old computers were varied and interesting looking and interesting to work with and fun to work with, and you could extend them by making your own hardware modifications.
http://oldcomputers.net/imsai8080.html (the real thing)
I figured I would search for s100 bus computer ads. Didn't find any. They are probably all in Dr. Dobb's Journal of Computer Calisthenics and Orthodontia (now I think available as volumes), Popular Electronics, and Byte, all of which are out of print now. I am a tinkerer - this is why you find so many machines listed here. My electronics skills were somewhat decent, but not at debugging - didn't have any training and practice on any proper equipment, but I had a logic probe & I knew a few things, enough to keep me going. And I learned whatever I could - I was a sponge. Everything I got, except 1 item and my tools and a number of parts, was used. I was always pulling chips off of boards because I couldn't afford to buy them at the store. I was always collecting boards that had TTL chips and memory chips wherever I could, but I avoided dymanic RAM, because I didn't know how to work with it. I was somewhat OK at timing diagrams, but not great, and trying to improve.
Before I was able to get a computer I traded for a DiskWriter (sorry, no picture). I had a nice big-screen terminal. All the DiskWriter did was basically log terminal I/O to floppies. And I used it. You could page through your file, tell it when to start writing and when to stop (I think it obeyed the standard control-S stop and control-Q start I/O, and probably control-C for break/cancel, there were a few others), and it had a lot of other cool features for a little lunchbox-sized case. It worked with any terminal (didn't matter what control codes or special cursor sequences you had), had 2 ports, one of which I used for a modem & I was on the BBSs (Bulletin Board Systems).
[BBSs were popular before web pages & newsgroups & blogs became popular. If you had a terminal or terminal program and a modem, that's all you needed. You were online. You would dial in to one of the local access numbers (or *the* local access number, depending on their arrangement with the phone company), and post or read messages, much like a newsgroup, but without the fancy graphic interface - it was all a text screen like a dos prompt. people sold their stuff on these boards or just talked about their computers, or shared a joke, or helped someone fix a problem (and there were plenty of those to fix - and you did it yourself).]
Later on I had an Adm-3A terminal for a short while. Solid thing. Fun to pop open the flip-top.
I had an SWTPC 6800 (SS50 bus) by SouthWest Technical Products Corp. - I think I used it with its companion terminal I traded some stuff for. I traded for everything because I didn't have any money when I was a teen.
When I got a real S100 bus machine (might have been an Imsai without the front panel switches, but I think it was some other MFR) I got all excited when I got a Solid State Music memory board. I had to solder ≅1000 socket pins. There were a lot of 2102 1k memory chips on this S100 card (an array of 8x16 or 16x16 chips with 16-20 pins each I think - a whopping 8 or 16K of memory)! I got a headache from all the soldering (hint - open a window next time!). And I wanted to use sockets in case one of the chips went bad. I can't remember if I ever got the board working. I don't think I did. I think I blew something. every s100 bus mfr wired theirs a little differently, so you had to be careful with your power & signals. Or maybe power was fine. I can't remember. That was about 16-20 years ago. They tried to make the board design flexible by allowing you to route signals with jumper wires & cut traces I think.
I also had a Processor Technology VDM-1 board for video. I was pretty excited about that, but I don't remember getting it to work very well, so I quit using it. I think I had a bad chip somewhere, probably blown RAM or TTL. I was a teen, and not quite smart enough (or rich enough) to debug it.
