This won’t mean much to most of you, but for us geeky old farts who grew up with the DEC PDP-8 and PDP-11 minicomputers, this could bring back memories of our joyous formative years!
There’s this ingenious nut in Switzerland named Oscar Vermeulen (I use the term “nut” affectionately, to mean a devoted super-geek) who has been building kits that let you assemble the physical console of either a PDP-8/I or PDP-11/70. The PDP-8/I was circa 1969 and was one of the most popular models of the low-end 12-bit PDP-8 series. The 11/70 was circa 1975 and was the high end of the PDP-11 series, almost a mainframe.
The idea is ingenious. The computing power behind the panels and switches is provided by a Raspberry Pi running the classic open-source emulator SimH, modified with special drivers for the lights and switches. Oscar is so exacting in producing the replica that there is even special timing to make the lights on the PDP-8 console dim and brighten like the actual incandescent mini-bulbs used in the PDP-8, instead of the typical LED instant-on/instant-off.
Also very cool is that these are 60% of the original size, making them large enough to be practical but small enough to keep on a bookshelf. The PDP-11 is definitely the more flashy-looking console, but the PDP-8 was really special in its elegant simplicity.
I’d be tempted to get on the wait list for the next batch myself, but I’m already up to my eyeballs in home projects. (I mentioned in the MMP thread that I specialize in sloth, which is true – it doesn’t mean that I don’t have home projects, it just means that they move with agonizing slowness! )
The 11/70 kit is sold out, alas.
We had one when I started with Western Electric research in 1980, but hidden in the computer center, so not much nostalgia for me. Now if he made a PDP 11/20 which I used in grad school we’d be talking. Especially if there was a way of reading the DEC tapes I still have.
I never used a PDP8, but, speaking of kits, a guy I met at a conference in 1975 had one in a suitcase.
I never used an 11/20, but a good friend at the time had one in his department and was the main user and programmer. At that time I was still working with PDP-8s and later with the PDP-11/40, whose console was very similar to that of the 11/70.
Unfortunately the only way you’d be able to use those DECtapes would be to use a real PDP-11 to image them to a file that SimH could access. Which is not so far-fetched. I’m running SimH on my PC, emulating a PDP-10. I really wanted to get the famous MacHack chess program for it, but couldn’t find it anywhere as a standalone download. But I eventually located a DECUS (DEC User’s Society) magtape distribution in PDP-10 Backup format that had been imaged as a file, that had MacHack on it. I “mounted” the file as a magnetic tape, and was able to extract the MacHack image using the PDP-10 Backup program.
A college friend has a business providing PDP-11 emulators that run on PC-class hardware. According to him, there are a surprising number of organizations that still need to run PDP-11s or the equivalent.
His value-added must be in consulting and support. SimH is open-source and free and is a very versatile emulator. I must say I’m surprised that PDP-11 applications are still in demand. It was a very popular machine in both academia and business, but it was a hell of a long time ago! The funny thing is that today any simulator running on a typical PC is going to be orders of magnitude faster than any physical PDP-11, or even any physical PDP-10 mainframe. I’m sure even one of the newer Raspberry Pi’s is far faster than an original 11/70.
OMG, blast from the past. I used to remember the bootup address by heart. In fact, I went to some kind of science museum with a bunch of my colleagues from Fermilab and all the computer exhibits were being run by PDP-11. Fortunately, the PCP-11 was mounted about 20 feet up, otherwise any one of us could have walked over and rebooted all the exhibits!
I remember having to program using “overlays” because the memory wasn’t enough to load the whole program at once. (We used it to collect high energy particle physics data.)
We used Teco and a line printer to program that sucker. I may still have a PDP hard disc in my garage…
Wow, I really do not want one. Not even to play with the switches!
I don’t know why you say that, but then, I had a special relationship with RSX-11M specifically which might have made me overlook its command language quirks. For me, much as the PDP-8 was the first and last computer that I truly and thoroughly understood every aspect of, RSX-11M was the last OS for which I ever had – or ever needed – a thorough understanding of the internals.* To me, UNIX is the ultimate in cryptic commands. The command language of the RSX-11 family was simply a more-or-less bog-standard implemention of DCL – the Digital Command Language – which was common to many of DEC’s OS’s across multiple platforms. Can you cite an example of something you think was cryptic?
* - Which also led me to a great admiration of the craft of Dave Cutler, its principal architect, who was later hired away by Bill Gates and became the architect of Windows NT, which became the kernel of all Windows systems from XP onwards.
