I'm sorry, but all I could think of was the products of the Sirius Cybernetics Corporation in The Hitchhiker's Guide To The Galaxy by Douglas Adams: "It is very easy to be blinded to the essential uselessness of their products by the sense of achievement you get from getting them to work at all. In other words-and this is the rock solid principle on which the whole of the Corporation's Galaxy-wide success is founded-their fundamental design flaws are completely hidden by their superficial design flaws." I wonder if you edited out the mandatory goat sacrifice to comply with RUclips regulations! Thank you for this. It was very enjoyable. It makes me have a great deal of sympathy for the people who had to do this on a regular basis.
It was normal when loading/running a test program in the hp1000 to jump to a specific address. Booting RTE/1000 did not require this step as the “disc” boot loader rom jumped to the correct location. I haven’t seen a hp1000 boot sequence in 32 years. I could do that with my eyes closed in the 1980’s. Well almost. I wrote a hp1000 boot loader that loaded a disc test program over HPIB from a hp1000 running RTE (real time executive). That 21 year old programmer was really proud of himself. Burned the boot loader rom as well. The loader used serial HPIB interrupts which required responding to polling by the host machine to determine who interrupted using a shared line. I could do almost anything with the hp1000. It used microcoding for the instruction set so I started playing with that as well. There wasn’t much I didn’t do with that computer. Those were fun times. We had several hp1000s running several chemical processing systems in a printed circuit fab shop in Boise. They were located in a computer room. On the wall next to the hp1000s was an emergency power shut off switch. A big red button without a cover. Late one night I got a call at home after a janitor bumped the switch with a broom handle instantly shutting down several processes and costing thousands of dollars in scrapped product. Needless to say a cover was quickly installed. Great memories. I did 10 years of programming on the hp1000. Pascal was my favorite language but I did a lot of FORTRAN 77 and assembly programming. One hp1000 controlled 24 hp6464c DC power supplies which were monsters used for electroplating. You should try to get one of those supplies for some high power projects.
My father was a tool & die machinist & maintenance guy who eventually got into installing & upgrading NC machine controls. In the mid-80s, I was in college & got to work with him on a project emulating a paper-tape reader & other peripherals with a DOS PC, interfacing with Texas Instruments controls. The full-featured control had 12 K of DRAM, and they needed to run G-code programs in the tens of megabytes, so the PC had to spool data to the control and react to control signals. Both serial & parallel interfaces were supported. As the code monkey, I had a control unit in my apartment for development & testing. It was the size of a small refrigerator, sounded like a modern server rack and had a toggle-switch interface on the front panel. Booting the monster involved unplugging everything else on that circuit, latching the bootloader (four 16-bit words) into memory and executing them to load the machine executive from paper tape. We had to breadboard enough components to fool the machine into thinking all of the limit switches, pumps, etc. were connected so the executive would think it was in a real machine tool. Perhaps not the typical father/son project, but watching this video brought back a lot of fond memories for me. Thanks for posting it!
This reminds me of the long routine getting the Data General Nove 1220 up & running at college, but the start-up procedure only involved flipping the appropriate toggle switches to set various registers, etc., through buttons on the disk pack drive(s) to actually telling the machine to "boot up." We didn't have to start up with paper tape because previous students had RTFM'd and loaded everything needed onto locations on the disk pack(s) and most importantly, documented everything required to start, run, and shut down the system. My Machine Design Professor had it tucked away in a "closet" in the (Mechanical) Lab building and he allowed some of us to use it for our "Senior Projects." Having access to an entire minicomputer with only one or two of us sharing it sure beat going to the Computer Center which had two (mini?-)computers (one upstairs, one downstairs) shared by nearly 100 students/terminals. One of those machines was an HP, but I don't recall which model it was. This was between 1979 & 1982, so the HP may have been an HP 1000-something like Jack has. But once I got access to a virtually _personal_ minicomputer, I never went back to the Computer Center! ;)
Maybe you can remember enough RE the data general to help the guy over at the tech tangents RUclips channel, to get up an running when his switch on occurs.
@@beingatliberty I watched his latest video and commented recommending a DG Guru to contact for possible PDFs of schematics & other docs that he doesn't already have. All I (think I) remember is turning the NOVA 1220 on first, toggling in appropriate data/commands, _then_ starting or maybe "loading" the platters, waiting for the appropriate "clunk" (the light was burned out), and finally (I think) hitting "run." I did subscribe to his channel to see what happens next. ;)
When I used Data General Eclipse computers, we sometimes had to enter what my co-workers called the "Hip Boot". We had a 20 Mb disk drive (required a forklift to move) and the Hip Boot would boot from that. Most of the time we didn't have to do that because boot PROMS were installed. The DG Nova computers we used for manufacturing test all had core memory. Test programs were loaded from a paper tape reader.
I'm so happy I finally get an idea of how you would operate these machines. Grateful for all you guys who have the knowledge and share the love for these historic pioneering machines. Without them I would not have been able to setup my PiZero2W as a Klipper control computer for my 3D printer tonight ;-). I love these videos, thank you Marc!
This brings back such pleasant memories of my first encounters with computers, about 1976 - 1978 or so, at Diablo Valley Collage in Pleasant Hill, California. I weaseled my way into a job as a computer lab assistant, so that I could kick everybody out of the lab and play with my own projects without interruptions late at night. We had an HP2000 Time Share BASIC System. There were two HP2100A processors. The main processor had 32K words of core for running the user's BASIC programs and most of the OS, with 2-second time slices in a sort of cooperative multi-tasking kind of thing. The I/O processor was nearly the same machine, but with 8K words of core, and the job of handling serial I/O to 32 user ASCII terminals. The system had the same disk drive, paper tape deck, and mag tape deck as Jack's system. It also used some sort of core drum memory device; a rotating drum that was chosen because it could help read and write program swap space on a regular periodic basis. I never knew much about it; it lived behind that smoky plastic and just did its job. We just called it "the drum." When I think of the machine, the first thing that comes to mind is always the buttons. The 1/2" (12mm) buttons on the front panel of the 2100B were just delicious. Big enough to feel right, and with a long, smooth, clickless travel so that you could make a clear statement that you wanted this one pushed. They were all binary, and lighted -- light turns on, light turns off. Entering an octal word was similar to what Jack showed. The trick for entering the long boot loader was getting into a frame of mind where you would see the previous word, the next word you have to enter, do an exclusive-or in your head, and push just those buttons that needed to be toggled. I loved those buttons. Every night... the lab assistant would send everyone home, put the appropriate reel of mag tape on the drive, and make a fairly fast incremental backup of the student's work stored on the hard drive. But every Friday night... a full backup to tape of the entire disk and operating system, which took maybe an hour or so. Then... key in the 64 words of boot loader through the front panel, in octal, tuck the three-inch roll of paper tape containing the operating system into that little pocket you can see to the left of the read head, and push the button. The tape reader was super fast, relatively speaking -- optical, so there were no mechanical issues to get in the way of blinding speed. So the paper tape would fly off the roll, through the read head, fly out into the air, and drop into a big snaky pile on the floor. What you did not want to do at that point was stomp around with your big shoes and smash the tape, because... I'm not sure if we had a second copy! So now with the operating system loaded into core, as far as the system knows, it's a brand new machine, fresh off the showroom floor. But the full backup is still in the tape drive, and we enter a few more commands on the teletype (now that the system is up!) to restore everything from that tape. As the student's work gets read back in, the disk drive gets repacked in a nice and orderly manner, eliminating fragmentation. When that's done, everything's running nice and smooth, ready for classes on Saturday or Monday morning. So there's miles of paper tape on the floor. Leave it to HP's ingenuity. On a little bracket mounted at the side of the paper tape reader, there was... I've never seen anything like it before or since. Think of a stylized electric knife handle, or something like that, sort of an olive green plastic handle-case sort of thing, containing NiCad batteries and a motor and a button. On the motor's shaft sticking out the end of the thing was a plate about three inches in diameter, with two pins sticking out near the center. You would fold the start of the tape around one of the pins, hold your free hand just right to keep the tape going where it was supposed to, and push the button. The miles of tape on the floor would slurp up into a tidy roll. No "spool" required -- the roll was self-supporting; just a roll. No complex mechanisms, just some creative "Why not?" engineering. It took probably some fraction of a minute to wind up the whole roll, if I remember right. The word "fun" applies! We also had an IBM System/3 -- not the new one, the old one, big as a car, 8 bits, 32k bytes of RAM, an IBM Selectric console, a 1402 card reader, a 1403 printer, and what I consider to be a *very* cool instruction set with variable word lengths. You could move or add or subtract integers or packed decimal numbers up to 256 bytes long. Brag about your 64 bit integers... these were 2048 bit integers! Loads of fun.
