Remembering The 1984 Unix PC. Why Did It Fail So Hard? (youtu.be) 62
"I love these machines," writes long-time Slashdot reader Shayde:
I was super-active in the Unix-PC Usenet groups back in the 90s... We hacked the hell out of them. They were small, sexy, and... they ran Unix!
Unfortunately, they were a commercial failure. There were so many things wrong with them — not just stuff that broke, but the baseline configuration was nigh on worthless. I recently was able to get another machine and got it up and running (with a few hiccups). I whipped up a video showing all the cool things it can do, but also running through what went wrong and why it ultimately failed.
The video shows the ancient green-on-black screen of 1984's AT&T Unix PC (with the OS running on a silicon drive emulation). The original machine had 512K of memory and a 10-megabyte hard drive described as slow, failure-prone, and noisy. There's also a drive for inserting floppy disks, and a separate MS-DOS board (with its own CPU) that could be plugged into the expansion slot — but the device was "remarkably heavy," weighing in aqt 40 pounds
See the strange 1984 mouse, and its keyboard with both a Return key and a separate Enter key. There's even plug-in ports for phone landlines. "It looked great," Shayde says in the video, showing off its Spirograph demo and '80s-era games like Pong, Conway's Game of Life, GNU Chess, "Trk", and NetHack. But besides slow startup times, it was expensive — in today's dollars, it would've cost roughly $15,000 — and suffered from Unix's lack of spreadsheets, word processing software and other office productivity tools at the time. At that price the Unix PCs couldn't compete with IBM's home computers and their desktop applications. "It just didn't have the resources, the software, the capabilities and the price point that made it attractive."
Unfortunately, they were a commercial failure. There were so many things wrong with them — not just stuff that broke, but the baseline configuration was nigh on worthless. I recently was able to get another machine and got it up and running (with a few hiccups). I whipped up a video showing all the cool things it can do, but also running through what went wrong and why it ultimately failed.
The video shows the ancient green-on-black screen of 1984's AT&T Unix PC (with the OS running on a silicon drive emulation). The original machine had 512K of memory and a 10-megabyte hard drive described as slow, failure-prone, and noisy. There's also a drive for inserting floppy disks, and a separate MS-DOS board (with its own CPU) that could be plugged into the expansion slot — but the device was "remarkably heavy," weighing in aqt 40 pounds
See the strange 1984 mouse, and its keyboard with both a Return key and a separate Enter key. There's even plug-in ports for phone landlines. "It looked great," Shayde says in the video, showing off its Spirograph demo and '80s-era games like Pong, Conway's Game of Life, GNU Chess, "Trk", and NetHack. But besides slow startup times, it was expensive — in today's dollars, it would've cost roughly $15,000 — and suffered from Unix's lack of spreadsheets, word processing software and other office productivity tools at the time. At that price the Unix PCs couldn't compete with IBM's home computers and their desktop applications. "It just didn't have the resources, the software, the capabilities and the price point that made it attractive."
It was price (Score:5, Informative)
You could get an IBM PC clone and put Xenix on it for a fraction of the price. There were plenty of nerds who wanted Unix, it was always hyped up as the next big thing, but nobody wanted to spend that much.
And why did Xenix tank? Because Xenix was also expensive. Around a thousand dollars for the base system which didn't even include a C compiler.
It wasn't until MINIX came out, and Coherent dropped in price, that Unix-on-a-commodity-system became practical. And even then, it took Linux and the various GNU distributions to actually become a well supported, common enough, OS for it to gain traction outside of neckbeards.
Re:It was price (Score:5, Interesting)
No (Score:5, Interesting)
Xenix was $500, which was the standard price for major software back in the day. That's what WordPerfect cost, etc. Yes that was insanely expensive in inflation-adjusted dollars however the productivity gains over a typewriter made it absolutely worth it. And Xenix was super cheap (comparatively) because Microsoft did a bulk buy from AT&T which drove the cost of their licenses down. AT&T was not a monolith trying to maximize profit across the whole corporation - the licensing division at AT&T did not want the hassle of selling jelly-bean software so they set the price high. Xenix was pretty popular for what it was -- common office applications were ported over Xenix and you can still run those Xenix applications on Linux today (in theory -- I am sure the iBCS code in Linux does not get the love it did 20 years ago).
