back to article PDP-10 enthusiasts resurrect ancient MIT operating system

Among the more interesting denizens of GitHub are the many projects devoted to rediscovering and preserving the history of computing – such as a system called the Incompatible Timesharing System for the legendary Digital PDP-10. If Wikipedia is correct, ITS (its name a play on an earlier MIT project, the Compatible Time …

  1. John Smith 19 Gold badge

    DECs funny numbering sytem

    The DEC 10 is actually much bigger than the PDP11, and deeply incompatible (36bits Vs 16), running at a whole 1MHz.

    OTOH the DEC 10 hosted some iconic AI stuff like SHRDLU. I'm guessing some early CAD software might have also turned up on it.

    Looking forward to seeing this work.

    Today the benchmark for this sort of performance would be a cheap smartphone. :-(

    1. Rob Willett

      Re: DECs funny numbering sytem

      I think the benchmark for this would be my cheap Fitbit or my coffee machine, never mind a cheap smartphone :)

    2. Richard Tobin

      Re: DECs funny numbering sytem

      The DEC 10 was also known for its Prolog system, much the fastest implementation at that time, which was developed in the AI department at Edinburgh University.

      It ran under TOPS-10 rather than ITS, however.

      1. John Smith 19 Gold badge

        "The DEC 10..also known for its Prolog system,..the fastest implementation at that time,"

        I'm guessing this may have been used to demonstrate that deterministic parsing of English was possible with only 2 units of lookahead in the mid 80's (Marcus's Parsifal used 3 but the Edingburgh these demonstrated you only needed 2, although they could be complex items).

        Another iconic bit of AI.

    3. GruntyMcPugh Silver badge

      Re: DECs funny numbering sytem

      DEC 10,.... the mention brought back some memories, but it was of a DEC 20 now I come to think of it, and it was the first multi user computer I ever logged into, in '86 at Trent Poly. That was it's last academic year in service, we moved onto the VAX Cluster after that.

      Trent Poly had four CAD stations hooked up to a VAX 11/750, they were quite large booths, with integrated screens, keyboard, and a joystick. Can't recall the name of the CAD package now though.

      1. Tony Gathercole ...

        CAD on VAX-11/750 at Trent Poly in 1986

        DOGS (Design Office Graphics System) from PAFEC.

        I oversaw commissioning of the system circa 1983; previously Trent ran a similar system on a PDP-11/40 under RSM-11M.

        Personally had moved on from Trent Poly by 1986 to ICI who ran multiple PDP-10s, mainly under TOPS-20 as DECSYSTEM-20s.

        1. GruntyMcPugh Silver badge

          Re: CAD on VAX-11/750 at Trent Poly in 1986

          DOGS! Boom! Thanks for that! This has brought back a cascade of memories,... was the 11/750 called Lace? All the machines had Nottingham related names, the Dec 20 was called 'Clough' after Brian Clough, and the VAX nodes were Robin, Marion, John and Tuck iirc.

          Hey, just sent you an invite via Linkedin, turns out we have people in common.

          1. asjackson

            Re: CAD on VAX-11/750 at Trent Poly in 1986

            Old thread which I just came across. I'm at Nottingham in CS and our student/staff Linux machines still have the same naming scheme :)

      2. mikale

        Re: DECs funny numbering sytem

        Was the CAD system made by Intergraph? Maybe AutoTrol?

    4. /dev/null

      Re: DECs funny numbering sytem

      DEC's PDP models were numbered in chronological order, with no indication of the different architectural families. So the PDP-1, 4, 7, 9 and 15 were 18-bit machines, the PDP-5, 8, 12 and 14 were 12-bit, and the PDP-3, 6 and 10 were 36-bit. The PDP-11 was something of an oddity being 16-bit.