Oh my - I learned HP Basic on a time share HP1000 system that was run by the school system in Oregon back in the 70s & 80s. I visited the computer center a few times and saw the machine - looked very much like that one. Good times.
Yes, I think the system was called the HP 2000 TSB (for Time Shared Basic), and it consisted of either two HP 2100, or the later versions had two HP 1000. One was the main basic interpreter and the other one took care of the communications with the many user consoles. Some early versions apparently had only one machine. Oh apparently the first version ran one on one HP 2116 with 16 users (see www.decodesystems.com/hp2000/#photos ). You must have run out of memory after programming just a few lines. I’ll have to try that one day. They were popular in schools.
You're definitely talking about the HP 2000s that most school systems in Oregon had at the time. Our system was shoved in a small narrow room between a classroom and a stairwell. The room was so small that two people could barely fit in there at the same time. We student hackers stole passwords by tapping a Radio Shack Model 100 to the phone line that was connected to the modem in the front office and waiting for someone to log in. It was important to turn capture on because the passwords were always control characters. We never got caught but I later found out that everyone knew about it.
Hah - fun bit of hacking there. IIRC the computer system was called OTIS and the computer center was a nondescript concrete building on Hwy 99 just on the northern outskirts of Eugene. I went there a few times with some fellow students and looked around and chatted with the guy running the show named Ben Jones. They also had some large IBM systems that were used for record keeping throughout the school system. I started using the HP Basic systems in 9th grade on an old Model 33 teletype that was jammed into a tiny soundproof cubicle in the back of a math classroom. In HS we had a dedicated computer room with two teletypes and an old Hollerith keypunch. Before I graduated those had been upgraded to video terminals and we also had some TRS-80 and Commodore PET machines. Somewhere along the way the HP BASIC at OTIS was upgraded and included interprocess communication statements - chaos ensued as students throughout the state were able to chat with each other. That got shut down after a while because the suits thought it was too disruptive. What a bunch of hooligans we were.
@@ebrombaugh There was also a Oregon state network that connected university computer systems. In Portland I was able to connect to the DEC System-10 at U of O through a local phone number which I spent hours on for free. Looking back at HP TSB, I wonder how it could possibly work reliably. There was no locking in their BASIC (that I saw) so writing to files on a multiuser system had to create unfixable bugs unless the programs used lock files. Someone could read old data from a file and then write over someone else's new data.
@@scottlarson1548 One of the other students at our school pioneered a BBS/chat system on the HP using shared files with some careful file locking implemented by hand. Kids from a handful of schools were able to post messages and chat in real-time with his software. The rest of us studied his approach and there were many variations on it being coded in our school. It was the wild west and we were learning a lot about these things.
I'm in the middle of what I thought were tedious computer setup steps. After watching this I gave my computer a little pat and said "Good computer, you're not so bad after all". Thanks for the video.
@@wishusknight3009 I like to collect and play with some proper esoteric systems. I have a couple(!) late model PDP-11's, and they're definitely a different beast to modern machines. Whilst i haven't got anything of this ilk, basically any sort of late 60's/early 70's minicomputer (and ealy toggle switch micros like the Altair 8800) are basically my white whale in computing. It's a whole different world unto itself
@@the123king Amiga's are not so easy now if one is on a budget. I love amiga's but for now I only use one on Mister. I sold the one i had a few years ago. Though I also like early motorola macs and PC's up to the 386. And have a few of the more popular types. The esotaric stuff I have is an everexx deskside workstation from the mid 80s that was originally a 286 and was upgraded to a 386 sx at some point. Really interesting machine,. Has a very elaborate accelerated graphics system and is all on a passive ISA backplane. Also has some of the earliest large scsi drives. I also have a few early laptopts and some of the more wacky early prebuilts. An x68000, An FMTowns that we have had since new... and a Colico Adam in perfect shape with a boatload of stuff with it, (though I dont really know what to do with that one tbh... No one wants them.) . A Compaq Quad xeon 400 server, an (presumably) Intel branded quad PPro server. And of course a lot of generic stuff that most all collectors have. Nothing too exotic really, but its a varied selection with some interesting bits. And quite a bit of console stuff., though nothing particularly special nor rare.
This is so hilaric. As a freelancer I entered a video editing room in 1988 on a sunday morning 9 a.m. to edit this tennis report for the evening sunday night sports programme. Most equipment I recognized but the character generator (for the titles) was new for me. As there was no manual, I had to call a person who was familiar with it (at sunday morning at 9). To get this machine alive I had to made the most ridiculous keypad combinations. After half an hour of instructions and trials it went working.
Hey great to see a live E-Series after all these years. I was a field engineer who used to service and repair the HP1000 M, E, and F series (as well as L and A series) in the 1980s to 2000s. It jogged a few dormant brain cells seeing you trying to boot from the 7970 (a long winded pain in the ass to head align). Seeing the E-series showing HLT77 (happy halt) made me smile. Never worked on a 7900 hardly, it was mainly 7905/06/20/25 storage on the MEF range, which I also spent many a day repairing and PMing. Great video. Thanks.
First job out of university was at HP Cupertino working in the DSD Lab on RTE for this machine. We had the entire HP1000 hardware and software team occupying one floor of 42U, Cupertino Integrated Circuit Operation (CICO) Fab (remember SOS?) below us in 42L. We could walk next door to the HP1000 (M/E/F) manufacturing line and QA in Building 41(or 40?). I learned so much in that job. Our team that did RTE-6 had just 6-7 people. Best job in my career, great team, great times.
Just dropping a comment to say thank you for documenting these vintage computer adventures you have, they are always facinating and fun to watch. Being 34 I never experienced this equipment myself, I had an uncle who worked on this and other stuff like it and would vividly describe it when I was a kid, so getting to see some of it running and doing it's thing, is really cool stuff Marc. Like I said thanks for documenting and sharing this.
Love this stuff. In the early 80s I programmed a HP 21MX in a large automated test system (4 or 5 full racks of instruments). Most of the instruments were controlled by HP-IB (IEE 488). Many of the vintage HP instruments in your videos were in the system. This was a fun and nostalgic video.
We had a 2116 in our repair depot in 1976. Plugging PCBs into the rack took a lot of force. My company, Measurex, still had 2116s being used at customer sites. One day I was given floating point ALU boards to test, giving an already out of date computer a new lease on life. We had lots of 2100 computers. Those ones I could lift.
@@CuriousMarc No, the name doesn't ring a bell. I was situated in Québec, and it was the late '70s. I liked playing around with HP 2100 computers then. I wish I still had the handbook for it. BTW, I have been watching your AGC videos. It appears to me the AGC was a bit-slice design and its machine code was microcode. Years later Lisp Machines came out of MIT. They also used a bit-slice architecture, and were stack machines; narrowing the gap between microcode and high level languages even more. I enjoy all your videos. I own a lot of HP test equipment, and like General Radio gear as well. I restored a GR 1710, a network analyzer.
@@CuriousMarc I guess I assumed it was a bit slice architecture after seeing a couple of AGU modules with "SLICE" in their names. Your series of videos on the AGC is very good. Anyone with some technical training and experience can follow what you did. It was a very competent team. Thanks for putting this on utube.
What a different world we have now. That tape drive was pretty cool; lot to set up there. Easy to take what we have for granted compared to this at one time. Beautiful machines and demo!
I'm 33 and work in IT. I remember almost 15 years ago when the company I worked for had to fly a contractor in from the other side of the country because he was the only person who knew how to work on our accounting mainframe. They had used the same system since the early 80's. I remember that it was a complete black box mystery to me what this cabinet was in our data center. I would talk nicely to it so that it didn't break because I knew that no one in our company knew how to fix this thing if it did break. Finally around 2012 we switched over to SQL reporting and again had to hire people who could emulate the features of our old mainframe in SQL and Business Objects etc.
When booting from a disc drive the lower 3 bits set the disc unit number to boot from. The disc controllers supported up to 8 drives per slot/controller (unit 0 to 7).
Brings back memory of 45 years ago of my time at HP repairing those suckers. Question was that a 7970 E or 7970B. Don't remember ever seeing an E on a 1000 system back then.
