I recently dropped in on one of the Vintage Computer Festival events, and it made me think about why people — including myself — are fascinated with old computer technology. In my case, I lived through a lot of it, and many of the people milling around at VCF did too, so it could just be nostalgia. But there were also young people there.
Out of curiosity, I asked people about the appeal of the old computers on display there. Overwhelmingly, the answer was: you can understand the whole system readily. Imagine how long it would take you to learn all the hardware and software details of your current desktop computer CPU. Then add your GPU, the mass storage controllers, and your network interface. I don’t mean knowing the part numbers, specs, and other trivialities. I mean being able to program, repair, and even enhance it.
Keep It Simple
Consider the relative complexity. An Intel 8080 CPU has 6,000 transistors and 76 instructions. An i7-940 has around 731 million transistors! It has multiple instruction sets and extensions so much so it is hard to get an exact count. There are multiple cores and complex memory caches.
Even the 8080 was complex compared to some of the old CPUs. The RCA 1802 was famous for being fully static and having a DMA system so you could enter data by hand without needing to burn an EPROM to get it started. The 6502 had about 3,500 transistors. It was deliberately made simple to gain speed and save costs. Both the 1802 and 6502 had simpler power and clock requirements, too.
Could these little machines do what our computers today can do? Not really, but they can do a lot. What’s more, you can readily learn to program them, design with them, and build them.
Retro Doesn’t Have to Mean Old
This leads to an interesting effect. I think people like these machines because they are simple, not because they are old. That means that replicas are valued more than in some other areas. A replica of an old book or car is of limited interest. But a replica Altair 8800? There have been many, and they are usually in high demand, especially since the actual computer is frightfully expensive and finicky to keep running today. Then there are the retro-style machines. The Hackaday Supercon 2022 Badge comes to mind. It isn’t an old computer, but it seems like one. Many people had a great time learning about the badge and hacking it.
Get Started
If you haven’t gotten the retrocomputing bug yet, there are a lot of ways to get started. First, you can emulate almost any old computer you like on a PC or even, sometimes, in a browser. This usually costs nothing, and you can get a good flavor of how a machine works. In addition, the emulators usually are decked-out systems with everything included. They also frequently include debuggers that anyone who used the real machine would have killed for back in the day. Think you’ll miss the front panel switches and lights? Maybe not. You can find emulation for everything from old mainframes, minicomputers, microcontroller boards, and pre-PCs like the venerable TRS-80.
The next step up is to build your emulator with something like a Raspberry Pi or an Arduino. You can also find FPGA recreations of old processors and computers. These machines can get very fancy or be very simple. It is hard to explain, but there is a big difference between booting CP/M in a window on your PC and booting it on a box with blinking lights on it. Doesn’t make sense, but it is still true.
When you are ready to move on from that, you can still buy many of these old processors either from inventories still hanging around or, in some cases, they are still made for some reason. You often have to blend new parts with old ones, but you can have the satisfaction of running a real CPU. Sometimes people try to make near-perfect recreations of old hardware. Sometimes they just shove all the support hardware into an Arduino for a cheap and easy build. Either way, you are still running on the authentic CPU.
Finally, you can find old machines and either spend a fortune or spend a lot of time repairing them. Maybe you’ll even do both! At least the repair work is feasible and enjoyable. Some computers were wildly popular and still easy to find. Others are very rare — you even occasionally see one that is one-of-a-kind. Those can take some detective work.
Do You?
Do you retrocompute? If not, why not? If you do, why? Is there old software you can’t live without? Is there a certain game that you like better than the modern counterparts? Do you enjoy being able to interface hardware to a machine without developing a PCIe interface? Or is it simply nostalgia? After all, people collect old cars, old radios, and old books. Maybe you don’t really need a reason. Let us know what you think in the comments.
Just because.
Oh and no nagging OS message from a corporate overlord, or forced reboot for unplanned emergency updates. Or ads.
If you run RTEMS on x86 you also don’t get ads. No need to use outdated and insecure CPUs.
Because it brings me back to a day when computing was fun, new, exciting, and made me feel optimism for the future.
Modern computing today is sterile, soulless, boring, and too pretty. It’s the same reason why when I’m sad, I’ll go to archive.org and surf the late 90s internet.
Well OK, but I totally disagree with the sterile, soulless, boring, and too pretty. We now have very supierior Programms, infinitely better HW and the advancements of AI will make things very interesting. Personally I owul love a PDP 8/ or 11 but I think retor computing ist , well, Retro.