Our subjective experiences vary. For me, the PDP-8, PDP-11, and also the PDP-10 and PDP-12, represent perhaps the best days of my youthful life – certainly the most fun times!
My first computer operator job was on the night shift in a furniture warehouse running a PDP-11/? (something.) I wanna say 11/05, but I’m not sure.
Went from there to IBM mainframes and programming.
Nope. Don’t miss it.
I’m used to seeing PDP-11’s in a 6’ tall rack with other components. Lots of them at NASA. When I bought my house in 1983 the previous owner was running some kind of consulting business from the back bedroom. It had several racks of PDP-11s.
I’ve read a bit about PDP-8’s back in the day although never used one directly; I did a bit of programming with a lower-end PDP-11 in the early 1980s and a PDP-10 in the early 1970s.
The PDP-10 had an instruction set that was . . . unique. One friend remarked that some of the mnemonics looked like something a startled Martian would say.
Circa 1972, Ralph Merkle (a co-inventor of public key cryptography) devised an algorithm for successive generations of John Horton Conway’s Game of Life that was very efficient. It operated on a bit-mapped board, updating an entire row at a time with a sequence of logical and Boolean operations. (I still have a printed listing of it, and I might even still have the original hand-written algorithm that he gave me.)
I coded that for a PDP-10 using “hextuple precision” – that is, 216 x 216 board, with each row stored as 6 36-bit words. @wolfpup , you could probably still run this on your PDP-10 sim. (I also did a version for the CDC-6400 on a 60x60 board.)
The machine I had access to had a CRT display that I figured out how to use, so it displayed the running generations on that. The program also had a single-step mode to advance and display one generation at a time. It also had a section written in FORTRAN to read an initial configuration from a text file, that could either be just a file of dots and spaces, or a variety of compressed formats that enabled elaborate configurations to be entered very easily.
This machine had a version of TECO that could display an entire screen-full of text with one command. Based on that, there was a macro package that turned TECO into a screen-oriented editor, which was about as diametrically opposite of basic TECO as you could imagine. I did a lot of enhancements to that package.
Young-uns today who never knew TECO will never know what they’ve missed.
The command language of the PDP-10 was also rather distinctive. It had the interesting feature that it contained a built-in primitive MAKE facility for compiling, linking, and running programs in the various languages that were available.
AFAIK, it was these early DEC PDP machines that first gave us the standard filename+extension syntax, in which the system itself used the extensions to determine which compiler or other program should be used with which files.
I still have a command-line version of TECO on my PC because it can do things that no other text editor can do, in terms of programmatically controlled modifications. TECO was nothing less than a programming language that was oriented to text manipulation. It could do anything.
I once wrote a version of RUNOFF (a text formatting program that produced chaptered and section-titled and justified text) entirely as a large TECO macro, just because TECO was available for the PDP-8 (under OS/8) but at the time RUNOFF was not.
My first experience with programming was largely on big iron, IBMs, CDCs, and a bit on a Cray-1. That was before I got to college, where most of my experience was on a PDP 20, running what we insisted on calling Twenex, though DEC didn’t much like that. But it had TECO and what I think was then a relatively new set of editor macros built on top of TECO, which went by the name EMACS.
Down in the basement the CS department still had a few PDP 11s they used, but I didn’t do much with them myself.
I worked at a lab in the mid-1970s that had their own home-grown editor that could do that. The underlying program wasn’t even an editor, it was just an app for manipulating text. Based on that, there was an entire quite elaborate text editor, and also a RUNOFF-like text formatter. The underlying language operated on arrays of character strings, and had a FORTH-like syntax. This was also accessible at the editor and formatter source-document levels, so end-users could write their own macros for their documents.
troff, the Unix text formatter, allowed for elaborate macro programming too.
Many Unix users were fond of emacs, another text editor with extensive programming abilities. But good luck fitting any of these elaborate apps on your typical PDP-8!
BTW, the computer that ah Clem hacks in Firesign Theatre’s “I Think We’re All Bozos on this Bus” ran PDP-10 command language. “Five jobs, two detached.”
Did this machine have a ROM boot? Our research machine did, the Ed-11 (the one we taught assembler on) did not, and you had to enter the first few lines of the bootstrap routine from the front panel.
If I could get my DECtapes read, and get the emulator, I could have my PC emulate the PDP-11 simulating the Lockheed SUE minicomputer running microcode emulating the LGP-21. 60 years of computing history in one run.
And it would probably run faster than the LGP21 did. Running it on the PDP-11 was faster, and that was 45 years ago.
)