      1. Rich.Alderson

        Re: DECs funny numbering sytem

        While it's true that the PDP-3 and the PDP-6/PDP-10 were 36-bit systems, they were not related to each other architecturally beyond that. The PDP-3 (never built by DEC, but one was built by a customer who bought the plans) was fundamentally a 36-bit extension of the 18-bit PDP-1; the PDP-6 and its follow-on the PDP-10 were very different from the PDP-1.

        For that matter, the PDP-1 differed from the PDP-4/7/9/15 family. It had a 6-bit instruction field, allowing 64 different instructions, while the other 18-bit systems had only a 5-bit instruction field for a total of 32 different instructions. Of course, the PDP-1 was limited to 12 bits of address (4096 18-bit words of memory), where the others could address 8192 words (13 bits) directly, and later members of the family had index registers allowing up to 128K (131,072) words of memory to be addressed.

  2. imanidiot Silver badge


    "Even to someone too young to be sentimental about the PDP-10, ITS isn't completely useless: it includes TCP support, a Telnet and Supdup (RFC 734) server and clients; as well as FTP and SMTP mail support. ®"

    So it's utterly and pointlessly useless?

    1. Ken Hagan Gold badge

      Re: So....

      Surely its uselessness is the point?

      Related and perhaps more useful projects would be: can you replicate the same functionality on the same hardware more efficiently by exploiting software techniques that we've learned (or perfected) since then, or can you build a better hardware platform using the same technology and the same budget. Such efforts might allow you to assess the achievements of the previous generations and pay an appropriate level of respect.

      Building *exactly* what they built is merely an act of homage, perhaps the sort of thing you'd do *after* you'd tried the other things and decided that "Yep, they knew their shit. I'm impressed.".

    2. P. Lee

      Re: So....

      >So it's utterly and pointlessly useless?

      A bastion smtp host?

    3. kkt

      Re: So....

      Not useless at all. So much of what we use computers for now is based on those concepts. Mail, moving files around, communicating across the net to do things on an architecture-neutral and manufacturer-neutral network. Transfers fast enough to be useful in minutes, not a nightly batch job.

  3. Dan 55 Silver badge

    It's emulators all the way down

    You run an emulator on a Unix system to run PDP-10, so you can run the PDP-11 emulator and run Unix.

  4. smithwr101

    PDP-10, PDP-11

    The Blinkenbone project has a number of simulated graphical panels including the PDP-11/70 and PDP-10 KL10.

    They use simh and add the flashing lights.

    1. deshepherd

      Re: PDP-10, PDP-11

      Went to the Computer History Museum in Mountain View in October and opposite the PDP-11 exhibit I saw there was a Data General Nova and I suffered a wave of nostalgia (first computer I used - we had one at school) ... had to resist the strong temptation to play with the front panel switches - at one time I knew the bootstrap code form memory ... though the full trip down memory lane would need several shelves of fanfold paper tape needed to for a OS reintsall onto the amazing large (?5MB) Diablo disk drive

      1. Doctor Syntax Silver badge

        Re: PDP-10, PDP-11

        "we had one at school"

        Stop it! We had one running the EDS analyser on the SEM.

        1. Version 1.0 Silver badge

          Re: PDP-10, PDP-11

          I still have a PDP-11/23+ running in my office, boots RSX11-M off a pair of RL02s (10Mb removable disk drives). RSX11 was a neat operating system and had DEC been willing to release it to the general public, it would have kicked MSDOS out of the ballpark.

          1. dbannon

            Re: PDP-10, PDP-11

            "I still have a PDP-11/23+ ......RSX11 was a neat operating system and had DEC been willing to release it to the general public, it would have kicked MSDOS out of the ballpark."

            No, instead they made the DEC Rainbow. From memory, floppy drives that were incompatible with the rest of the world, other strange stuff.

            Sigh, anyone from back then can produce such a list of what DEC should have done differently. If only they had listened (back then) to what we say now .......

  5. Conrad Longmore

    Ah.. Multics

    Still have my "Multics Commands and Active Functions" manual somewhere..

