store the IOWD in ADR.

06/ XCT TBL1-RD+1(A)
Same effective address, "SKIPL CKSM,ADR" loads the IOWD into
accumulator CKSM, and skips next instruction because its
negative.

07/ XCT TBL2-RD+1(A)
Not executed first time around. At the end of the tape, a JRST
instruction will be read in instead of an IOWD. (JRST is opcode
254, which is postitive). TBL2-RD+1+RD+1 is TBL2+2, which is
ADR. The JRST instruction which was just read in is executed,
and that causes the PC to jump to the beginning of the program.

10/A: SOJA A,RD+1
Set the PC to RD+1, subtract one from index register A (so it
now has RD in the right half, then jump to the original address
(RD+1).

04/ JRST .-1
Not set, keep looping until the bit does get set

State RD+0 = Reading in data words

03/RD: CONSO PTR,10
Read paper tape reader status, skip if "DONE" bit is set

04/ JRST .-1
Not set, keep looping until the bit does get set

05/ DATAI PTR,@TBL1-RD+1(A)
Index register A has RD+0, indexing TBL1-RD+1+RD+0 is TBL1+1,
which is the ADD CKSM,1(ADR) instruction, therefore the
effective address is one greater than what ADR points to. Store
the data in memory.

06/ XCT TBL1-RD+1(A)
Same effective address, "ADD CKSM,1(ADR)" adds the word read in
to the additive checksum in accumulator CKSM.

07/ XCT TBL2-RD+1(A)
The address is TBL2-RD+1+RD+0 which is TBL2+1. That location
has "AOBJN ADR,RD". Add one to both halves of accumulator ADR.
If the result is still negative, loop back to RD (location 3).
If non-negative, continue on at location 10.

10/A: SOJA A,RD+0
Set the PC to RD+0, subtract one from index register A (so it
now has RD-1 in the right half, then jump to the original
address (RD+0).

State RD-1 = Reading in checksum

03/RD: CONSO PTR,10
Read paper tape reader status, skip if "DONE" bit is set

04/ JRST .-1
Not set, keep looping until the bit does get set

05/ DATAI PTR,@TBL1-RD+1(A)
Index register A has RD-1, indexing TBL1-RD+1+RD-1 is TBL1+0,
which is the CAME CKSM,ADR instruction, therefore the effective
address is ADR. Store the expected checksum in ADR.

06/ XCT TBL1-RD+1(A)
Same effective address, "CAME CKSM,ADR" compares the calculated
checksum in accumulator CKSM with the expected checksum stored
in memory location ADR. Skip the next instruction if they're
equal.

07/ XCT TBL2-RD+1(A)
The address is TBL2-RD+1+RD-1 which is TBL2+0. That location
has "JRST 4,ST" which is a HALT instruction. If the previous
compare instruction failed, set the program counter to ST and
halt the CPU. This allows the operator to back up the paper
tape reader and try again. If the CAME succeeded, this HALT is
not executed.

10/A: SOJA A,RD+1
Set the PC to RD+1, subtract one from index register A (so it
now has RD-2 in the right half, then jump to the original
address (RD+1). This jumps to location ST1, which resets the
finite-state machine.

Dispatch table for finite-state machine

11/TBL1: CAME CKSM,ADR
In state RD-1, read expected checksum into ADR, then compare
calculated checksum with expected checksum.

12/ ADD CKSM,1(ADR)
In state RD+0, store data word into memory, then add data word
into running checksum.

13/ SKIPL CKSM,ADR
In state RD+1, store IOWD or JRST in ADR, then load that word
into accumulator CKSM and skip if the word is negative.

14/TBL2: JRST 4,ST
If the checksum comparison fails, halt the CPU, with ST in the
PC.

15/ AOBJN ADR,RD
In state RD+0, increment the IOWD and jump to RD if more to go.

16/ADR: JRST ST1
This is the last word of the RIM10B loader. When the hardware
read-in process is completed, this instruction is executed to
start the program.

17/CKSM=ADR+1
The additive checksum is calculated using this accumulator.

STORING BOOTSTRAP IN CORE MEMORY

The FM ENB switch enables Fast Memory, causing references to the
accumulators (locations 00 through 17) to go to RAM instead of core
memory. When FM ENB is off, the above bootstrap can be toggled into
locations 01 through 16. (Locations 00 and 17 need not be
initialized.)
[End of http://www.inwap.com/pdp10/rim10b.html]

jmfb...@ma.ultranet.com

unread,
Feb 25, 1997, 3:00:00 AM2/25/97
to

In article <5enk2j$c...@shell3.ba.best.com>,
<Ultranet snipped more discussion>


Aaaahhh!!! JRSTs, HRLIs, CAMEs. Isn't it wonderful to speak a familiar language
for a change?

/BAH

Carl R. Friend

unread,
Feb 27, 1997, 3:00:00 AM2/27/97
to

Smith and O'Halloran wrote in
article Nr. <5enk2j$c...@shell3.ba.best.com>:
>
> [RIM loader code saved and snipped]

>
> STORING BOOTSTRAP IN CORE MEMORY
>
> The FM ENB switch enables Fast Memory, causing references to the
> accumulators (locations 00 through 17) to go to RAM instead of core
> memory. When FM ENB is off, the above bootstrap can be toggled into
> locations 01 through 16. (Locations 00 and 17 need not be
> initialized.)
>
Thanks for that wonderful bit of code! That's one for the files.

In any event, this technique would only work on the KA-10. Later
-10s used semiconductors exclusively, so there wasn't any way to save/
load anything to the lowest of low core.

As an interesting sidelight, the KI-10 had four sets of AC (or FM,
if you will) selectable by the operating system or by front panel
switches (IIRC, in the same little panel that had the margining con-
trols and ReadIn Device switches). Whether DEC ever made use of the
added ACs, I'm not sure, but doing so would certainly have helped
during context switches.

Tom Moser

unread,
Feb 27, 1997, 3:00:00 AM2/27/97
to

In article <33157086...@swec.com>, "Carl R. Friend" <carl....@swec.com> writes:
> Smith and O'Halloran wrote in
> article Nr. <5enk2j$c...@shell3.ba.best.com>:
> >
> > [RIM loader code saved and snipped]
> >
> > STORING BOOTSTRAP IN CORE MEMORY
> >
> > The FM ENB switch enables Fast Memory, causing references to the
> > accumulators (locations 00 through 17) to go to RAM instead of core
> > memory. When FM ENB is off, the above bootstrap can be toggled into
> > locations 01 through 16. (Locations 00 and 17 need not be
> > initialized.)
> >
> Thanks for that wonderful bit of code! That's one for the files.
>
> In any event, this technique would only work on the KA-10. Later
> -10s used semiconductors exclusively, so there wasn't any way to save/
> load anything to the lowest of low core.
>

I don't think this is true. I will be very embarrased if I am wrong but
I believe that on the KL you could just PMAP another page to page 0 and
then any writes to location 0-17 in that page would go to the shadow ACs.

Maybe someonw with a KL running TOPS-20 can try this - I haven't written
code for a PDP-10 for 12 years so you might need to tweak this but you
will get the idea.

HRLZI T1,.FHSLF /* These 2 lines might be to
MOVE T2, [.FHSLF,,5000] /* the wrong ACs (source vs dest)
MOVX T3, PM%RD+PM%WR
PMAP% /* map page 0 to page 5

Now any references to 5000-5017 will end up in the shadow ACs but will NOT
appear in the ACs and refs to 5020-5777 will be visible in page 0.<