S100 (IEEE-696) bus machines were the most standard bus of the CP/M machines. Cards were somewhat interchangeable because there was a standard. The Imsai computers like you seen in the movie War Games were S100 bus boxes. The power supplies back then were usually heavy because the transformers were huge (linear power supplies). And so were the capacitors - you could spot weld with those things. I know someone who uses those to rejuvenate dead NiCd batteries (blows the crystal shorts that grow between between the plates over time). Manufacturers had interesting names like Cromemco, Pertec, Persci, Proteus, CompuKit, Altos, MicroBee Applied Technologies, Morrow Designs, Itty Bitty Computers, Itty Bitty Machines, etc. and some were even funny. You can probably guess where Solid State Music (Solid State Microsystems, then today's EMU Systems) likely got its name from - I think it's because their first board was an audio board for the S100 (Digital-to-Analog converter?). Later on you could get switching power supplies, but you had to be careful about providing a load or your PS would burn up. Today's switching supplies are much more forgiving. Welcome to the "boat anchor". There were the kludges and the kludges, whatever your system was, depending on the sound it would make when dropped in a lake.
For a while I had another S100 bus minicomputer with a couple of disk drives all packaged together, but the thing was not small (size of 2 trash compactors). Can't remember if it worked. but it had at least 2 8" full height disk drives, maybe 3. You need to realize minicomputer means mini in terms of mainframe size (Vaxen, IBM, CDC). Must have weighed at least 100 pounds.
things people did
In those days before the keyboard, people discovered that you could put your AM radio next to your computer and it would make noise. so they wrote programs that made tunes. and it made the magazine! (Dr. Dobb's Journal of Computer Calisthenics and Orthodontia, now known as Dr. Dobb's Journal).
It's interesting to go back through those early magazines and see what their first programs were. they had to fit them into 256 bytes of RAM. and they spent all day toggling the data in and debugging the programs. games like "mugwump" and "hunt the wumpus". and tic-tac toe.
these games came only to those who could afford a teletype (and who had the fortitude to oil & maintain it). An acquaintance of mine said he learned leaps and bounds when he got a teletype. he set it up next to the bathtub so the paper would spill into it, and print away and debug his programs.
Some teletypes didn't have keyboards (PSR-33?). some did (model KSR-33). later they had punched tape (cool! you can save your programs!).
later on came terminals. serial port standards were not quite defined. in fact, everybody was making their own port pinouts (their own "standard"), so things were a mess, and you had to custom wire everything.
if you blew an IC or power supply, you got the replacement chips and capacitors or whatever you needed and you would fix the board yourself - sometimes you could send the board to the manufacturer and they would repair it for your for a fee. boards were expensive, costing several hundred apiece and you could get them preassembled or you could get it as just the board or as a kit.. soldering one of those is a several day session in which you need lots of fans because the soldering fumes would make you nauseous. S100 bus frames were also expensive and about the only standard available that was the same besides I think SS-50 and the later Apple II bus).
What I did
Later I got a Xerox 820-II computer (not a PC!) but I needed some modifications (or maybe it was because I wasn't able to get a case), so gutted and modded a large terminal which wasn't supported much. I got a switching power supply for it, and some 8" full-height drives and a disk controller. I used this for a fair while. Eventually I got a double-density controller and double-density half-height drives.
And I kept that system busy. Eventually I messed up one of the drives due to an electrical accident. I disassembled, reassembled, and transported that system around. I even siliconed the connector so I wouldn't wring off a wire from too much use. Eventually that happened elsewhere.
You had to learn how to manually terminate your drives. (see http://www.mdfsnet.f9.co.uk/Archive/info-cpm/ for CP/M articles - you learn a lot - CPM emulators can be had).
I had Microsoft Basic-80 (a popular BASIC when it came out) with the reference card - I think I've lost the reference card, but the language hasn't really changed, it compares very well to Visual BASIC.
I got a Tektronix 4013 graphics terminal, and was I ever excited. I was able to get various boards and options for it, including a tape drive and tapes for it which held 300MB I think, and a graphics tablet. I was already tinkering with trigonometry functions and 3d plotting programs back then, and whatever graphic program I could get my hands on in BASIC and convert, I did. 3d wireframe stuff, and some pretty wild curves.
It had an APL keyboard I think. APL was a weird symbol language. All Greek & math symbols. But I used the keyboard as ASCII (can't remember how I did it, probably a board modification on the keyboard).
The manual set was huge. you could send control codes to have it draw graphics, and it used Tek's unique storage tube display to display things. You couldn't leave things on the screen too long or they would start to flare & get fuzzy.
the electron beam would draw a vector on the phosphors and the way the tube was set up, the phosphors would stay lit for a long time. If I remember correctly you didn't want to leave anything on the display for too long or it would burn into the screen. When you cleared the screen with the button or a control character it would flash the display.
It drew ASCII and APL characters into place, just like a plotter, unlike today's CRT's, which scan. I had a super-8 camera and some really really old film, so I tried to make a movie using it and the Xerox (Xerox wasn't always a copier or printer).
I modified the camera single-shot switch off the printer port with a relay so that the computer could control it, and filled up a roll with moving, changing wild trig curves in BASIC. Haven't been able to duplicate what I did because I didn't print it out. and I lost the program. I think I still have some 8" disks around.
I got some volumes of Dr. Dobb's Journal of Computer Calisthenics and Orthodontia (nothing do do with dentistry, everything to do with old computers and programming , mods), and some ACM journals, and some issues of Computer Music Journal. I started learning about IIR and FIR digital filters and different types of filters, the technique of using a time window in a digital filter, etc., and FFT's and DFT's and eagerly started attempting to implement them. I even got hold of a program for my Xerox II CP/M machine called Music Man I think, which was a music sequencer - it did realtime FFT's and you picked the "organ drawbar" settings for the sound you wanted, and programmed in the notes.
I got every piece of sheet music I could get my hands on that had up to 3 (or was it 4?) notes at once (4-part harmony), such as choir music, bach's music, etc. My mother had a box of sheet music I could tinker with, and she bought me a book of bach's chorales.
To do all this music, I had designed and built my own audio card, which was basically a DAC on a plug-in board. the plug-in interface was similar to the apple II proto boards, so I used one of those, or I had wired it into the parallel printer port interface, or one of the other parallel ports, I am not sure which, I think I tied it into the bus with some TTL logic (I had gobs of data books). later I tried to add an ADC, and had hopes to add multiple DACs for more music. I had an amplifier from an arcade game (my friend fixed them and had lots of parts, and we traded stuff).
There were 2 types of disks, hard sectored with many holes in the hub, and soft-sectored, with 1 hole in the hub. Most everybody wanted the soft-sectored disks & higher density drives - they were really impressed if you could get double-sided quad density drives on your machine (1 megabyte on an 8" DSQD disk!). 128K on hard sectored or SSSD (single sided single density) disks, but if you wanted compatibility with other people's systems like the Northstar Horizon and some others, you used hard sectored disks. When you got software from the SIG/M library, you might get either. At the CPMUG meetings, people would bring their machines and copy stuff from the library that they wanted. And copy they did. People would take turns hosting the library at their house I think. Guys often took SSSD disks and used a hole punch to notch the side of the disk to allow writing to the bottom side of the floppy. saved money.
LBM was the popular archive method. ZCPR3 was a pay-for CPM OS of choice. Z8E was a really cool free assembler. Any BIOS you got probably came with the source code (unlike today). Why? because you probably wanted or needed to modify it. So hobbyers were buying eprom burners. Everybody wanted one.
Microsoft's Basic-80 was very popular. XMODEM and KERMIT were the most-used file transfer protocols - ymodem & zmodem eventually came out and was made popular. Today it's HTTP and FTP and P2P-file-share-stuff. PIP was the command used to copy files. BDOS was the Basic Disk Operating System on which the OS was built. The Osborne 1 was a luggable. but it had a small, small screen.
The Kaypro was another 5.25"-based computer - I think it was also a luggable, but with a bigger screen. Everybody in my CP/M user group (which at one point stood for Computers and Programmable Machines instead of Control Program for Microsomputers) admired the Kaypro and Osborne. Occasionally you would see an NEC or TRS-80 laptop or drum machine, 8" floppies aglore from various other characters in the group and whoever weas hosting the vast CP/M library of floppies at the time (and everybody wanted to make copies). CP/M was the first standard OS for microcomputers. it's been a long time, I should probably look this up on wikipedia, instead of a BIOS you had BDOS, the BASIC Disk Operating System I think.
Radio Shack came out with the model 100 this really interesting mini-notebook-sort-of-thing that went out of production and out of style. it had a built-in terminal program and a decent amount of memory, modem that I wanted but never got. it's still a lightweight compared to today's laptops.
I had a TRS-80 model 1 for a while, didn't get very far with it, but I did tinker around with the graphics on it. I can't remember, but I think I had a coco for a while (or at least I had access to one).
I also had (and modified) a Timex Sinclair ZX-81. I quickly broke/wore out the membrane keyboard and gre tired of it, so I examined the traces and drew a wiring diagram and bought a Cherry raw 64-key keyboard from Radar Electric, my favorite electronics store.
I think I mounted the keyboard so the wires jutting out the bottom wouldn't touch the table & break. And I was happy (sort of - couldn't save my programs). You always needed more RAM, and if you did connect the RAM pack on the back, the solder-tinned connector would get jiggled & you would lose your program/crash, or the connector would oxidize on you & break its connection, and again, you would lose your program or crash or lock up.
it had the ever popular 4-MHz Z80 processor in it. I dreamed of a Z-180 or even a Z-280 processor box. There were a few late-model CP/M machines that had those, but they were higher-end.
A friend traded with me to get an Ikegami 19" open-frame color monitor, which I used on my PC. because I knew nothing about ground loops & related special power supply issues, I lost my 20-20 vision on that monitor - my fault. I was sitting with my face 6 inches from it (only way to view it), and the screen was fluttering at 60Hz vertically, for hours on end, and getting headaches (that's a signal!). Couldn't focus.
I refused reading glasses. That was foolish. To this day my distance vision is fuzzy (20-60), but I can still read 4-point text. I appear to be unable to bring it back to 20-20. Boy did that monitor get warm.
I think because it was open-frame and I didn't care for the idea of getting zapped by the flyback, I found some sort of metal holed cover for the top & sides, although some of it was relatively unprotected and just hanging out.
You know, those things they stick in big computer rooms on mainframes way back when? This thing must have been about 4.5'-5' tall and about a yard wide. huge. It took some effort to take it apart and haul it upstairs and reassemble it. Was it heavy - each part! It must have weighed 300-500 pounds total. I was glad (but always worried) that the floor or stairs didn't cave in (creaked somewhat though).
I found a way to tweak the handshake signals with a little TTL chip or two so I could hook it up to my computer (can't remember, but I think I had an XT clone then). Did my 10-page college paper on it just in time. The thing had a pretty decent speed, but boy was it loud! it had its own sound-dampening hood and a loud fan to cool it. You could adjust the paper width (cool!) up to 14" and I could use paper with regular holes.
It didn't have a full ASCII character set (this was for a mainframe) though, so a number of special characters just didn't print and I think sometimes you got a funny replacement. I couldn't get a replacement ribbon, so I discovered much earlier I could oil the ribbon to make the ink last.
Eventually I couldn't use it anymore and I needed the space (my room was filled with boxes of circuit boards, power supplies, parts, and electronic data books). That was an OUTRIGHT BLAST to use - too bad I didn't know any way to get more ribbons for it, or I would still be using it! what would be really fun is to have a 12-core workstation (not a desktop) hooked up to one of those massive room-sized color laser printers and do some massive printing and listen to it all doing its thing. I think I could go for that. I'm just not quite the copier maintenance man yet. I know a bit about about clearing print jams from printers.
Let me tell you about that 10-page college pager incident: - there's a fixit story behind it. Before the line printer was banished to my upstairs bedroom (what a haul!), I was tinkering around with it, like I do with everything else I own, and I remember pulling various (warped) boards from the cage, inspecting them, and putting them back in. Who knows, maybe I was vaccuming out paper dust, too. I was a clean-nut about my equipment.
I put the boards back in and fired up the printer to print my report before I had to go to class. The fan whirred, but the band didn't move! no clacking of the printhead! My heart sunk. I turned it off. I was already under a lot of college stress, and I started weeping and crying out to God for help. That paper was due that day (midterm?).
I calmed down after a short while and after 5 minutes started going through the printer again, looking everything over. I pulled all the boards out, but this time I noticed a small piece of square black plastic junk at the bottom of the board cage. What's that doing there? I pulled it out and looked at it. it had writing on it.
2200. Never seen anything like that before. Resistor? Maybe. Capacitor? Ceramic Resonator? I looked those boards over, and right in some sort of timing section there was a missing part that fit exactly. I checked the schematics real hard. It was a capacitor, and in the clock/timing section that drove *everything*.
So that's why everything stopped!. I figured I happened to have the equivelant capacitors I could wire up in parallel and replace it (the leads/legs were completely broken off, so the old one was useless). This time I plugged the board in while bending it so that the cage didn't break off the timing caps in the process like before.
I fired it up again, and ... weee-neee-neee-neee-neee-neee-neee-neee... I was so happy I was praising the Lord because he helped me fix my printer and I got my report done on time. Just before I had to leave. Boy was I thankful.
Before this I had an Anadex DP-9000 or DP-9501 dot-matrix printer. Sorry, no pix today. It was really interesting looking though. It was pretty fast for a dot-matrix. but I think it jammed a lot.
I went to the CP/MUG meetings and did a lot of trading. even demonstrated some programs like MusiCraft, which could play 3-notes simultaneously and did so by doing a reverse-fourier transform to generate the sound. all I needed was a DAC (Digital-to-Analog Converter) on one of the Z80 I/O ports. And I made the ciruitry based on some data sheets.
Got the sound amplifier & speaker from an arcade video game from a friend I think. Or I had it connected to my stereo. Whatever worked. It made organ-like sounds (a Hammond BX3 or CX3 organ's drawbars works on the same basic principle - each drawbar is its own frequency like a fourier transform's harmonics), and I had fun tweaking it and making different timbres.
I also wrote some sound effects programs of my own. I tried making a multiplexed, multi-DAC w/mixer, multi-input-ADC board, but it was never finished (something wrong with it) and too many wire-wrap wires to debug. I can only go so far with circuitry before my brain gives out if I do something wrong. It was a mess of wires in a small space, using wire-wrap sockets, because I had so many bus connections to multiplex. would have been really nice if I could have afforded a circuit board.
In those days I designed the basic structure of a processor that never got off the ground - it's similar to the Java chip that's out now.
When I had that XT clone I was able to get a SCSI controller and a bunch of full-height 5.25" SCSI hard disks and some power supplies to run them. My friend said it sounded like jet engine taking off when they started. I had a future domain controller. the particular model I had had a bug in the firmware... once in a while it would zero out your partition table. I even got a later version of the firmware, but it didn't fix the problem.
When I discovered it was the SCSI controller, I set out to find a different one or get a different hard disk system. ended up with all kinds. MFM, I think an RLL, and an 157MB ESDI disk. The ESDI was the best disk I had in a long time. but I kept running out of disk space. :-)
Eventually when I had Windows 95 I ended up with an HP 1300T SCSI magneto-optical drive (bare), it took 1.3GB or 1.2GB cartridges that when you slid back the metal window (like a huge floppy) darkly looked like a rainbow with spokes because it was hard-sectored. had a couple of cartridges. it was slow, but MO drives could keep data for a LONG time.
you could also get WORM disks for it cheaper if you were only interested in backup/archiving. areas of the disk could go bad on you without warning like a ZIP disk or a flash drive, because blocks/sectors are only good for so many writes (usually about 1000). Then your data is lost. and I did lose a few files because I used it so much.
so I ran scandisk on it, which only makes things worse. :-/ What can you do? These optical drives are mainly used for medical imaging & archiving documents. the cartridges are 1/4" thick and the drive and cartridge is 5.25". the media is flippable - it has a movable write-protect tab on both sides.