  6. Richard Tobin


    In 1992, Unix System Laboratories (USL) sued Berkeley Software Design Inc for, among other things, violating USL's alleged trademark on the word "UNIX" in their phone number, 1-800-ITS-UNIX. Disappointingly, MIT didn't sue them for using "ITS".

  7. Dave 15

    pdp11 emulator... I am sure in the back of my garage....

    Certainly there used to be, complete with toggles on the front panel for putting in the bootstrap and a separate hard disk... slightly large by modern standards for its somewhat small capacity

  8. ForthIsNotDead



    I just don't know why, or what I'd use it for. I know I want one though!

  9. JustNiz

    Pleas remind me which PDP-11 operating system(s) used pip?

    as in:

    pip /li



    1. Steve Aubrey

      Believe pip (Peripheral Interchange Program) was in RSX. RSTS had a different animal (all from pre-1982 memories, so grain-of-salt time).

      1. JustNiz

        Thanks Steve,

        I used a PDP-11/75 for my first real programming job. which actually would have been around 1982.

      2. pkoning

        PIP existed all over the place. TOPS-10, DOS-11, RT-11, RSTS-11 and RSTS/E, RSX, IAS, the list goes on and on.

      3. Tcat

        yup. And C/PM as well. The first time I tried it, it took so long, I thought the computer locked up.

    2. PhilBuk

      Don't forget RT-11 and there was one that started with OS (OS-11?), similar to OS-8 (which also used PIP).


    3. PghMike

      There may have been others, too, but I definitely ran pip on TOPS-10 systems when I was in high school in the early '70s.

  10. The_Idiot

    Ah, me...

    ... memories of lying on the floor each morning at the Tech College I worked in, flipping lines of bat switches in sequence to boot up the PDP-8. Bugger, I'm getting old...

  11. Anonymous Coward

    How should the history of computing be recorded? Discuss.

    It's incredible to me that humankind's most significant invention since the airplane is also it's least documented and has led to everything it created being similarly at risk. Historians of AD2100 will realize there is a giant void starting in around 1990 where no historical records survive. Historians of AD3100 will have long debates about whether a 3.5" floppy disk is really a type of weapon or a digging implement. If you don't believe go to your nearest archaeological museum and marvel at how little we know about societies that did not use permanent written records.

    1. Pirate Dave Silver badge

      Re: How should the history of computing be recorded? Discuss.

      " Historians of AD2100 will realize there is a giant void starting in around 1990 where no historical records survive. "

      When they figure out how to read the last extant copy of the Internet Archives (on a stack of DVDs, found in a janitor's closet at the Field Museum in Chicago) , they will quickly understand why no written history was kept. Porn, cats, and the fall of Western civilization, that's all there was.

      1. Captain DaFt

        Re: How should the history of computing be recorded? Discuss.

        And then when the lawyers show up From Disney Galactic to sue them, the janitor, the Field Museum in Chicago, The company that recreated the DVD reader to read the DVDs, and the company that made the boxes they were stored in for $INCREDIBLY_VAST_SUM for copyright infringement, intellectual theft, and DRM breaking, then they will realize why there is a giant void starting in around 1990 where no historical records survive

  12. Pirate Dave Silver badge

    If nothing else...

    that diagram of Arpanet from 1977 is worthy of printing and hanging on the wall in my office with the caption:

    "This used to be the ENTIRE INTERNET."

    1. DNTP

      Re: If nothing else...

      They display incomprehensible technical printouts like that as public artwork in Boston and Cambridge, every few blocks.

      Oh wait that's the MBTA bus network map and it is actually worse.

  13. Oengus

    PDP-11 Emulator

    Yeah... Now I can run Underworld again... I have the source listing somewhere...

    Also, I can run the control program for a Nuclear power plant. I remember a couple of years ago a company looking for PDP-11 Assembly Language programmers to work on control systems for Nuclear power plants.