You must be talking about that Redmond OS…. I have no such issues with the Linux OS I am running on all my systems :) .
And yes … Just because is a good answer. I fire up Oscar’s 1/3 size PDP 11/70 front panel now and then to load a retro OS and play around with it. Just because. When I retire I’ll probably do more ‘retro’ stuff as it can be fun. Keep my modern Linux workstation around though too. A person can do both!
I use that Redmond OS, and for some reason there are no random crashes or screens of death. Magic!
And whenever I tried to use Linux, something either didn’t work at all, or had driver/configuration issues. So YMMV.
I admit I’ve not seen rashes or screens of death on the M$ boxes at work for several years now. The company I work for is a Redmond shop and ‘pays’ accordingly.. But every week I am forced to do a reboot for updates. Grrr. At home, I only update my systems on my schedule. Nothing comes down automatically. No ads. I do hear complaints from co-workers who still have that Redmond OS on their computers at home though :) . I just smile. No aggravations here!
Those driver/config issues in Linux, in my experience, have become a thing of the past (yes, I remember those days when there were problems, been there). The latest Ryzen powered laptop I bought (last year), I wiped the installed OS, and installed Linux. Everything worked out of the box. Upgraded Ram to 32GB, installed another 2TB SSD, runs beautifully. Nice! Same with the previous laptop. My home desktops are all Ryzen 5000 series powered with NVidia graphics cards. Solid as a rock. The home server is using a Ryzen 2400G and runs 24×7 without a hiccup serving up files. Only time it goes down, is if we are on vacation, I shut it down. All the hobby SBCs I use work fine with 64bit and 32bit PI OS and Debian.
Anyway, YMMV but my experience has been smooooooth sailing for quite a few years now! Loving it. Pleasantly boring so to speak :) .
I’ve been using Linux since Slackware was downloaded on a stack of floppies BTW. Around Windows 7 (best/last OS I’d purchased), I moved to Linux for all home systems for good. The pocket-book bleeding went away.
…and no web browsers you can’t uninstall without a third-party tool!
I like to play new games for Commodore 64, they’re better than the ones created while the machine was still being produced. Started learning assembly, and yeah it’s hard, but persistence is all.
That’s because in the day the games were made by workers, today they’re made by fans.
80s game development was a largely individual exercise. These “workers” were artisans with almost complete autonomy. The history of Activision is good reading in this area.
If games are better now it’s due to more available time, advances in human knowledge and better dev tools. 8bitguy spent 3 years on Petscii Robots – that’s about how long the video game boom of the 80s lasted. HSW had about 4 weeks to build ET. (Much respect to both these guys!)
No disrespect to the 80’s programmers but when somebody does something for the pure love of it it tends to show.
Not only. A modern dev environment with cross-compilers, emulation, step-by-step debugging makes it way easier to focus on the logic and content, even if you’re working in pure assembly code.
Not my thing, but I did notice two different estate sales in north Atlanta this week, one had an Amiga 1000 and the other had a commodore 128 and I thought to myself, “I wish I knew what to post that, so someone in retro computing in atl could go grab them…”. Not sure if links are allowed, just Google estate sales (.net), most start on Friday morning around here
‘you can understand the whole system readily’ exactly! I’d love to see a computer you could build for under £100 of discrete components, I think it would make a fantastic teaching aid in schools and colleges. I keep trying to build this (https://blinkingcomputer.org) but really haven’t got far in over 30 years of work…
This!
I wanted to build an emulator of Odra 1305 using CPLD, but apparently these are now more obsolete than 7400 series logic. I also would love to make my own computer using TTL or CMOS logic chips.
Z80, PIO chip, EEPROM, SRAM, UART, ~40 LEDs, 5 or 6 7400 serie buffers/latches, an address decoder (74LS138 and some AND gates) and a serial to USB converter. Connector that goes to some io pins on an arduino to program the eeprom the first time with a boot loader and poof! Done. It obviously could be greatly simplified, but that give you the idea.
Replacing Z80 with your own custom CPU would be more fun. And quite educational, too. Or using some less known CPU, especially something obscure, would make such a design more unique. The only problem is the fact that obscure or uncommon parts can get pricey, so one option would be to emulate the weird architecture with logic chips.
That’s what I’m doing. I have the instruction set so built the software side out to FORTH and bare bones Python. I have a novel two transistor DRAM cell (uses the capacitance of MOSFET) and I’m working on a 5 transistor per bit ripple carry adder that has no transistor switching in the ripple carry stage. There will be a LED for everything and it’ll step slow enough so you can see it all work.
There are Microcontrollers that include several peripherals in a single die, but they don’t give quite the same experience.
Being hard wired, you get no control over I/O addressing. You only get to specify which peripherals, and RAM size, when you pick a part number – and assume that part is available in stock. Then there’s the PCB design and fabrication skills required, especially for BGA devices. SMD takes up less space, but the ability to populate a board, sections at a time, can be an issue if you use a solder reflow oven.
Oh, and is that new chip 5V tolerant? Just about everything in older designs used a single 5V supply.
Have worked with a number of system-on-chip microcontrollers that had more on chip peripherals than the package had pins for, so as part of the cold start code you get to poke values into an “I/O multiplexer”, to choose which functions go on which pins. Feels weird coming from the days when designing a system you only had to concern yourself with what to put *in*
I’m keeping an eye on a couple – https://feertech.com/microbeast/index.html and https://rc2014.co.uk/ but haven’t got the time to get either at the moment.
FYI, RC2014 boards/kits are available on Tindie.
Because 20th century technology was still exciting, no black box? 🙂
As I’ve said very publicly – The reason I am into ‘retrocomputing’ is the community and people I encounter that have the same involvement as I do. I don’t care what aspect anyone focuses on within retrocomputing, just that they are enjoying themselves. This also means that they aren’t required to have identical experiences (and age) or like the same systems that I do. It’s the people for me, not the inanimate electronics.
Retro computing groupies!
It’s amazing!
I tell them ‘I’ve got one of the first 200 pre-release Amigas and the first dev kit’ and their panties just fall off.
Great. Niw I’m naked.
I can understand bare metal 68000 assembly progamming; Microsoft Visual C++ not so much.
I do understand the appeal of retro computing compared to a “normal, modern PC”.
What I find much harder to understand, is the appeal to retro computing compared to some kind of microcontroller board. A few years ago I helped with conversion of an 68000 SBC from eagle to KiCad, and it had a lot IC’s and ran on 16MHz or so. And all that fits in a single microcontroller. Other common projects is to do something with an Z80, and then add an atmega328 as “peripheral”. That Atmega has more of almost anything that a Z80 has (you’d need a bit bigger uC to beat it on Ram/ Flash and the old home computers did have a video output) (RP2040 can do quite capable DVI output, as show cased on Hackaday several times).
But there are so many uC’s that are still small enough to understand, while capable enough to do some real work. I know uC’s sometimes run emulators, but why not do it more natively? For example port the Acorn Archimedes OS to an STM32. Add a keyboard and a small TFT and you can do native development on a handheld and battery powered platform. STM32 can also load software (for example uSD) into RAM and then execute it. And they go up to at least 768kB of RAM, and that should be more RAM than anyone ever needs :). Plenty of room for programs, while keeping the “bios” and “OS” in Flash. They have built-in USB, and sometimes Ethernet too, along with plenty of other niceties.
I guess the biggest hurdle is the lack of a preexisting community, combined with quite a lot of effort needed to get a project like that started. I do not have the capability to start such a project myself, but if it existed, combined with a “healthy community”, I would definitely be very interested and attempt to contribute too.
Since this 2019 pandemic event microcontrollers became absurdly expensive. In 2021 I sold my well used STM32F1 (ZL27ARM) dev board for $170. I bought it new in 2014 when I was a student, it cost about $40 bundled with book about STM32 chips. Atmega 328 cost $2.5 in single quantities in 2013. In 2021 I desoldered it from my old linefollower robot and sold it on allegro (polish e-bay) for $20. I never thought someone would pay this much for used 8-bit MCU.
We’re living in a clown world now with this economic meltdown and I decided to pick up gaming and streaming hobby instead of electronics. Last year I bought my new gaming rig on credit and within 6 months it paid for itself with YT and Patreon.
I too prefer discrete logic parts over microcontrollers.
They don’t contain buggy software, they don’t have patent issues, they are free to be cloned by third-party manufacturers.
I’m thinking of 4000 series CMOS and 74 series TTL.
Another “thing” is the trouble with GALs/GALs of the past.
They had their protect bit enabled often and their programming couldn’t be saved. This made PAL/GAL chips a nightmare for hardware restoration.
Many retro/vintage electronics people still remember this and don’t use them anymore, if possible.
Microcontrollers are bearing the same risk.
There are people who know how to decap those old chips and disable the protection stuff while still being able to read the data out. Or in many cases for simpler chips that contain no internal state (or only very simple internal state) you can brute force the internal logic just by feeding it inputs and seeing what the outputs look like.
All the above. I like the nostalgia. I grew up on Commodore, TRS-80, Apple II etc… The basic programming is extremely easy to understand and even the various monitor programs for machine language programming. I have also learned many things this second time around that I was too young to learn then and modern computing… Well I don’t know if its the complexity more or the locked down nature of modern computing in the name of profit. But as much as you wouldn’t teach a kindergartener differential equations without basic number skills and elementary math. I can see why learning on systems like that to get a foundation on modern systems makes a lot of sense.
While at university, a friend and I created a commercial BASIC for the RCA 1802. It was written in assembly language and hand-translated into machine code. I can’t remember how many times we had to enter kilobytes of code by hand until we had a proper running system, but it was tedious to say the least.
When it was finished, we burned it into an EPROM and licensed it to an 1802 distributor. I can’t say we got rich from the sales but it did create some welcome extra income.
It was the first time I had income form programming but it whet my appetite for computer jobs, and the same year I started a consultancy where I ended up writing everything from the software for a heart monitor via video firmware for a Eurobus-based CP/M system (which also included configuring CP/M to run on the system) to writing a BIOS for a PC clone. Oh, and a lift/elevator control system based on the Motorola MC14500 1-bit computer. Yup, ONE BIT computer.
Also because nostalgy. I got my first computer late, with my first paychecks at my first job after university. But before that I played as a child on relatives ZX Spectrums or local clones. Not enough though. I also attended in the vacation between high school and university a BASIC course on another local clone of the same ZX.
Now I have bought two original ZX Spectrums.
But I lack time to enjoy them (and one crashed).
So, besides being easy to understand and program, they are the reminders of our glory days.
Unfortunately not everybody has a temple for his computers and other machines that defined us and our dreams.
Long live the skull and wrenches!
the thing for me is that “old software” is not bad software, it’s perfectly fine. Sure, an old image editor isn’t going to have all the neat features that newer image editors do, but that’s fine. Old software is imho a work of art, that deserves to be appreciated. Just because the machine they were programmed for may be old, doesn’t make the software any less useful.
I agree with that. I’m still using some Windows 3 era software, too.
Why not, after all? If it’s still useful and doesn’t cause any trouble, it’s okay.
With WineVDM, many of these 16-Bit oldies can run directly on Windows x64 desktop, for example.
Why? Because I never stopped using these machines.
+1
(for me that extends to pretty much all electronics. I never stopped using tapes, CRT TVs, VCRs, old computers and old software.)
Slightly different approach from my side:
I sold my whole CD collection about two years before they became obsolete.
Obsolete¿
Well, to each his own. They are really cheap at 2nd hand sources, and I keep a portable player in a Faraday Cage in the event of a zombie apocalypse.
I didn’t know zombies liked music.
Yes that is exactly what I mean. Now they are not worth half a dime a dozen. That is the meaning of obsolete.
When I was young I had to work for 6 to 8 hours to earn enough to buy a single CD.
Pink Floyd called it “Riding the gravy train”. Back then (either LP, CD, whatever) it was about 50ct for production, 50ct for the artist and the rest is for the big fat man with the cigar.
The final nail in the coffin was Napster. Napster was trying to be very reasonable, and when they got sued out of existence I never paid a dime for music anymore.
An 8086 running DOS 5.5 was the last computer I could claim to understand. Wish I still had it (and all my work on it). “Repair”? Did the stores all close and no one told me?
Let’s say V30. 🙂 The i808x was so slow it hurt.
I’m into history of technology and computers. For years I wanted to build either a replica/emulation of older computer, or some part of one, like delay line memory. Recently I was thinking of making an electronic lock operated with 3D printed “punched” cards. I even have a clever design in mind that makes it impossible to copy by taking a photo…
I like the weird architectures that old computers had, home computers in the 70s and 80s were like the Cambrian explosion, lots of weird experiments because nobody knew where things were going, and having one today is like having a (commodore) pet trilobite.
Things with 18 or 36 bit words seemed especially weird. Also seemed weird, while “buzzing out” a TRS-80 (to make hand-drawn schematics, that later got lost !) that there were so many 74xx chips that had 3’s of things in them, when there were perfectly good devices in only slightly bigger packages with 4’s of things. Eventually the light bulb went on, the dots were connected, it became apparent that if you were building a CPU from scratch in a world of 3-ish MSI devices, you could easily end up with an 18 bit computer. Not sure if that really was what happened, but it did seem very possible to construct something like a 6502 or 8080 but with 9 bits, which I wanted to do for the longest time. Well, maybe that’s part of the appeal of retro tinkering, to explore strange parallel universes of possibilities.
Huh. I use here, sometimes a rescue. It is a TRS-80 Model 102. I use it to talk to a Raspberry PI OS, on a variety of devices from across the water. I use it to talk to Slackware Linux from much earlier, itself installed inside a copy of VirtualPC, itself running on a laptop that I’d gotten at VCF East several years earlier. Oh and it runs Windows Seven-SP1.
I bought an Amiga 4000 for my 30th birthday because I could only dream of it when I was a student.
21 years later it’s still in my attic, Never booted up. I’m not sure I’ll find the time, and courage, to fix it before I retire, but I have my dream computer :)
I bought a mini-C64 just for fun recently. The o’ basic prompt brought back lots of memorys.
38911 of them? =P
As mentioned, programming in assembly on the old machines was not an issue… just time consuming, but also fun at the same time. The only real negative was the eprom erase/burning cycles. However not having access to a logic analyzer connected to the data/address bus and control signals can certainly consume a lot of debugging time. These days once can buy a really nice older HP high end logic analyzer for cheap. These were $20K+ at the time which would be multiples of that now. I think I saw one locally going for a few hundred $ with pods. Ah, the good old days where things were simple and one had to be very innovative and resourceful to create products with very limited memory and CPU resources… unlike today.
On a side note… I still have my Commodore Pet 2001 with chicklet keyboard and it still runs and loads programs… via a cassette no less! I was thinking of selling it, but it is simply too cool and brings back some great memories in my early days of “trying” to write programs. It is funny to think back when there was no Internet and/or access to virtually unlimited info and yet one could find various books and other resources without too much difficulty. All it took was reading various monthly electronic or other publications (BYTE) and ordering items via mail. The tech kids of today have it far, far easier than we did and yet I find it odd that the rate of innovation is not as high as it was during the early stages of the computing and embedded systems technologies.
I bit tougher to innovate I think because a lot of areas are already covered. Thinking software as that is my line… Spreadsheet? Been done. Word Processor? Been done. Database? been done. Cad system? Done. About anything you can think of has already been addressed that people use. Sure they can be refined to make better, but to come out with a ‘new’ never seen before product or idea that is going to change the world? It is not like the 70s and 80s or even 90s heyday when around every corner you could spot something cool and new — probably out of someones garage or workshop!
” It is funny to think back when there was no Internet and/or access to virtually unlimited info and yet one could find various books and other resources without too much difficulty.”
I don’t mean to disagree, that’s how probably how it was for most laymen back in the 80s.
However, there already were a online services and databases for the more informed people.
Here in Europe, we had national services like Minitel (France), Prestel (UK) or BTX (Germany, Swiss, Austria etc).
Other countries had similar services, I believe.
America had CompuServe from 1983 onwards, I believe. And Promenade/Quantum-Link (Q-Link) from the mid-80s onwards.
Then there was Packet-Radio on ham radio, which was in wide use since roughly 1985..
It was pretty much like dialing into BBSes or mailbox systems on the telephone system, except that it was via radio. There were even a few satellites carrying Packet-Radio transponders.
Anyway, I don’t mean to sound like a smart-ass here, I know that the digital world was a quiter. But I can’t hide what I know did exist.
And I didn’t even mention the X.25 networks that served as an invisible infrastructure for a lot of things (like banks, ATMs, airports, power plants etc.)
FujiNet…. Atari, CC65.
Wow, this is quite a neat question.
For me I was in a unique position at the time, it was 1993-1994 and the idea a child could actually own their computer was quite preposterous. My family wasn’t rich, but we weren’t poor either. I was also thankful and very lucky that my parents were always supportive of my hobbies and learning, gave me unconditional love, and spoiled me as much as was possible with my interests.
A used 386 or 486 powered computer would run you thousands of dollars and as a result was quite out of me and my families budget. Where I lived did not have a hot spring of old computers to pick from, nor anywhere you could buy new parts. The used market was overinflated and pricey too. From my very first experience touching a computer in pre-school (An Apple ][ Plus btw), and being accused of breaking it (I did no such thing, however I might have gotten it into a test mode or caused AppleBasic to scroll some garbage across the screen), put in the corner, my name on the chalkboard, and not getting my snack that day I KNEW I had to have my own computer. There was no way I could learn what I wanted to learn and do what I wanted to do with adults constantly over my shoulder afraid I was going to break their equipment, and I didn’t want that kind of pressure either. I knew from the keyboard alone I couldn’t really cause damage to the computer itself I couldn’t undo, but I really also wanted to learn the hardware side of it, since that’s what excited me the most. It was a magical black box and I REALLY wanted to understand how it worked and how I could tweak them and make them better.
My first choice would have been a ][ just like the one I got disciplined over, but Apples were expensive, even Apple ][‘s went for more money than I could possibly earn at my age. I learned as much as I could using my elementary school’s computers and the volunteer who became sort of a mentor to me and helped me out when I was in over my head. I became the goto person at my school for any computer problems! It felt wonderful to be able to help people out and to learn at the same time.
I mowed lawns, I did chores, I fixed stuff around the house, I fixed computers for friends and family, I did everything I could to earn money. With everything I made, my father agreed to take me to a local computer swap meet. I told him I was going to buy a computer there no matter what, and he smirked, chuckled, and gave me encouragement, even though he knew it was a ice cube’s chance in hell I’d manage to get anything on my budget. I found a guy with stacks of IBM 5150 PC’s for sale, and we agreed upon the price of $42 for one complete system… that is an IBM 5150 with 128KB of ram, 2 360k floppy drives, a monochrome video card, a green screen monochrome monitor, a clicky clacky Model F keyboard, and a 5 1/4″ floppy which he said ‘must have gotten too close to his IBM DOS 3.3 master disks, but it was just a blank floppy nudge nudge, wink wink.
I took this underpowered monstrosity home, played with DOS, played with GW-BASIC, and started taking it apart, putting it together, learning by trial and error what worked and what didn’t. We had a computer shop that had a bathtub full of junk computer parts at fixed costs, and I got good at snagging good things to upgrade my computer with and other junk I could build or upgrade other computers with. I picked up a “Pocket PCRef”, made by the same people as the ‘little black book’ the ‘Pocket Ref, and it gave me specs to hard drives, floppy drives, numbers to companies, which companies didn’t exist anymore, beep codes, bios codes, what the DIP switches on my IBM 5150’s motherboard did, and a lot of other very important info I could always carry with me, whether it be when I was scrounging parts or doing work for someone on their computer. Everything back then was hands on and had to be manually configured, so having that little book with me was a livesaver. I didn’t have the internet either back then, so I was quite literally on my own to lean. By the time it was time for my to build my first computer, I was 9 years old, I already had the IBM 5150 I bought at age 7 for myself upgraded to 640KB RAM, a 42MB ST251 MFM hard drive, CGA graphics and monitor, 2 half height 360KB TEAC floppy drives, and as many crazy interface cards as I dared pack it with. Putting a 3 1/2″ 720KB floppy drive in it forever gave me the hatred of DD 3 1/2″ disks and the incompatibilities with HD 3 1/2″ drives, but that’s a story for another time.
So to answer the question, I went “Retro” out of necessity, and I really went in depth not on a Macintosh or a Apple ][, but a original IBM PC because that’s what I could snag and get a hold of. Unlike today they were considered obsolete and best used as a door stop. The guy who got $42 from me probably thought he got the better end of the bargain to be honest! I collected so much junk over the years because people wanted it gone, and I thought it was neat. I got every computer from that era I ever wanted to play with simply because it was cheap or free to do so, and at some point my little hobby became popular and even acceptable. Although I always wanted a mainframe, the gold reclamation people always had bigger pockets at auctions, and I think my parents would have legitimately had a very long conversation with me should I have actually won any I bid on.
That being said I like showing the old stuff off to anyone who wants to play with it because it’s fun. Kids find the old machines fascinating, fun, and easier to approach than today’s high level systems since so much of the hardware is easier to access and understand what’s going on, and older folks remember actually having to rely on the old tech to get work done. The old machines were fun, unique, and gives you a tight box to work in, with lots of limitations built in you have to get creative to do things with. Working inside a box is sometimes the best way to spark creativity, or learn what is possible with today’s computers if you take mind of your resources. I always tell those who want to play with computers “Give me a lever long enough and a fulcrum on which to place it, and I shall move the world.” The long enough lever, as well as the fulcrum on which to place it are today’s computers, but only if you can learn how to properly utilize them. It’s always been my opinion that old tech teaches you the potential of new tech, and new tech can be used to make old tech do tricks never conceived of by the machine’s original designers.
This has been my hobby for a very long time, and I wouldn’t have it any other way. I will say my little threadripper workstation is the first machine I built had which I didn’t have to overclock and tweak every little thing about, but I think that just means I am getting old. I am looking forward to seeing what the upcoming generations do to run circles around us and do amazing things with the tech. If I can help them along the way by keeping my old machines working well enough and with enough of their funny quirks just as they were when I was growing up, I wish to do that as long as possible. I don’t think I could get in the hobby now though as the barrier to entry is far too expensive, so I try to keep a supply of older parts and machines I can give to those who are most interested to learn.
I don’t retro compute in any serious way, but my interest is mostly historical and cultural. These were the machines that shaped a culture.
I’m interested in how different levels of tech affect the user, and if the tech of the time had anything to do with the fact so many remember the 90s and early 2000s as being a happy time.
I’m also interested in the psychology of why people were so quick to abandon optimizing whatsoever, long before tech could handle it(I’m pretty much fine with electron now!).
And why nobody seems excited about modern tech. Many even hate it and want zero tech.
I like retro tech for the same reason I like oil lamps. Living history I’m glad we don’t depend on, and also glad to have enough privilege to be able to preserve these this as cultural artifacts.
Thank you for sharing your story with us, it’s very fascinating to read. Also my deepest sympathy for your struggling with those adults/teachers, I think you’re a really fine being and didn’t do anything wrong back then. More than often, these people forget that the students/kids are important and that the equipment is secondary. Also kudos for tinkering with the stuff at young age and learning from your own experience. That’s the spirit! 🙂
Simplicity is a big one for me, but also efficiency. It seems wasteful to use gigabytes of RAM and billions of transistors to write a document when an 8 bit micro can do the same job.
Learning to code in such meager environments really helps build your fundamentals, which can be applied back to your day job on modern hardware. This old geezer just checked in some lightning fast Java code that’s about 10x smaller than what our milennial contractors write ;)
Do you retrocompute? If not, why not?
Not anymore. Too many hipsters and youtube content-creators drove the prices of these vintage machines to stupidly insane prices. When a sub 1Mhz Commodore goes for the same price as a pc core i7, it’s time to move on and download an emulator.
You’re not quite fair here, I think.
Did it cross your mind that the prices are high, simply because the vintage computers had been dumped over and over since the 1980s?
Dude, we’re living in the 2020s, 40 years away from the 80s. Times have changed.
There are not much vintage computers left to begin with, after such a long time. 😔
The “hipsters” and YT dudes we often see are simply people who have an emotional bond with vintage computing, I think. They’re the ones who care.
I think it’s shortsighted to blame them, just perhaps due to feeling a bit jealous that they own these computers and we don’t.
I mean, it’s not their fault; It’s not right to blame them just because they either do successfully restore old computers or because they didn’t throw theirs away in first place like the majority did.
There are no modern monochrome monitors (except e-ink).
So if you like to see a monochrome picture “natively”, you need an old monochrome CRT monitor or a black/white TV.
That’s something that’s being forgotten, I think.
Modern TFT/LCD monitors can’t do pure monochrome, merely RGB.
However, the human eye also has separate photoreceptors (rods) for gray/brightness.
That’s why our vision can turn monochromatic in the night,
so we can still make out something under difficult situations.
I think those TFTs used in resin printers would be good as high-resolution, greyscale screens, with the proper backlight added.
E-ink and monochrome OLED’s – like the SSD1306 I use for the 65uino. I wonder if you can get bigger monochrome OLEDs.. Monochrome LCD’s probably won’t count as “modern” :)
But honestly, I’m thrilled every time I turn on my SBC’s green channel only “VGA” output.
Not quite the same as green phosphor … but I let myself be fooled.
Yes! All my HaD.io projects. Because making a hobby feel like time travel is fun! Or at the very least imagining a world where we stuck with DIPs.
And 8 data bus lines are just more manageable. Learning something is fun when you can actually feel some progress with less effort – you can get a 6502 running on a breadboard faster than you can setup the STM32 toolchain.
And keeping things out of the landfills is more fun than filling a closet with new Nucleo boards. To me at least :)
Nostalgia for the long line of still in mint condition classic PCs I simply threw away, something I now deeply regret, and nostalgia for the then entirely new fascination with machines which were simple enough to learn down to their minute details, unlike now.
I don’t have to give much thought as to why I am interested in the things that I am interested in. The reasons are basically self-evident. What I think about more, and what gives me the s%$#s when it comes to mingling with like-minded enthusiasts, is the peer pressure that I am invariably put under to mutually affirm a boring narrative of self-justification.
Constructively contributing to the subject at hand in a genuine way, especially if individualistically, always:
1) Makes you a target of envy and ostracization to a degree that is never absent.
2) Takes a backseat to the group preference to mostly spend its time circle-jerking about just how frigging awesome it is to be part of a group interested in the subject.
Like, as if, on the flip-side of the coin, constantly encountering non-like-minded individuals who for some reason think that you owe them an explanation for why you might be even remotely passionate about something that they think is uninteresting or just plain boring isn’t bad enough already.
It was nostalgia for me with the PlasMa project…
http://www.philizound.co.uk/plasma/plasma.html
I wanted to rekindle that old hands-on key-switch programming and see the inner workings in action, but there was no way I could afford or accommodate real old hardware. An emulation with some basic hardware made it possible.
I can sort of say I do it for a job.
I program C and a little C++ on RTOS’s for cellular communications platforms.
As the silicon gets smaller, and a little faster, I find myself doing the same thing on what is turning into grains of sand. It was a VAX 11/750 back in the day, now it’s an ARM M7. 16MB of RAM tends to keep the code concise.
Yes OK I’ve done my time on ZX Speccy and similar back in the day, but that just keeps my code tight and purposeful.
Yes using old thin client from 2013 and old 2008 portable with Debian. ;-)
As an art form, its your classic trend following fad fans vs historical forms. No different than music, for example.
Must be why I listen to Classical Music while working. Including programming those older fellows.
So that’s why I’m listening to Glenn Miller music!
The picture is of my S-100 exhibit from VCF East in 2022. I got into the hobby partially from nostalgia about the older systems I used to use, but also from a desire to own the systems that I couldn’t afford back in the day. I enjoy showing them off at VCF events, and volunteering at various museums to keep the old tech running.
It is indeed. And I got into the hobby because I missed out on the chance to own in a sense the bigger examples of the DEC family members that are also shown there.
I used to love CP/M and my old Z80 system. I still have my original TRS80 model 100 in the closet. Remember one of the great thing about these systems is the lack of malware being written for them as well as virus……Like a fine wine the get better with age.
Yes, i have fun with older systems.
PDP8s, PDP11, VAX, Alpha ( BIG IRON racksized class ), also SGIs ( up to 4 rack sized systems Origin2000 and Altix4700 ), also SUNs from smaller 3/50 to E6k5 and some other Unix irons.
They make more fun then todays systems with boring standard hardware.
Also one sees how technical solutions were developed and some today very common things like f.e. SSDs have been in the enterprise systems decades ago and now are mainstream.
Also f.e. Inmemory Databases have been around since f.e. Oracle 7 with big Alphaservers with 28GB ram in the 90s and some years ago they have been advertised like reinventing the wheel — funny…
My computers from yesteryear never got compromised by malicious software when I connected to a BBS over a dial-up modem. So computing was also a rather safe undertaking back in the day. Now a days, every which way you turn with your computing activities, there’s always a potential threat from malicious actors that has to be actively and constantly guarded against, and constant privacy invasion going on as perpetrated by the biggest technology companies.
And there is no point in a computing device today that is not connected to the Internet. It’s looked upon as basically a worthless paperweight.
Which means in a way that computers have long since ceased to be the main attraction – interconnectedness is what everyone seeks in their technology experience, and so the “connected experience” has transcended the “computing experience”. This very website and comment thread, of course, speaks to that as example.
That extremely sophisticated derivatives of those early computer microprocessors are what facilitates this is now merely incidental to most people. Only hard core gamers really care about advances in CPU and GPU capabilities and performance specs. For the masses, that has long since ceased to be a “thing” that registers any attention whatsoever.
Hell, even an hdd controller board if orders of magnitude more complex as a warehouse full of retrocomouters.
I recently dicked around with a Seahate F3 drive (and even that is 10+ years old), and it‘s mind boggling complex…
Every singe part in modern conputers is a whole computer in itself…