Funnies and old folklore
my favorite old laptops
I would have to say that my favorite laptops of all time were 2 models: the Toshiba laptop with (2) 720k floppies and a 640x200 LCD screen, it was a Toshiba T 1200, lasted for 8 hours on a charge (old computers web site says NiCD batteries, but for some reason I remember reading in the manual that they were lead-acid, but since they are cylindrical I think the other authors of the web page are right anyway, looks more like NiCD. but NiCD for 8 hours? I don't think so. Probably lead-acid. which could explain the 15V charging adapter).
There was a HD model, but I didn't get the one with the 20MB HD. DOS in ROM, so "instant-on!".
Very nice 'clack' feel keyboard (like the Cherry brand of keyboards), better than the ones we have today. Nice backlit screen. Had a manual, too. But this one had a german keyboard, which took some getting used to.
But the batteries degrade after 1 year and won't charge anymore (well, it was used). (for restorers: The batteries are rather long versions of D cells that have solder tabs instead of buttons on the end. you may still be able to get these from digikey.com or mouser.com or mcmelectronics.com or newark.com) 1MB Ram. Black and white.
But you can't buy 720K floppies anymore, and floppies are only good for 3-5 years. I think the thing has a serial port and parallel port.
Maybe you can mod this thing to run FreeDOS in ROM instead, if you can get past all the 386 programs, and put 1.44MB floppies and controller on it (well, the thing has a bus IF in the back)?
The other favorite was the ultraportable lightweight TRS-80 model 102/100 (the original laptop). The model 102 has a nice keyboard, and it lasts for 20 hours on a set of AA batteries[!] (I hear it lasts for 8 hours on NiMH).
you can still buy a model 100 or 102 for really cheap. the only useful interface it has to the real world today is its RS-232 port [it's not useful anymore].
it has 32K RAM, which is quite a lot of typing, considering that 1-2k is about 1 page of typewritten text.
The only thing I would change about it is its ram, which I don't think is NVRAM and it is definitely not Flash. I would mod it to at least use NVRAM, hopefully in a socket, if I could get by with that. I never owned one.
It has a 40x8 character display and it does 240x64 graphics. black and white.
You can still buy these things - people want them, so don't throw them away, sell them! It is a useful computer.
You can buy a PC interface cable from club100.org catalog.
check out club100.org - they have lots of software.
Now I think I could just get by with a PDA that can PC-sync.
but it wasn't fun if were doing cobol with punched cards, then I hear debugging was a real pain, and some folk never got to see a mainframe, at the college all they saw was a slot in the wall into which they fed their cards. I got to see the mainframe parts.
that was at portland community college in the 60's, and possibly a lot of other colleges. you would get a core dump printout if your program broke. cobol debugging was all by eye and machine code and 370 assembler (pretty high level for assembler) and jcl (Job Control Language).
as of 2007 they have an object-oriented cobol compiler by MicroFocus for the PC. when one engineer heard that he nearly fell off his chair laughing...
I think for a while some years ago you could even get IBM 370 mainframe-on-a-card for the PC for doing development while the mainframes were still hanging on. I think they are all gone now, replaced by linux MPP machines or whatever they are using now.
the 5MB disk platters were 16" and had about an 8" hole that I think was attached to a hub (platters usually are). they were huge, they came multiple platters in a pack you could install into the dishwasher-sized disk drive. tape drives were refrigerator-sized, if you have ever seen movies or tv shows of old.