Nice... I used the same hp setup to read punch tapes to run manufacturing equipment. We modified the teletype board to run faster by swapping out the crystal and changing a few of the capacitors. There was no real take up reel for the punch tape, so it would heap up on the floor as it was loading. We would stand real still as the tape would cover our feet since we had to manually help it at times. Then quickly rewind it back without it getting knotted up. We also had an HP2100 and that same tape drive. The tape drive died. So I had all our tapes read by and external company and emailed. I then re-wrote the bootstrap to read data from the rs232 card and fill memory with what was found. I also did a 1's and 0's check on each memory location first(magnetic core). Wrote it in HP RTL assembly on paper, converted it to binary (octal) and hand enter the code. Worked like a charm.
So fun to see one of these running again. I used an HP1000 in my early days in the Air Force in the late 80s. We used a test system controlled by this computer to test missile, and missile component readiness. Our interaction was pretty basic as I recall, mainly inputting a number into the S-register iirc (111102 maybe??) Store->Run.. something like that, then moving to the terminal for the rest of the test. The program was loaded off a (I think) 14-inch patter drive, quite advanced from the paper drives I suppose.
Marvelous! We had these in ‘76 at the then Westinghouse (now Northrop Grumman) Baltimore plant. They drove the PC board testing equipment. We had a number of these units and they were upgraded over the years
Ahhh... paper tape. Now that brings back fond memories. Much later we had 8" floppies to boot the Bull Level 60 mainframe. Those big beasts were slow... but fascinating. What was frustrating was that it had many, MANY blinkenlights, as a proper computer should, but behind a interlocked panel in the main CPU cabinet. Couldn't open it during normal operation. Sad!
I love your channel, Sir! I find your humor wonderful too. I passed your channel onto someone I know who used to work in the Apollo program. Thank You!
Wow, this brings back memories. Used to look after an industrial control system with two HP 1000’s running rte, linked to ten HP2250’s connected via HPIB extenders, which interfaced to about 1000 different I/O and controlled an offshore drilling rig! If I remember rightly it was 141200 in the S register to boot from the 7912 disc drive, but it’s been a while…. Consoles were the awesome 2645’s. Had 7970 tape drives too, but that was on the HP3000. This channel brings back great memories, and always entertainingly done!
Den? The level of combobulation in Jack's lab surpasses mine by a few orders of magnitude. I'm damn envious of all that nerdy stuff down there. HP 5512A? Now that's a thing Nakazoto would love. Same for me :) Thanks for sharing a story of how HP got into computers - probably the only industry some people know HP for.
Looks like a similar setup I used. I used HP1000 mini in college. Modified an 8008 assembler with extra instruction set for Z80 microprocessor. Punched a few switches to use ascii tape reader to boot it. Once booted, system was on the hard drive with the huge head coil in it. We did not have the mag tape.
My first paying job as a computer repair technician was servicing these HP computer systems back in the 1970's for an HP reseller. The HP 2116 was the workhorse until the 2100 and 21MX became more affordable. Compared to other mini-computers back then, HP equipment was over-built, which is probably why examples like this are still running today. Hideously expensive compared to today's commodity PC hardware. The concept of the CPU halting and being manually controlled seems so foreign but was common place back then. When micro-processors came out they had dynamic RAM memory for their internal registers and thus had to be clocked constantly to "refresh" the data, thus the BIOS boot ROMs on power up and immediate dialog on a terminal. Wow this brings back memories. It would have been nice to see the 2116 in action with all of it's front panel lights. You can actually "single step" a program and watch each instruction manipulate the data in the registers as they are displayed. In fact when a CPU would not boot because of an internal logic, bus or core memory failure, we would have to "toggle in" simple instructions and single step them to see if that would give us a clue to the problem. Such as a dropped or hung data bus or address bus line. Hard to believe these systems supported multiple users, running programs written in Basic, on only 32KB of main memory.
The 2116, 2115, 2114 and 2100 are core memory machines, you load the last 64 words of memory regardless of memory size. Protection was via a switch on the 2116, 2115, 2114 and a push button on the 2100. More fun to load the boot prog manually. 9500 test systems were fun to work on.
Many years ago, as a CS student, I had a summer job at a research lab which had a 2100 and a 21MX. I mostly worked with the 2100, loading their custom OS from paper tape and debugging it from the front panel. One annoyance about the 2100: the front panel used incandescent bulbs, which burned out. I had to do all my debugging with one dead bulb, which for some reason the lab wouldn't replace (I'm guessing they had let the maintenance contract lapse). Fun times!
Our hp2100 constantly lost bulbs. Needed a plastic tube inserted around the bulb to pull them out. After a few times I started to replace them myself. Also had a lot of custom interface boards inside which learned to fix.
What great memories. My first computer in high school (1977) was an HP2000E based on the same hardware. Them my assembly and microcode classes were on HP21xx machines at Chico. Lots of fun.
I learned basic on a HP 21MX time share in 7th grade. A year later, I wrote a false login prompt password stealer and stole the administrator account A000 password and it’s secondary password. A few weeks later I got busted and suspended for the first hack ever in my school district. I would love to see more videos this computer that was stuck in a glas-housed room of my school . Hopefully you can make many more !!!!
A colleague has been looking at our similar dual machine with washing machine disk drives (which do sort of spin up) and a printer but no paper tape. Just booting the thing could make a really good demo.
I worked on HP 21116 based Automatic Test Systems (ATS). You punched in the boot loader through the front panel. This loaded another boot program through the paper tape (we used metal mylar) reader. Then you loaded the operating system from disk. The was HP Test Oriented Disk System (TODS). The main language used was HP Test BASIC. HP Test BASIC contained all the function calls necessary to control HP instruments and scanning devices. Before HP-IB so they used custom HP interface cards. Before HP RTE.
I know I _”shouldn’t be able to”_ but I can hear the whine of that CRT, and I’m 47 years of age. Apparently the average Joe can’t hear such high frequencies unless he’s a child or a teenager, but let me tell you this - I took a hearing test a couple of years ago, and my audible range cuts off just around 18Khz. Btw I have Asperger’s, so that’s a very significant factor. I can also hear my kitchen clock ticking from the next room, around a corner.
It's not an accurate diagnosis for being on the spectrum, for sure, but a surprising number of us have that high frequency sensitivity. Helpful thing to have as a technician.
@@jnelson4765 Indeed! High-pitched sounds are a pain to many aspies but to me it has never been "noise", it's just the sounds that tell if things are working like expected :)
@@jnelson4765 for sure. High frequency _and_ high volume sensitivity. I can hear things going-on outside and among neighbours that seemingly no one else can.
Wow - we take so much for granted about how modern computers are just ready for input and running things after they start up, running a zillion processes automatically. Very informative!
It really tripped me up when you were reading sequences of octal digits (100) as if they were decimal ("one hundred") rather than octal ("sixty-four"). Apart from that, I really enjoyed this video!
That’s how you read octal. Unlike hexadecimal, you can pronounce it like a regular number. It’s just base 8, that’s all.That’s the one and only beauty about it. Except for that, it’s miserable.
Brought back memories of an HP-1000 w/ RTE-IV B used at a chemical plant in Tacoma, WA during the early 1980's. Not sure how much it really helped in improving process efficiency. That plant closed down at the end of 1985. Last time I saw an HP-1000 in actual use was as an adjunct to some chemical laboratory devices in about 1998. Still going strong, but spares were an issue.
When I was a kid I was couple of times at my fathers work, he was technician for those computers in former Czechoslovakia - soviets tried to copy everything from the west, but they were not able to reverse engineer the IC and lot of other stuff, so they were making these vastly big computers taking the lot of space and still breaking. In Slovakia they were able to smuggle first PC computer from West Germany and to reverse engineer, thus first computers born in Slovakia. Limited BASIC but it was incredible.
Fascinating stuff..I always wondered what the procedure was to get these kind of machines up and running. We take so much for granted now. And it's great to still see these amazing complex machines still in use.
I am so glad that I didn't start using computers in the days when you had to poke values in the program counter by hand, in binary, before you could run anything!
While an undergrad EE student in the early 80's, I rescued an old varian 620 which was going to be scrapped. I pushed it into the corner of a lab and proceeded to get it working. It was built entirely from wire wrapped 7400 TTL and many of its instructions were malfunctioning. I managed to get the CPU and disk controller working but one of the three 8K core memory stacks was giving me a lot of grief - so I couldn't get it's "vortex" OS to boot (it would load but, well, there was a big chunk missing). Before I could sort the memory out, the guy responsible for the lab decided to evict me (without notice) and put up a sign saying people could help themselves to my machine for parts. It disintegrated before I even became aware of this plot 😭 I was somewhat annoyed by this...
I find this stuff very rewarding on old computers, I find it extremely frustrating and obtuse when it happens on a modern linux ... cause it means usually I am at that level cause something that has been working fine for years got borked during an update and now its not even an option
For quite long along this piece of history I kept thinking that I had been around the HP 1000 once in my life. Until it finally dawned on me that I had mixed the 1000 and the 3000. But then I also realized my fake familiarity may have been due to similar actions that had to be taken for booting a DEC 11. Never had to do it myself, but saw it happening a few times in the long route and then a second time when a more knowledgeable guy said: "I don't think we have to do it that way - I think we can use a shorter setting, let me try..." And he did, with just half a dozen entries and the DEC woke up just fine. That time at least there were more than one way to "skin the cat". Also, by that time I had some experience with the marvelous HP 9100 and had my own HP 35. Maybe even the HP 67, but never became any sort of computer guru. All my exposure was centered on learning just "enough to be dangerous" about Z80, Z180, MC6800, '6803 and '68000. In the meantime, I got my HP-87, HP-15 and HP-42S as well as the binary/octal/decimal/hex calculator HP-16. The '15 lives on as simulation in my various personal computers, including my Android phone, iPad Mini and MacBook Pro. Oh yes, the very first computer I was "near" was a British Elliot Mark something that ran either Autocode Mark II or some sort machine language. I pal of mine became an expert on the machine code. I barely managed the Autocode to pass my exercise and then promptly forgot it. Its input and output were originally punch cards and then paper tape. I have become ambivalent about either liking or hating some of this in my past.
This brought back memories of using the front-panel switches to boot UNIX/RT on the PDP-11/70. The kernel would read the switches on every clock interrupt so the quickest way to bring the machine down was to flip all the switches up, resulting in a "Panic: Zero."
The 9100 is my Holy Grail - the only machine that beats it is the AGC, and that “belongs in a museum” (like the ones you’ve worked on). Marc, your channel is a joy - please keep up the great work. I’m wondering: do you dabble at all in “mere” HP calculators? That’s the extent of my collecting…I’ve repaired the card readers on an HP67 and in the accessory for the 41C (the extent of my tinkering).
Old school computer room work, load some bits into a memory address, bootstrap bootstrap, the smell of this gear in operation, is more like an electronic medical centre - worthy of a labcoat to operate, some faint ozone generated or something, and halon gas systems to protect the data from fire.
Brings back memories... we had HP 1000s before we replaced them with 3000s. And that was over 30 years ago. 16 bit binary boot programs were a great advance over the 60 bit octal boot programs you had to set with switches for the Control Data machines...
For me the nostalgic part here is the terminal (CIT-220+) that is the same that I got from an auction in about 1991 when my university installed the first "net" in the student houses (dormitories) i.e. RS-232 lines (I think it was 9600 baud) that used the wiring of an old PA-system that was in every room. I do not know what they had used the PA-system for really, perhaps some radio programs because there was a selector switch for many "channels". But now we had access from our student apartments, with the dumb terminal, to the university UNIX servers and therefore email, usenet groups and more :) I still have that terminal in storage and last tested it about two years ago, still working.
At 11:10, the terminal's keyboard came into full view for more than a second and caught my eye: hey, I have exactly the same! Looking closer at the terminal, I recognized a C. Itoh CIT220+. I have a CIT224, obviously equipped with the very same keyboard. It still serves as the console terminal for my server running under Debian.
Some ancient HP test instruments are ridiculously hard to find in a decent condition nowadays. It's nice that you guys restore and repair those rare elusive beasts. It's interesting that some classic HP computer basically had the early form of multi-boot setup, in this case which ROMs to boot the bootloader from.
First 21mx computer was 2105, MX stood for Measurex who asked hp for a smaller computer for process control. For example, paper mills used the Measurex control for paper thickness.
My first job working at Bell Labs, back in the day, was programming HP 2100 and MX computers to do IC layouts back in the late 70s and early 80s. Good times!
Oooh -- the HP 9825! I would get to play with these things a few times at a GF's high school -- great fun these things are! I think the one I used may have been a 9860, but it was a similar form factor.
I have recently managed to get Puppy Slacko7 64bit to increase screen brightness on this old Toshiba laptop from utmost gloom to 100% with a shell script that runs automatically at boot. So I concur with your Linux command comment! Easy when you know how, but the road to rejoicing was somewhat rocky as I am a computing dilettante who refuses to... Better that I end the rant here.
I maintained HP 2113MX computers in the USAF for several years, and , if there is one thing I have memorized, it's RPL states I used on 9 different systems 40 years ago, and thinking to myself, "store, preset, ibl, preset , run!"
I also had to tear down and reassemble several of these 'Rubik's cubes' chassis, as the front and backplanes would loosen up their grip to the mainboard, and, their grips on their assorted interface cards; a quick 2 hour tear down, and judicious use of a small flat heard screw driver to rebend all the little spring loaded connectors in the slots, and maybe the computer would then be good for the next year....or ...maybe not!
I used to see the old HP1000 computer at the University electric engineering dept. gathering dust . After we acquired a brand new HP3000 .This computer was forgotten and never run again
Learned Assembly on this machine using card decks. Was the primary CS weedout course @ Chico State back in the day. I also worked on the 2000 and 3000 series. The 1000 was a bitch to program since it only had 5 usable memory registers of which E was just a one bit overflow.
ya.. for all those that 'interfaced' with a tty terminal with paper tape and telephone modem.. you got to do the telephone modem part.. like with 16 lines going in..
More memories. I didn't like the move from octal to hex, the computers I used had opcodes that seemed to align with octal, but as assemblers and debuggers became more popular, the meaning of the binary opcode was not as important.
Hardest part was generating the operating system. I would start it before I left for the day and it would usually be finished in the morning. Since the disc drives were voice coil, there were a few song programs.
I just missed toggling in a boot loader as our college got rid of it's Data General Nova the year before I started.... I feel I really missed out on living through that period of history.... maybe I'd think differently if I had lived through it. ;)
I'm sorry, but all I could think of was the products of the Sirius Cybernetics Corporation in The Hitchhiker's Guide To The Galaxy by Douglas Adams:
"It is very easy to be blinded to the essential uselessness of their products by the sense of achievement you get from getting them to work at all. In other words-and this is the rock solid principle on which the whole of the Corporation's Galaxy-wide success is founded-their fundamental design flaws are completely hidden by their superficial design flaws."
I wonder if you edited out the mandatory goat sacrifice to comply with RUclips regulations!
Thank you for this. It was very enjoyable. It makes me have a great deal of sympathy for the people who had to do this on a regular basis.
It was normal when loading/running a test program in the hp1000 to jump to a specific address. Booting RTE/1000 did not require this step as the “disc” boot loader rom jumped to the correct location. I haven’t seen a hp1000 boot sequence in 32 years. I could do that with my eyes closed in the 1980’s. Well almost. I wrote a hp1000 boot loader that loaded a disc test program over HPIB from a hp1000 running RTE (real time executive). That 21 year old programmer was really proud of himself. Burned the boot loader rom as well. The loader used serial HPIB interrupts which required responding to polling by the host machine to determine who interrupted using a shared line. I could do almost anything with the hp1000. It used microcoding for the instruction set so I started playing with that as well. There wasn’t much I didn’t do with that computer. Those were fun times. We had several hp1000s running several chemical processing systems in a printed circuit fab shop in Boise. They were located in a computer room. On the wall next to the hp1000s was an emergency power shut off switch. A big red button without a cover. Late one night I got a call at home after a janitor bumped the switch with a broom handle instantly shutting down several processes and costing thousands of dollars in scrapped product. Needless to say a cover was quickly installed. Great memories. I did 10 years of programming on the hp1000. Pascal was my favorite language but I did a lot of FORTRAN 77 and assembly programming. One hp1000 controlled 24 hp6464c DC power supplies which were monsters used for electroplating. You should try to get one of those supplies for some high power projects.
My father was a tool & die machinist & maintenance guy who eventually got into installing & upgrading NC machine controls. In the mid-80s, I was in college & got to work with him on a project emulating a paper-tape reader & other peripherals with a DOS PC, interfacing with Texas Instruments controls. The full-featured control had 12 K of DRAM, and they needed to run G-code programs in the tens of megabytes, so the PC had to spool data to the control and react to control signals. Both serial & parallel interfaces were supported.
As the code monkey, I had a control unit in my apartment for development & testing. It was the size of a small refrigerator, sounded like a modern server rack and had a toggle-switch interface on the front panel. Booting the monster involved unplugging everything else on that circuit, latching the bootloader (four 16-bit words) into memory and executing them to load the machine executive from paper tape. We had to breadboard enough components to fool the machine into thinking all of the limit switches, pumps, etc. were connected so the executive would think it was in a real machine tool.
Perhaps not the typical father/son project, but watching this video brought back a lot of fond memories for me. Thanks for posting it!
In the CNC world this is called "Drip Feeding"
This reminds me of the long routine getting the Data General Nove 1220 up & running at college, but the start-up procedure only involved flipping the appropriate toggle switches to set various registers, etc., through buttons on the disk pack drive(s) to actually telling the machine to "boot up." We didn't have to start up with paper tape because previous students had RTFM'd and loaded everything needed onto locations on the disk pack(s) and most importantly, documented everything required to start, run, and shut down the system. My Machine Design Professor had it tucked away in a "closet" in the (Mechanical) Lab building and he allowed some of us to use it for our "Senior Projects." Having access to an entire minicomputer with only one or two of us sharing it sure beat going to the Computer Center which had two (mini?-)computers (one upstairs, one downstairs) shared by nearly 100 students/terminals. One of those machines was an HP, but I don't recall which model it was. This was between 1979 & 1982, so the HP may have been an HP 1000-something like Jack has. But once I got access to a virtually _personal_ minicomputer, I never went back to the Computer Center! ;)
Maybe you can remember enough RE the data general to help the guy over at the tech tangents RUclips channel, to get up an running when his switch on occurs.
@@beingatliberty I watched his latest video and commented recommending a DG Guru to contact for possible PDFs of schematics & other docs that he doesn't already have. All I (think I) remember is turning the NOVA 1220 on first, toggling in appropriate data/commands, _then_ starting or maybe "loading" the platters, waiting for the appropriate "clunk" (the light was burned out), and finally (I think) hitting "run."
I did subscribe to his channel to see what happens next. ;)
When I used Data General Eclipse computers, we sometimes had to enter what my co-workers called the "Hip Boot". We had a 20 Mb disk drive (required a forklift to move) and the Hip Boot would boot from that. Most of the time we didn't have to do that because boot PROMS were installed. The DG Nova computers we used for manufacturing test all had core memory. Test programs were loaded from a paper tape reader.
I'm so happy I finally get an idea of how you would operate these machines. Grateful for all you guys who have the knowledge and share the love for these historic pioneering machines. Without them I would not have been able to setup my PiZero2W as a Klipper control computer for my 3D printer tonight ;-). I love these videos, thank you Marc!
This brings back such pleasant memories of my first encounters with computers, about 1976 - 1978 or so, at Diablo Valley Collage in Pleasant Hill, California. I weaseled my way into a job as a computer lab assistant, so that I could kick everybody out of the lab and play with my own projects without interruptions late at night.
We had an HP2000 Time Share BASIC System. There were two HP2100A processors. The main processor had 32K words of core for running the user's BASIC programs and most of the OS, with 2-second time slices in a sort of cooperative multi-tasking kind of thing. The I/O processor was nearly the same machine, but with 8K words of core, and the job of handling serial I/O to 32 user ASCII terminals.
The system had the same disk drive, paper tape deck, and mag tape deck as Jack's system. It also used some sort of core drum memory device; a rotating drum that was chosen because it could help read and write program swap space on a regular periodic basis. I never knew much about it; it lived behind that smoky plastic and just did its job. We just called it "the drum."
When I think of the machine, the first thing that comes to mind is always the buttons. The 1/2" (12mm) buttons on the front panel of the 2100B were just delicious. Big enough to feel right, and with a long, smooth, clickless travel so that you could make a clear statement that you wanted this one pushed. They were all binary, and lighted -- light turns on, light turns off. Entering an octal word was similar to what Jack showed. The trick for entering the long boot loader was getting into a frame of mind where you would see the previous word, the next word you have to enter, do an exclusive-or in your head, and push just those buttons that needed to be toggled. I loved those buttons.
Every night... the lab assistant would send everyone home, put the appropriate reel of mag tape on the drive, and make a fairly fast incremental backup of the student's work stored on the hard drive.
But every Friday night... a full backup to tape of the entire disk and operating system, which took maybe an hour or so. Then... key in the 64 words of boot loader through the front panel, in octal, tuck the three-inch roll of paper tape containing the operating system into that little pocket you can see to the left of the read head, and push the button. The tape reader was super fast, relatively speaking -- optical, so there were no mechanical issues to get in the way of blinding speed. So the paper tape would fly off the roll, through the read head, fly out into the air, and drop into a big snaky pile on the floor. What you did not want to do at that point was stomp around with your big shoes and smash the tape, because... I'm not sure if we had a second copy!
So now with the operating system loaded into core, as far as the system knows, it's a brand new machine, fresh off the showroom floor. But the full backup is still in the tape drive, and we enter a few more commands on the teletype (now that the system is up!) to restore everything from that tape. As the student's work gets read back in, the disk drive gets repacked in a nice and orderly manner, eliminating fragmentation. When that's done, everything's running nice and smooth, ready for classes on Saturday or Monday morning.
So there's miles of paper tape on the floor. Leave it to HP's ingenuity. On a little bracket mounted at the side of the paper tape reader, there was... I've never seen anything like it before or since. Think of a stylized electric knife handle, or something like that, sort of an olive green plastic handle-case sort of thing, containing NiCad batteries and a motor and a button. On the motor's shaft sticking out the end of the thing was a plate about three inches in diameter, with two pins sticking out near the center. You would fold the start of the tape around one of the pins, hold your free hand just right to keep the tape going where it was supposed to, and push the button. The miles of tape on the floor would slurp up into a tidy roll. No "spool" required -- the roll was self-supporting; just a roll. No complex mechanisms, just some creative "Why not?" engineering. It took probably some fraction of a minute to wind up the whole roll, if I remember right. The word "fun" applies!
We also had an IBM System/3 -- not the new one, the old one, big as a car, 8 bits, 32k bytes of RAM, an IBM Selectric console, a 1402 card reader, a 1403 printer, and what I consider to be a *very* cool instruction set with variable word lengths. You could move or add or subtract integers or packed decimal numbers up to 256 bytes long. Brag about your 64 bit integers... these were 2048 bit integers! Loads of fun.
Oh my - I learned HP Basic on a time share HP1000 system that was run by the school system in Oregon back in the 70s & 80s. I visited the computer center a few times and saw the machine - looked very much like that one. Good times.
Yes, I think the system was called the HP 2000 TSB (for Time Shared Basic), and it consisted of either two HP 2100, or the later versions had two HP 1000. One was the main basic interpreter and the other one took care of the communications with the many user consoles. Some early versions apparently had only one machine. Oh apparently the first version ran one on one HP 2116 with 16 users (see www.decodesystems.com/hp2000/#photos ). You must have run out of memory after programming just a few lines. I’ll have to try that one day. They were popular in schools.
You're definitely talking about the HP 2000s that most school systems in Oregon had at the time. Our system was shoved in a small narrow room between a classroom and a stairwell. The room was so small that two people could barely fit in there at the same time. We student hackers stole passwords by tapping a Radio Shack Model 100 to the phone line that was connected to the modem in the front office and waiting for someone to log in. It was important to turn capture on because the passwords were always control characters. We never got caught but I later found out that everyone knew about it.
Hah - fun bit of hacking there. IIRC the computer system was called OTIS and the computer center was a nondescript concrete building on Hwy 99 just on the northern outskirts of Eugene. I went there a few times with some fellow students and looked around and chatted with the guy running the show named Ben Jones. They also had some large IBM systems that were used for record keeping throughout the school system.
I started using the HP Basic systems in 9th grade on an old Model 33 teletype that was jammed into a tiny soundproof cubicle in the back of a math classroom. In HS we had a dedicated computer room with two teletypes and an old Hollerith keypunch. Before I graduated those had been upgraded to video terminals and we also had some TRS-80 and Commodore PET machines. Somewhere along the way the HP BASIC at OTIS was upgraded and included interprocess communication statements - chaos ensued as students throughout the state were able to chat with each other. That got shut down after a while because the suits thought it was too disruptive. What a bunch of hooligans we were.