  14. Stuart Dole

    PDP-8, PDP-11...

    Ah - memories. Booting OS/8 off the toggle switches and paper tape, then an actual hard drive! This was my first taste of assembler programming. I wrote a real time process scheduler, and integrated it with an FFT routine for brain wave analysis in real time. Then we graduated to 16 bits and RT-11. Our local DEC salesman would slip us the source code tape for RT-11 to help with writing device drivers for lab equipment - what a gem! Then bringing up TSX - a multi-user RT-11 emulator and running lots of Teco terminals and somehow we got a version of nroff/troff running under RT for creating and printing documents.

    Then RSX-11M, which came with the source code. I remember reading through it and thinking - here are the tracks of a mighty elephant. Indeed, David Cutler went on to create VMS and then NT. But I think RSX-11M was the high point. On a smallish PDP-11 we ran a dozen video terminals (ADM-3a's, I think) for document processing, as well as millisecond data collection in the lab on dozens of A/D channels, all without a hiccough. A friend who was running an early Unix on an 11/70 (a big machine) was losing time because it fell behind on the 60Hz clock ticks - it turned off interrupts every time it had to think about memory mapping, which was often and long.. But Unix was a lot better for getting a program up and running - it had this strange language called "C". And yeah - these were megaherz machines.

    There were a few PDP-10s around, but I never did much with them. 36-bit words?

  15. PghMike

    weirdest UI ever

    One thing you should know, if you'd never used it, was that the ITS "shell" was a binary debugger.

    You'd login by typing "mlk$u" where "mlk" was your user name, and "$" was the Escape character. No password was required (or even allowed; thanks RMS). You could then type something like "4/" and see what's in location 4 (which was a register on the PDP-10). There were ways of running commands like macsyma, too (":macsyma", i.e. put a ":" in front of the command line).

    Even stranger, you could type $$^R, and it would unprotect the OS, so that you could patch the running kernel from your shell.

    I am not making any of this up.

    1. kkt

      Re: weirdest UI ever

      That's not even the strange part. The strange part is that ANYnoe on the net could telnet in and make an account, and once they were there nothing at all stopped them from reading, changing, or deleting any file on the system. "Security" was a "do you really wanna do that?" message. The strange part is that the environment worked, hardly ever did people abuse this trust.

      A different and better time.

  16. Nameless Faceless Computer User

    I used to dial into the PDP-10 using an ASR-33 and a modem which had a pair of suction cups.

  17. Rich.Alderson

    PDP-11 emulation on ITS?

    I suspect that there's a misunderstanding in the report. SimH is a family of emulators, among which are a single model of the PDP-10 (the KS-10 processor), and the entire family of PDP-11 systems (from the original bare bones PDP-11/20 to the final PDP-11/84 and PDP-11/94, with all the stops in between).

    There are cross-system tools for writing PDP-11 software under ITS, but they're intended for running on "real" -11s.

    On the other hand, I might have missed something, since my particular interests in ITS are ITS itself, EMACS and Lisp, and I haven't explored far beyond those.

    1. diodesign (Written by Reg staff) Silver badge

      Re: PDP-11 emulation on ITS?

      No, there really is a PDP-11 emulator in the ITS source code.


  18. Zzznorch

    PDP-11 and VAX-11/780

    I had a chance to use a PDP-10 back in the spring of 1977 when I was in 9th grade but that was only because a friend let me play a computer game on the ASR-33 teletype. My first programming course was in the fall of 1977, when I was in 10th grade, and we used a timeshared PDP-11 running RSTS/E through a 300 baud modem connected to an LA36 DecWriter II. That is the nostalgic system for me as well as VAX/VMS that I worked on in the 1980's at my first and second jobs. Occasionally I fire up the SiMH software to play a little with either RSTS/E or VMS. I was much more involved with VMS and even wrote several kernel mode programs having been tutored by one of the then guru's on the internals of that operating system.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon