Hackaday owes a lot to the hobbyist electronics magazines of yesteryear. Back in the day, Popular Electronics and Radio-Electronics would publish projects and articles about DIY electronics – more or less the same editorial purview we hold today. Some of these projects would become full-fledged products, and you need only look at the Altair for what can happen at this confluence of publishing and engineering.
One of the more popular companies to come out of these hobbyist trade magazines was SWTPC, or Southwest Technical Products Corporation. This was the company that brought one of the first microcomputers to the masses with the SWTPC 6800. This wasn’t just a homebrew microcomputer company – there were Nixie clocks, test gear, and stereo preamplifiers – all things that could easily find a place on the pages of Hackaday today.
This year at the Vintage Computer Festival East, [Michael Holley] brought out the test gear he’s been collecting for the past few decades. These are machines that wouldn’t be out of place on any DIY electronics blog today. This is by all accounts the pre-history of the maker movement.
Interesting items in the exhibit include the Popular Electronics Digi-Vista, a digital clock laden with Nixie tubes from the December 1970 edition. The construction of this clock seems bizarre today – the Nixies are mounted at a right angle to individual boards connected to a backplane. Today, when you can build a Nixie clock as an example of how to lay out a PCB. Back when you could buy traces at Radio Shack, not so much.
But of course the most well-known product from SWTPC was the 6800, a backplane microcomputer based on the Motorola 6800. Combine this with SWTPC’s improvement on the TV Typewriter from a few years before, and you had a complete computer system a year before the Apple I was available. This was the height of consumer technology at the time.
And don’t forget the power amps from SWTP.
I’ve still got my Tiger.
Yup, Tiger .01’s were the benchmark for a very long time. “It’s almost as good as a Tiger” was a high compliment. I still have a bunch, and also, after a few decades of refinement, a few I built that use FETs in the output and are in some sense “better’ – though the only way one could tell is with some rather esoteric test gear that took more time to build and calibrate than it was worth…The (maybe) improved model had DC-4mhz bandwidth and .0027% IMD measured at 21 & 22 khz 50:50 – but sounded the same as far as anyone could tell. The only thing wrong with the originals is you could fry them with a high frequency sine wave at full power, or overheat them if abused too much.
I had an SWTP amplifier back in the day. I lost it in a move in 2003, or I might still be using it. It worked perfectly for many years.
Yes, I grew up in the SWTP era. But you’re wrong about one thing: this wasn’t the “prehistory” of the maker movement – there was a whole maker movement in the 1970s and 80s, that was cut off eventually in the 90s when the personal computer industry cut off access to our computers with a neat little innovation called the Universal Serial Bus. When USB took over and both serial and parallel ports started to disappear from our computers, the “build or buy” decision became easy because the barrier to building grew to “border wall” heights.
There were microcontrollers out there throughout, as Intel with their 8048 series and Motorola with their 6805 and 6811 series fed the embedded market, but the age of ever-faster and ever-cheaper personal computers pushed hobbyist computing almost off the map. And there was still a barrier to entry even in the microcontroller field, with programmers and compilers being expensive, since they were aimed exclusively at the embedded market, to the exclusion of hobbyists. It was only with the emergence of the Microchip, Parallax, and Atmel microcontrollers, and more importantly, free software tools (and reasonably simple hardware) for programming them brought electronics back to the masses, enabling the current maker movement.
It wasn’t just that big industry was crowding us out, though; I think that basically, the people who in previous years would have become electronics hobbyists became computer hobbyists instead, and once you had a personal computer, what other hardware did you need, anyway?
I agree, there was always an electronics maker movement. I was a teen in the 70’s, and R-E was may favorite magazine. I built my ZX-81 from a kit. And done forget people kept working on cars (a friend of mine put a Ford engine in his ’57 Chevy, after making an adapter plate to match to the transmission). I do think that the electronics hobbyist movement dipped in the 90’s, as everyone was getting a grip on SMT parts (It helped kill Heathkit), I agree that PC grabbed lots of interest in the 80’s (you can learn programming on a hand-me-down computer just as well as the newest model, but you could get cast-offs for a song). I also lamented the loss of serial & parallel ports, as well as no peek & poke commands in Visual Basic. But that also represented opportunity to those who could figure out ways around it.
I think what has happen in the 2000’s is that we have seen a convergence of all makers, brought together by the internet (via websites like this) as well as maker spaces.
“I also lamented the loss of serial & parallel ports, as well as no peek & poke commands in Visual Basic.”
Oh, don’t get me started! The disappearance of peek and poke commands wasn’t just Visual Basic; it was anything that ran in protected memory, or more to the point, on any processor from the 80386 or 68020 on. Sure, it was great for business computers, because it meant that a program could crash without taking the whole system down. But it also meant that if you wanted access to hardware ports other than what was commercially available, you would have to write your own drivers for it, and who knew where to even START with that? In the case of Windows 3.1 and up, that meant buying Microsoft C/C++ 5.0 with the Windows SDK that IIRC was in the $800 range. It came with a stack of books literally three feet tall. Kind of daunting for your typical hobbyist.
Ya know, I had two Tiger B’s. Could never get them to stop oscillating above human hearing and blow the output transistors.
:(
Yup, the B’s had issues…Mine too. You had to change the RC network on the gain stage and be real careful how you loaded those…a little fooling with the grounding often helped too. The real winner was the .01 model.
The use of backplanes with plug-in modules wasn’t as odd as it seems. That’s how systems were built in those days, from mainframe computers down to the PDP-8 minicomputers. And PCB layout wasn’t a computer-based thing. You could buy photosensitive copper-clad board, draw your own pattern onto clear acetate, and expose, “develop”, and etch boards using that method. Or you could silkscreen a pattern resist onto plain copper-clad boards and etch that. This meant that making a number of identical modules was relatively easy, and the wired backplane was the logical method of interconnecting them.
Today’s methods, whether sending Gerber files to China (or to OSH Park – thank you Laen) or feeding a CNC router, are more conducive to single-board designs because we can easily replicate a pattern many times and interconnect them on a single PCB.
There were additional reasons that card rack and back-plane computers were built that way.
In development when you inevitably get some part of the circuit wrong, you only have to fix and remake that one part and not the whole thing.
The unit could be built and expanded progressively in stages.
Parts like memory were expensive. Having a generic memory card enabled the purchaser to only buy what they need or can afford at that time. A 64kB computer could have 3 or 4 16kB cards.
Fault diagnostics are much easier if you can swap cards to narrow the fault down to a specific card very quickly.
Upgrades only need upgrade one card, saving the expense of replacing all the other circuitry.
Better part redundancy. If you have many computers they will break down from time to time. Any unit that is not repaired, immediately becomes a parts reserve to maintain the others.
It saves PCB real estate as it is much easier to rout boards individually rather than rout the entire system on one board.
Reduced downtime – replace a faulty card immediately and get the system up even before the card or fault is repaired.
A have looked at a lot of this old technology and at first glance you always think you can make many changes to improve on the design. Quite soon the reality sets in that the older designers were very intelligence people and that things were done the way they were for a reason and there is little or no room for improvement.
If these methods (modular design) were used today, there would be much less waste and land fill and loss of valuable resources like the precious metals used to create these things, less human exposure to the chemical waste and byproducts.
But *most* significantly, if these modular methods were used today hen there would be less profit for the corporation that sell them.
This may be of interest –
https://hackaday.io/project/20132-retro-rack-and-backplane-computer
Those edge connectors on that clock are not very reliable. Tinned or copper is just crap. Those Peterson pins in that 6800 are just as bad. One tiny point of contact. They were the bane of Gulbransen organs, the tinned stuff was in late Hammonds. Turns out with time all the hard to service hard soldered boards have held up longest without service. That senator that held up vacuum tubes and pronounced the ailing air traffic control system was right, it had tubes but the connectors and modular stuff was the problem. I still remember a Quasar TV “with the works in a drawer” as having lotsa connector problems.
The SWTP 6800 was my first computer back in ’76. (I guess I am showing my age here).
Backplanes were common on computers those days. We had them on mainframes when I worked at Sperry Univac. Later I designed memory boards, serial and printer interfaces, and hard disk drive controller boards for DEC LSI-11 (micro versions of the PDP-11) on the Q-bus. As mentioned before it offered an easy upgrade path.
The SWTP 6800 used the SS-50 bus. The Altair 8800 and some other computers used the S-100 bus. SS-50 memory boards were about twice as expensive as S-100 boards. I ended up taking an S-100 memory board and making an adapter to work with the SS-50 and saved about $100. Back then $100 was a lot of money.
” Back in the day, Popular Electronics and Radio-Electronics would publish projects and articles about DIY electronics – more or less the same editorial purview we hold today.”
One feels old when their past is described as “back in the day” and seen with nostalgia and rose-colored glasses.
I’m not sure what’s “rose-colored” about it. It really WAS quite nice that not only did authors like Don Lancaster and Forrest Mims and others develop these circuits and share them with us, but also provided the bare PCBs and either the hard-to-find parts or complete parts kits for many of them, for very low prices. Hell, they even printed full-size PCB masks and drill templates for those of us who wanted to make our own PCBs. Not as convenient as Gerber files, but those, you know, hadn’t been invented yet. This WAS the open-source movement of its time. Early microprocessor projects even included full source code, right there in the pages. You could argue that, well, you did have to pay for the magazine, but that was the cheapest form of distribution at the time.
This was SO much more flexible and affordable than the Heathkit option, which I sometimes think has been over-romanticized – Heathkit kits were often more expensive than their commercially-built equivalents.
” Heathkit kits were often more expensive than their commercially-built equivalents.”
I don’t know if that was always true, certainly by the mid 80s given the influx of Asian manufactured products they seemed way out of whack, but through to late 70s I believe they were very competitive.
It may not have been literally true – Heathkit products had a very good reputation, which may even have been justified. But this was part of their marketing plan – you can’t make money on kits that compete with asian products, so they aimed for the mid-to-high-end markets, where they COULD compete. It was no accident that their flagship amateur radio products were made to look like the Collins S-line. Were they comparable to Collins radios? Not from what I’ve heard, but they were something you could be proud of when you finished building them, at a fraction of the cost. So I think it would be a stretch to call them “competitive”. They made the IMPRESSION of being competitive with the high-end market.
Do you need a little balance to counter the nostalgia?
– Magazines only came monthly at most. There was no daily project article fix. (although.. this did have one advantage… less time reading about others projects means more time to build one’s own)
– Making in general was more expensive than today. For most of the more interesting magazine projects you really needed to be confident in your abilities before starting because the investment just in parts was enough you really didn’t want to mess up!
– Many if not most of the magazine projects would use at least one hard to find part. The author would of course just happen to have a supply! This was how they made their money. In the author’s defense we all need to make a living somehow and the magazines didn’t pay enough to cover it.
– There were no open, real time forums. Well.. maybe if you lived in a big enough city to have a local Prodigy or CompuServe number or a really big BBS. Magazines had columns that answered letters but your question had to be interesting enough to warrant magazine space if you wanted an answer. As for all the open source back and forth design improvements we see in shared designs on the internet today… forget it!
“– Making in general was more expensive than today. For most of the more interesting magazine projects you really needed to be confident in your abilities before starting because the investment just in parts was enough you really didn’t want to mess up!”
I kinda dispute that, it’s relative, basic tools have got really cheap, however, so has manufacturing. Back then manufactured tech items were hugely expensive, just an 19″ Color TV was probably $800-$1000 in todays money. Building one yourself could in fact halve the cost, so even though you were laying out several hundred 2017 dollars in components it was “worth it”. Contrast that today’s situation where you have to be making for the hell of it, because 95% of the time it’s hugely cheaper to just buy one.
“Many if not most of the magazine projects would use at least one hard to find part. The author would of course just happen to have a supply!”
I don’t think this was as mercenary as you imply. I did notice the hard-to-find-parts aspect of many projects, but I think what was going on was that some of the writers got most of their ideas from “new products” articles in the trade magazines, and application notes from the manufacturers. Naturally, each manufacturer wanted to showcase the parts that only THEY were producing, so we’d see all sorts of articles for projects using unijunction transistors, tunnel diodes, PIN diodes, and all kinds of other things that never saw the inside of a Radio Shack or a consumer product.
But can you blame them? Back then, new devices were coming out on a monthly basis, and nobody knew which ones were really going to take off. Several people have commented about the Tiger amp featured in Popular Electronics. I remember THAT being a hard-to-find-parts project, because the article was showcasing a brand-new pair of complementary silicon power transistors, at a time when many solid-state amps depended on using a silicon NPN and a germanium PNP in the output stage because silicon PNPs were unobtanium. Well guess what? That turned out to be THE way to make power amps, and eventually there were many choices of PNP/NPN complementary pairs available.
You guys know that Pop’Tronics and R-E issues are still available on line, right? See
http://americanradiohistory.com/
If anyone’s interested, I’ve created Word files cataloging the issues through about 1978 or so, covering the advent of the hobby computer.
IMO, we hobbyists took a severe blow when manufacturers like SWTPC, MITS, and Heathkit figured out the math: You could offer a hobby computer kit and sell maybe 3000, or you could offer a PC and sell 300,000,000. Tough decision.
I call it the temporal anomylizer, SWTPC’s ambiance synthesizer. A pair of bucket brigade chains with lotsa knobs. You don’t have to run input to it, just the feedback and delta-time/echo knobs and everyone is tripping in it’s sounds. Best Hamfest item ever. None of the op-amps were bypassed or high f limited. Lotsa garbage till I added caps. Also added peak limiter to both channels. Might be similar in those untamed tigers, I remember someone in collage that had similar woes.
Here’s a download link for the schematic and block diagram for the ambiance synthesizer. Get it before the HaD Effect destroys it!
http://www.aronnelson.com/DIYFiles/up/ambience.zip
In action: https://www.youtube.com/watch?v=F_FNmtEM3XM
A nixie clock is my next project, I’ve been putting it on the back burner too long now. I want one badly they look amazing.
https://hackaday.com/wp-content/uploads/2017/04/thmb.jpg
I have the exact same thing somewhere. All these were just before my time, I got started in electronics in the early 80s and had Radio Shack project kits like 50-in-1 to play with. The microlab I got was a gift from someone who is much older than I am and last I heard still at Adobe.
LOL! The first thing I thought was, “Blue LEDs in that decade? Really? Must’ve been retrofitted”, but then I realized that those are actually incandescent indicators!
To add to the nostalgia, I remember when I was around 13 borrowing my dads typewriter to create official looking letters to request samples of some of the earliest ICs from companies like Plessey. They actually sent samples of what were then very expensive devices. I found out about them through Wireless World, a great magazine in the U.K. I think I got some of the earliest logic gates and some op amps that were predecessors to the uA702. My plan was to build a pocket analog computer as a slide rule replacement because, at that time, digital computers were rather large! I was also very proud to have Wireless World publish a letter from me about improving capacitive discharge ignition systems for cars by using an Oscillator based on a logic gate for thr inverter. The circuits at that time were all discrete and this was back when cars used points. It is sooo much easier now.
Heh, I’ve got some of those compendium books of circuits from 70s or 80s, and yah, think I saw a nand gate oscillator CD ignition in one… wonder if it was yours… (They generally had permission from magazine publishers et al to reprint, not sure if they sought yours.)
I’m old enough to remember SWTPC stuff. It was rather pricey, but not outrageous. Compare this with the likes of what Elektor offers today. Their prices are so high for kits, I consider them to be a Rip-Off. And that’s after stripping-off the usurious added cost for the likes of VAT (tax) in the EU, and the fact that Elektor has a legacy physical print-magazine to still support.
I have for sale a very well configured and working software development system complete with two 8 inch floppy disk drives, and 22 floppy disks. It has 8k RAM, a serial port, a parallel port, and a disk interface card.
All the documentation which was accumulated from initial build up to the current spec is included, together with 3 boxes of spare chips. The floppy drives come with their engineering manuals.
There is also a binder full of software listings for many of the programs which are included on the disks with their source code. Both ‘C’ language compiler and Basic interpreter are on the disks.
The system boots up into the “SWTBUG” monitor ROM, from which the Flex disk operating system can be started by typing the single letter ‘Z’, made possible by an included EPROM.
I don’t have a web site or any real idea of how to sell it.
Oops, forgot to say: SWTPC 6800.
Man, I would sure love to own that digi-vista design nixie clock pictured above. I collect units of this vintage and would provide a good home!
There is a 6809 one on ebay:- https://www.ebay.co.uk/itm/South-West-Technical-products-SWTP-computer-with-6809-CPU-assorted-ICs-books/401971606346
But you can buy a new ‘reproduction’ SWTP 6809
https://www.corshamtech.com/product/ss-50-6809-flex-complete-system/
I studied electronics at a “vocational high school” (they still have those?) back in Ohio in 1975 & 6. I had long been into building things as a kid, including designing & etching my own circuit boards, even double sided, using only a schematic. Subscribed to Pop Tronics and a couple other similar mags whose names escape me. Back then, probably the biggest “kit” company was Heathkit, especially popular with ham radio fans (they still have those??? ;-) I built a 2.5 digit nixie tube digital multimeter of theirs which I used extensively. The “.5 digit” was just a vertical neon bulb capable of only displaying “1”! About that time, the whole electronic enthusiast sector seemed to be moving to those absurdly clunky, virtually useless “computer kits” that began appearing. Oh… how neat… a TYPEWRITER… SNORE! I had zero interest in those, but was really into audio. I built 2 SWTPC “Universal Tiger B” power amps (WITH the power meter add on, of course!) a Dynaco PAT 4 preamp (also a kit), and combined them with a high quality single play turntable and a couple JBL loudspeakers. 75 very clean RMS watts/channel was nothing to laugh at in those days. That stereo really “rocked” and was the envy of all my high school buddies! I began listening to Jazz as well. Fell in love with that after hearing Miles Davis’ “Kind of Blue” while experiencing some of the “lousy-by-today’s-standards” cannabis which was around then. Remember all those stems & seeds??? And the idea pot had to be BROWN and “Columbian” to be good??? Kids! ;-) I hated it when SWTPC switched to computers. No more cool audio! Used that setup for years, eventually switching to a Hafler DH-500 (total 510+++ watts!) & Hafler EQ. David Hafler was the co-founder of Dynaco. Hafler also offered kits, but I stumbled across these units used, and factory assembled. I still use the Hafler gear, driving a couple Ohm loudspeakers I actually rescued from a trash bin! No preamp, just the EQ, power amp & speakers, playing high quality digital files (mostly .flac’s) from my PC. Haven’t build any audio in awhile, but NOW I do build my own PC’s, from the MOBO up. Yeah, I’m an old guy, and though I do have an Android “smart” phone, I basically hate it on most levels. Incredibly fiddly interface, and it seems like almost every “app” is incessantly trying to cram ads in your face or otherwise “weasel” something or other. I still prefer a full size keyboard, screen, and my i7 based PC!