@@ebrombaugh There was also a Oregon state network that connected university computer systems. In Portland I was able to connect to the DEC System-10 at U of O through a local phone number which I spent hours on for free.
Looking back at HP TSB, I wonder how it could possibly work reliably. There was no locking in their BASIC (that I saw) so writing to files on a multiuser system had to create unfixable bugs unless the programs used lock files. Someone could read old data from a file and then write over someone else's new data.
@@scottlarson1548 One of the other students at our school pioneered a BBS/chat system on the HP using shared files with some careful file locking implemented by hand. Kids from a handful of schools were able to post messages and chat in real-time with his software. The rest of us studied his approach and there were many variations on it being coded in our school. It was the wild west and we were learning a lot about these things.
I'm in the middle of what I thought were tedious computer setup steps. After watching this I gave my computer a little pat and said "Good computer, you're not so bad after all". Thanks for the video.
Me too. This is something else to see.
@@wishusknight3009 I like to collect and play with some proper esoteric systems. I have a couple(!) late model PDP-11's, and they're definitely a different beast to modern machines. Whilst i haven't got anything of this ilk, basically any sort of late 60's/early 70's minicomputer (and ealy toggle switch micros like the Altair 8800) are basically my white whale in computing. It's a whole different world unto itself
@@the123king My "white whale" starts with early 16 bit home computers and pc's. So a little past the need to strap a boot loader.
@@wishusknight3009 Eh, Amigas were my white whale when i was younger, but then i bought about 5 and got bored of something so easy to obtain
@@the123king Amiga's are not so easy now if one is on a budget. I love amiga's but for now I only use one on Mister. I sold the one i had a few years ago. Though I also like early motorola macs and PC's up to the 386. And have a few of the more popular types.
The esotaric stuff I have is an everexx deskside workstation from the mid 80s that was originally a 286 and was upgraded to a 386 sx at some point. Really interesting machine,. Has a very elaborate accelerated graphics system and is all on a passive ISA backplane. Also has some of the earliest large scsi drives.
I also have a few early laptopts and some of the more wacky early prebuilts. An x68000, An FMTowns that we have had since new... and a Colico Adam in perfect shape with a boatload of stuff with it, (though I dont really know what to do with that one tbh... No one wants them.) . A Compaq Quad xeon 400 server, an (presumably) Intel branded quad PPro server. And of course a lot of generic stuff that most all collectors have.
Nothing too exotic really, but its a varied selection with some interesting bits. And quite a bit of console stuff., though nothing particularly special nor rare.
Wow - this is a labor of love to keep these old machines running, and to remember how they work. Jack has a very impressive collection there.
This is so hilaric. As a freelancer I entered a video editing room in 1988 on a sunday morning 9 a.m. to edit this tennis report for the evening sunday night sports programme. Most equipment I recognized but the character generator (for the titles) was new for me. As there was no manual, I had to call a person who was familiar with it (at sunday morning at 9). To get this machine alive I had to made the most ridiculous keypad combinations. After half an hour of instructions and trials it went working.
Hey great to see a live E-Series after all these years. I was a field engineer who used to service and repair the HP1000 M, E, and F series (as well as L and A series) in the 1980s to 2000s. It jogged a few dormant brain cells seeing you trying to boot from the 7970 (a long winded pain in the ass to head align). Seeing the E-series showing HLT77 (happy halt) made me smile. Never worked on a 7900 hardly, it was mainly 7905/06/20/25 storage on the MEF range, which I also spent many a day repairing and PMing. Great video. Thanks.
First job out of university was at HP Cupertino working in the DSD Lab on RTE for this machine. We had the entire HP1000 hardware and software team occupying one floor of 42U, Cupertino Integrated Circuit Operation (CICO) Fab (remember SOS?) below us in 42L. We could walk next door to the HP1000 (M/E/F) manufacturing line and QA in Building 41(or 40?). I learned so much in that job. Our team that did RTE-6 had just 6-7 people.
Best job in my career, great team, great times.
That place even sounds wonderful! Thanks Jack!
Just dropping a comment to say thank you for documenting these vintage computer adventures you have, they are always facinating and fun to watch. Being 34 I never experienced this equipment myself, I had an uncle who worked on this and other stuff like it and would vividly describe it when I was a kid, so getting to see some of it running and doing it's thing, is really cool stuff Marc. Like I said thanks for documenting and sharing this.
When I was a medical service rep I used paper tape to load the arrhythmia central station.
Love this stuff. In the early 80s I programmed a HP 21MX in a large automated test system (4 or 5 full racks of instruments). Most of the instruments were controlled by HP-IB (IEE 488). Many of the vintage HP instruments in your videos were in the system.
This was a fun and nostalgic video.
We had a 2116 in our repair depot in 1976. Plugging PCBs into the rack took a lot of force. My company, Measurex, still had 2116s being used at customer sites. One day I was given floating point ALU boards to test, giving an already out of date computer a new lease on life. We had lots of 2100 computers. Those ones I could lift.
Measurex? Maybe you know our restoration buddy Ed Thelen?
@@CuriousMarc No, the name doesn't ring a bell. I was situated in Québec, and it was the late '70s. I liked playing around with HP 2100 computers then. I wish I still had the handbook for it. BTW, I have been watching your AGC videos. It appears to me the AGC was a bit-slice design and its machine code was microcode. Years later Lisp Machines came out of MIT. They also used a bit-slice architecture, and were stack machines; narrowing the gap between microcode and high level languages even more.
I enjoy all your videos. I own a lot of HP test equipment, and like General Radio gear as well. I restored a GR 1710, a network analyzer.
@@williamogilvie6909 Ed was in California I think. The AGC is neither bit sliced nor micro-programmed.
@@CuriousMarc I guess I assumed it was a bit slice architecture after seeing a couple of AGU modules with "SLICE" in their names. Your series of videos on the AGC is very good. Anyone with some technical training and experience can follow what you did. It was a very competent team. Thanks for putting this on utube.
This was very interesting. Glad there are computer hobbyists who keep the old tech alive to teach us younger folks about it!
What a different world we have now. That tape drive was pretty cool; lot to set up there. Easy to take what we have for granted compared to this at one time. Beautiful machines and demo!
I'm 33 and work in IT. I remember almost 15 years ago when the company I worked for had to fly a contractor in from the other side of the country because he was the only person who knew how to work on our accounting mainframe. They had used the same system since the early 80's. I remember that it was a complete black box mystery to me what this cabinet was in our data center. I would talk nicely to it so that it didn't break because I knew that no one in our company knew how to fix this thing if it did break.
Finally around 2012 we switched over to SQL reporting and again had to hire people who could emulate the features of our old mainframe in SQL and Business Objects etc.
When booting from a disc drive the lower 3 bits set the disc unit number to boot from. The disc controllers supported up to 8 drives per slot/controller (unit 0 to 7).
Brings back memory of 45 years ago of my time at HP repairing those suckers.
Question was that a 7970 E or 7970B.
Don't remember ever seeing an E on a 1000 system back then.
Thank god for miniaturization and GUIs!
Nice... I used the same hp setup to read punch tapes to run manufacturing equipment. We modified the teletype board to run faster by swapping out the crystal and changing a few of the capacitors. There was no real take up reel for the punch tape, so it would heap up on the floor as it was loading. We would stand real still as the tape would cover our feet since we had to manually help it at times. Then quickly rewind it back without it getting knotted up. We also had an HP2100 and that same tape drive. The tape drive died. So I had all our tapes read by and external company and emailed. I then re-wrote the bootstrap to read data from the rs232 card and fill memory with what was found. I also did a 1's and 0's check on each memory location first(magnetic core). Wrote it in HP RTL assembly on paper, converted it to binary (octal) and hand enter the code. Worked like a charm.
always love the music choice also love my uchicago alums!!
yes chicago is the coolest city in the world methinks
Pulp non-fiction. Excellent video, enjoyable narration, and information that almost nobody will use. Made me smile, though!
So fun to see one of these running again. I used an HP1000 in my early days in the Air Force in the late 80s. We used a test system controlled by this computer to test missile, and missile component readiness. Our interaction was pretty basic as I recall, mainly inputting a number into the S-register iirc (111102 maybe??) Store->Run.. something like that, then moving to the terminal for the rest of the test. The program was loaded off a (I think) 14-inch patter drive, quite advanced from the paper drives I suppose.
Marvelous! We had these in ‘76 at the then Westinghouse (now Northrop Grumman) Baltimore plant. They drove the PC board testing equipment. We had a number of these units and they were upgraded over the years
Ahhh... paper tape. Now that brings back fond memories. Much later we had 8" floppies to boot the Bull Level 60 mainframe. Those big beasts were slow... but fascinating. What was frustrating was that it had many, MANY blinkenlights, as a proper computer should, but behind a interlocked panel in the main CPU cabinet. Couldn't open it during normal operation. Sad!
I love your channel, Sir! I find your humor wonderful too. I passed your channel onto someone I know who used to work in the Apollo program. Thank You!
Wow, this brings back memories. Used to look after an industrial control system with two HP 1000’s running rte, linked to ten HP2250’s connected via HPIB extenders, which interfaced to about 1000 different I/O and controlled an offshore drilling rig! If I remember rightly it was 141200 in the S register to boot from the 7912 disc drive, but it’s been a while…. Consoles were the awesome 2645’s. Had 7970 tape drives too, but that was on the HP3000.
This channel brings back great memories, and always entertainingly done!
Den? The level of combobulation in Jack's lab surpasses mine by a few orders of magnitude. I'm damn envious of all that nerdy stuff down there.
HP 5512A? Now that's a thing Nakazoto would love. Same for me :)
Thanks for sharing a story of how HP got into computers - probably the only industry some people know HP for.
Looks like a similar setup I used. I used HP1000 mini in college. Modified an 8008 assembler with extra instruction set for Z80 microprocessor. Punched a few switches to use ascii tape reader to boot it. Once booted, system was on the hard drive with the huge head coil in it. We did not have the mag tape.
My first paying job as a computer repair technician was servicing these HP computer systems back in the 1970's for an HP reseller. The HP 2116 was the workhorse until the 2100 and 21MX became more affordable. Compared to other mini-computers back then, HP equipment was over-built, which is probably why examples like this are still running today. Hideously expensive compared to today's commodity PC hardware. The concept of the CPU halting and being manually controlled seems so foreign but was common place back then. When micro-processors came out they had dynamic RAM memory for their internal registers and thus had to be clocked constantly to "refresh" the data, thus the BIOS boot ROMs on power up and immediate dialog on a terminal. Wow this brings back memories. It would have been nice to see the 2116 in action with all of it's front panel lights. You can actually "single step" a program and watch each instruction manipulate the data in the registers as they are displayed. In fact when a CPU would not boot because of an internal logic, bus or core memory failure, we would have to "toggle in" simple instructions and single step them to see if that would give us a clue to the problem. Such as a dropped or hung data bus or address bus line. Hard to believe these systems supported multiple users, running programs written in Basic, on only 32KB of main memory.
What an amazing computing wonderland!! Awesome video. Jack is the best! 👌
The 2116, 2115, 2114 and 2100 are core memory machines, you load the last 64 words of memory regardless of memory size. Protection was via a switch on the 2116, 2115, 2114 and a push button on the 2100. More fun to load the boot prog manually. 9500 test systems were fun to work on.
Many years ago, as a CS student, I had a summer job at a research lab which had a 2100 and a 21MX. I mostly worked with the 2100, loading their custom OS from paper tape and debugging it from the front panel. One annoyance about the 2100: the front panel used incandescent bulbs, which burned out. I had to do all my debugging with one dead bulb, which for some reason the lab wouldn't replace (I'm guessing they had let the maintenance contract lapse). Fun times!
I would've replaced it myself! I'm only kidding. But I would be tempted...
Just take the bulb out of HALT/CYCLE or INSTR/STEP. Those were momentary and really did't need bulbs. Or even FETCH or EXECUTE. Not really necessary.
Our hp2100 constantly lost bulbs. Needed a plastic tube inserted around the bulb to pull them out. After a few times I started to replace them myself. Also had a lot of custom interface boards inside which learned to fix.
Another gem of a video. Thanks !! Always look forward to new content.
Dave, I've "shrinkified" the processor. Another golden Marcism!
What great memories. My first computer in high school (1977) was an HP2000E based on the same hardware. Them my assembly and microcode classes were on HP21xx machines at Chico. Lots of fun.
I learned basic on a HP 21MX time share in 7th grade. A year later, I wrote a false login prompt password stealer and stole the administrator account A000 password and it’s secondary password. A few weeks later I got busted and suspended for the first hack ever in my school district. I would love to see more videos this computer that was stuck in a glas-housed room of my school . Hopefully you can make many more !!!!
The USAF made us plan/write out, enter manually, and run a few subroutines in actual machine language instruction coding...
A colleague has been looking at our similar dual machine with washing machine disk drives (which do sort of spin up) and a printer but no paper tape. Just booting the thing could make a really good demo.
Do you mean IBM 2321 Data Cell Drive (strip file) ?
I mean an HP 1000.
I worked on HP 21116 based Automatic Test Systems (ATS). You punched in the boot loader through the front panel. This loaded another boot program through the paper tape (we used metal mylar) reader. Then you loaded the operating system from disk. The was HP Test Oriented Disk System (TODS). The main language used was HP Test BASIC. HP Test BASIC contained all the function calls necessary to control HP instruments and scanning devices.
Before HP-IB so they used custom HP interface cards. Before HP RTE.
I know I _”shouldn’t be able to”_ but I can hear the whine of that CRT, and I’m 47 years of age.
Apparently the average Joe can’t hear such high frequencies unless he’s a child or a teenager, but let me tell you this - I took a hearing test a couple of years ago, and my audible range cuts off just around 18Khz.
Btw I have Asperger’s, so that’s a very significant factor. I can also hear my kitchen clock ticking from the next room, around a corner.
It's not an accurate diagnosis for being on the spectrum, for sure, but a surprising number of us have that high frequency sensitivity. Helpful thing to have as a technician.
@@jnelson4765 Indeed! High-pitched sounds are a pain to many aspies but to me it has never been "noise", it's just the sounds that tell if things are working like expected :)
@@jnelson4765 for sure. High frequency _and_ high volume sensitivity. I can hear things going-on outside and among neighbours that seemingly no one else can.
Reminds me of my Heathkit H8 where you had to enter the initial commands in split-octal on the front panel to load Basic in via the cassette player!
Why doesn't the H8 get any love from the vintage computer guys?
I actually study startup processes; how they have changed from the first computers to the present day. This material is really helpful. Thanks.
Wow - we take so much for granted about how modern computers are just ready for input and running things after they start up, running a zillion processes automatically. Very informative!
It really tripped me up when you were reading sequences of octal digits (100) as if they were decimal ("one hundred") rather than octal ("sixty-four"). Apart from that, I really enjoyed this video!
That’s how you read octal. Unlike hexadecimal, you can pronounce it like a regular number. It’s just base 8, that’s all.That’s the one and only beauty about it. Except for that, it’s miserable.
Brought back memories of an HP-1000 w/ RTE-IV B used at a chemical plant in Tacoma, WA during the early 1980's. Not sure how much it really helped in improving process efficiency. That plant closed down at the end of 1985.
Last time I saw an HP-1000 in actual use was as an adjunct to some chemical laboratory devices in about 1998. Still going strong, but spares were an issue.
When I was a kid I was couple of times at my fathers work, he was technician for those computers in former Czechoslovakia - soviets tried to copy everything from the west, but they were not able to reverse engineer the IC and lot of other stuff, so they were making these vastly big computers taking the lot of space and still breaking. In Slovakia they were able to smuggle first PC computer from West Germany and to reverse engineer, thus first computers born in Slovakia. Limited BASIC but it was incredible.
Fascinating stuff..I always wondered what the procedure was to get these kind of machines up and running. We take so much for granted now. And it's great to still see these amazing complex machines still in use.
It's always fun going to Jack's place. His slide rule collection is the best. Looks like he's moved more stuff in since I last had a chance to visit.
I am so glad that I didn't start using computers in the days when you had to poke values in the program counter by hand, in binary, before you could run anything!
While an undergrad EE student in the early 80's, I rescued an old varian 620 which was going to be scrapped. I pushed it into the corner of a lab and proceeded to get it working. It was built entirely from wire wrapped 7400 TTL and many of its instructions were malfunctioning. I managed to get the CPU and disk controller working but one of the three 8K core memory stacks was giving me a lot of grief - so I couldn't get it's "vortex" OS to boot (it would load but, well, there was a big chunk missing).
Before I could sort the memory out, the guy responsible for the lab decided to evict me (without notice) and put up a sign saying people could help themselves to my machine for parts.
It disintegrated before I even became aware of this plot 😭
I was somewhat annoyed by this...
I find this stuff very rewarding on old computers, I find it extremely frustrating and obtuse when it happens on a modern linux ... cause it means usually I am at that level cause something that has been working fine for years got borked during an update and now its not even an option
For quite long along this piece of history I kept thinking that I had been around the HP 1000 once in my life. Until it finally dawned on me that I had mixed the 1000 and the 3000. But then I also realized my fake familiarity may have been due to similar actions that had to be taken for booting a DEC 11. Never had to do it myself, but saw it happening a few times in the long route and then a second time when a more knowledgeable guy said: "I don't think we have to do it that way - I think we can use a shorter setting, let me try..." And he did, with just half a dozen entries and the DEC woke up just fine. That time at least there were more than one way to "skin the cat". Also, by that time I had some experience with the marvelous HP 9100 and had my own HP 35. Maybe even the HP 67, but never became any sort of computer guru. All my exposure was centered on learning just "enough to be dangerous" about Z80, Z180, MC6800, '6803 and '68000. In the meantime, I got my HP-87, HP-15 and HP-42S as well as the binary/octal/decimal/hex calculator HP-16. The '15 lives on as simulation in my various personal computers, including my Android phone, iPad Mini and MacBook Pro. Oh yes, the very first computer I was "near" was a British Elliot Mark something that ran either Autocode Mark II or some sort machine language. I pal of mine became an expert on the machine code. I barely managed the Autocode to pass my exercise and then promptly forgot it. Its input and output were originally punch cards and then paper tape. I have become ambivalent about either liking or hating some of this in my past.
This brought back memories of using the front-panel switches to boot UNIX/RT on the PDP-11/70.
The kernel would read the switches on every clock interrupt so the quickest way to bring the machine down was to flip all the switches up, resulting in a "Panic: Zero."
The "Store" button needs a beeper to confirm contact was made in the pushbutton switch!
I was thinking exactly the same thing. It really does.
The 9100 is my Holy Grail - the only machine that beats it is the AGC, and that “belongs in a museum” (like the ones you’ve worked on). Marc, your channel is a joy - please keep up the great work. I’m wondering: do you dabble at all in “mere” HP calculators? That’s the extent of my collecting…I’ve repaired the card readers on an HP67 and in the accessory for the 41C (the extent of my tinkering).
Holy cow! I learned to program BASIC, FORTRAN, RPG, COBOL, and Pascal on an HP1000 in 1981/1982 at a community college in California.
Old school computer room work, load some bits into a memory address, bootstrap bootstrap, the smell of this gear in operation, is more like an electronic medical centre - worthy of a labcoat to operate, some faint ozone generated or something, and halon gas systems to protect the data from fire.
This should be required viewing for those who complain about Windows taking too long to load.
Brings back memories... we had HP 1000s before we replaced them with 3000s. And that was over 30 years ago. 16 bit binary boot programs were a great advance over the 60 bit octal boot programs you had to set with switches for the Control Data machines...
For me the nostalgic part here is the terminal (CIT-220+) that is the same that I got from an auction in about 1991 when my university installed the first "net" in the student houses (dormitories) i.e. RS-232 lines (I think it was 9600 baud) that used the wiring of an old PA-system that was in every room. I do not know what they had used the PA-system for really, perhaps some radio programs because there was a selector switch for many "channels". But now we had access from our student apartments, with the dumb terminal, to the university UNIX servers and therefore email, usenet groups and more :) I still have that terminal in storage and last tested it about two years ago, still working.
At 11:10, the terminal's keyboard came into full view for more than a second and caught my eye: hey, I have exactly the same!
Looking closer at the terminal, I recognized a C. Itoh CIT220+. I have a CIT224, obviously equipped with the very same keyboard. It still serves as the console terminal for my server running under Debian.
Does he have a red infrared light to keep the computers toasty and happy in the computerarium?
This model was introduced five years after the first Xerox Alto was built? Incredible.
Yep. That’s how far ahead of the curve the Alto was…
It’s amazing we ever got anything done in them days
Some ancient HP test instruments are ridiculously hard to find in a decent condition nowadays. It's nice that you guys restore and repair those rare elusive beasts. It's interesting that some classic HP computer basically had the early form of multi-boot setup, in this case which ROMs to boot the bootloader from.
I did my first F77 and Pascal programming on an HP 1000 running RTE/6 if I remember correctly.... 1984
First 21mx computer was 2105, MX stood for Measurex who asked hp for a smaller computer for process control. For example, paper mills used the Measurex control for paper thickness.
*pans camera* "where's the DEC stuff, where's the DEC stuff... HOLY CRAP A STRAIGHT 8!"
Yup, this man has the right credentials.
My first job working at Bell Labs, back in the day, was programming HP 2100 and MX computers to do IC layouts back in the late 70s and early 80s. Good times!
o my, if you keep going back in time on how things work, you are very close to how they made the pyramids 🤔 good job
Ah, yes. The feeling when you remember that one obscure option for tar and it actually works. Well done!
Oooh -- the HP 9825! I would get to play with these things a few times at a GF's high school -- great fun these things are! I think the one I used may have been a 9860, but it was a similar form factor.
"when we finally get a command right on linux" truer words were never spoken, hahah
The entire time I watching this my internal monologue was screaming AUTOMATE THE BOOT PROCESS!!!
I have recently managed to get Puppy Slacko7 64bit to increase screen brightness on this old Toshiba laptop from utmost gloom to 100% with a shell script that runs automatically at boot. So I concur with your Linux command comment! Easy when you know how, but the road to rejoicing was somewhat rocky as I am a computing dilettante who refuses to... Better that I end the rant here.
When I was in school I used a HP 2100 for simulating solar photo chemistry (in the EE department) and Conway's game of life.
Show of hands, please: who noticed the RPM of the tape reel was nearly exactly synchronized with the Jeopardy Think Music beat? 🤔
I am soooo jealous - worked with the 1000 for many years, and I am currently building a simulator with a real live front panel
Thanks Jack for inviting us into your vintage HP Computer lair.
When booting a computer required magic spells - and anyone succeeding was considered a wizard...
en.wikipedia.org/wiki/Clarke's_three_laws -> 3 :-)
I don't know if you meant to do this, but at the 12:10 minute mark when the Jeopardy theme is playing, the RPM of the tape is on beat with the music.😂
At Raytheon, I used an HP 1000 using the RTE-II Real Time Executive to write FORTRAN programs
love the nerd jealousy gents yours has a printer hmmmph...brilliant love your content thanks again
I maintained HP 2113MX computers in the USAF for several years, and , if there is one thing I have memorized, it's RPL states I used on 9 different systems 40 years ago, and thinking to myself, "store, preset, ibl, preset , run!"
I also had to tear down and reassemble several of these 'Rubik's cubes' chassis, as the front and backplanes would loosen up their grip to the mainboard, and, their grips on their assorted interface cards; a quick 2 hour tear down, and judicious use of a small flat heard screw driver to rebend all the little spring loaded connectors in the slots, and maybe the computer would then be good for the next year....or ...maybe not!
I used to see the old HP1000 computer at the University electric engineering dept. gathering dust . After we acquired a brand new HP3000 .This computer was forgotten and never run again
Learned Assembly on this machine using card decks. Was the primary CS weedout course @ Chico State back in the day. I also worked on the 2000 and 3000 series. The 1000 was a bitch to program since it only had 5 usable memory registers of which E was just a one bit overflow.
ya.. for all those that 'interfaced' with a tty terminal with paper tape and telephone modem.. you got to do the telephone modem part.. like with 16 lines going in..
Like installing dos 6.22 and wfw 3.11 from floppies. Compared to having a micron or so dos / wfw cd-rom disk to install.
Future generation will look at these systems with awe and bewilderment.
Ignored the short. Because shorts are trash.
Waited for this video! And this one I shall enjoy even with the silly vertical first bit.
More memories. I didn't like the move from octal to hex, the computers I used had opcodes that seemed to align with octal, but as assemblers and debuggers became more popular, the meaning of the binary opcode was not as important.
Awesome 😀
Hardest part was generating the operating system. I would start it before I left for the day and it would usually be finished in the morning. Since the disc drives were voice coil, there were a few song programs.
Aww…Memories! My first computer program was a biorhythm program running on the HP 9825 using HP Basic. The rest is history. :)
I just missed toggling in a boot loader as our college got rid of it's Data General Nova the year before I started.... I feel I really missed out on living through that period of history.... maybe I'd think differently if I had lived through it. ;)
The terminal is a VT-220 clone from C.Itoh.