Minix was never a commercial operating system, though very late in its life it did get used for some specific use cases such as the Intel Management Engine. One reason Minix wasn't popular in the 1980s was because it wasn't licensed to be. Two, because it was designed to replicate Seventh Edition Unix for educational purposes. If you've ever tried to port over software to Minix you've found very quickly that calls which were quite standard and expected by the late 1980s (e.g. select, or named pipes, sockets, or the curses library) were not implemented at all, or were implemented in a perfunctory way which didn't match the contemporary Unix API.
I don't know that Coherent ever had much in the way of applications, you'd really be relying on it being compatible with other unices.
I think the bigger reason Unix never took off in the PC mass market is that there was no reason for it. In law firms, there is an obvious advantage in document sharing and centralizing backup and version management, and it was easy to quantify productivity gains that justified the cost. (I.e., lawyers billed by the hour, but secretary cost was covered by the lawyers' billable hours.) For most users, this was not the case -- if you needed a heavy duty Unix system for CAD or whatever that's what you'd buy. Otherwise MS-DOS did the job. There were always tons of "make the PC more industrial-grade" products like DesqView but the reality is that very few people needed that stuff until Microsoft brought it to the masses with Windows.
Re: (Score:2)
iBCS support in the Linux kernel was removed in January 2008, and it was already unusable at that point (so wasn't deprecated, just removed).
Re: (Score:3)
The idea was that IBM PC Compatibles won for a lot of reasons. But the clone market and piracy was the real reason. I remember the first time I saw the original manuals for an iBM PC 5150 and I was shocked that one manual was gobs of printouts of the source code to the PC BIOS. The Compaq made a clone very EARLY. CP/M was way too damn expensive... I can go on and make a history lesson, but the people old enough to remember have t
Re: (Score:3)
You missed the #1 gigantic reason that eventually got Microsoft in trouble with antitrust and then they invested in Apple to get out of it. Microsoft made exclusive deals with computer vendors where they could get the OS and later the OS bundled with Word and other products at a huge discount if you didn't sell any competitor's products. I worked for and we sucked at Microsoft's teat. MS is probably half the enemy to consumers Apple is today, but in the 90s, they were the devil. They still have the absolu
Re: (Score:2)
I don't hate Microsoft,
Weird after all their illegal skullduggery that set back computing a decade.
I think they should win competitively
They can't.
in gaming lately, they are losing, from what I've read and seen
Losing how? They have orders of magnitude more users in that department, even if Linux has increased in popularity by a large percentage since Windows 11 was released. And many games just won't run on Linux, especially a significant percentage of the most popular multiplayer games. I am gaming exclusively on Linux now, but the numbers just aren't that high. Yet? Hope so.
Re: (Score:2)
Those AT&T V20 clones essentially used a cloned/pirated 8086 processor produced by NEC that ran at higher clock speeds and had some fancy extensions that almost no software ever used.
I thought that was interesting. My friends asked if my computer was legitimate more than once. I was like "It's an AT&T computer, it's cool."
The NEC V20 series was WAY faster than any 8086.
They still produce them to this day.
Re: (Score:2)
Stuff like Netware was shit, but it was also cheap. From a business perspective, if it does what the business needs, even if it's a nightmare for admins and users, they aren't going to spend any more money on something that is technically better.
Re: (Score:2)
From a business perspective, if it does what the business needs, even if it's a nightmare for admins and users, they aren't going to spend any more money on something that is technically better.
You're right in that up-front costs often drove the decision-making, but what they didn't take into account is that software that is a nightmare to use means the admins get less done in a day, and the users keep requiring tech support, which means the company has to hire additional employees, so in exchange for a "cheap" one-time software purchase they are now saddled with expensive additional staffing costs, in perpetuity.
Re: (Score:2)
Sure, but they collected the bonus for reduced costs already, and if enough work isn't being done they just crack the whip harder or recommend replacing people with AI. They are not rational actors.
Re: (Score:2)
I still remember many many years ago, I had to use a hacking tool called burglar to get in to a netware server that had expired every password including the admin password. Novell support wasn't at all helpful.
Re: (Score:3)
Xenix was $500, which was the standard price for major software back in the day.
Windows 3.1 debuted at $149. OS/2 2.1 was $249, so for $400 you could get both OS/2 and Windows 3.1, and run software for both. There was approximately no desktop software for Xenix, and most people didn't have enough machine to run X11 gracefully anyway, which took a LOT more RAM than Windows 3.1 and a bit more than OS/2. So, no, and also no.
I think the bigger reason Unix never took off in the PC mass market is that there was no reason for it.
There were lots of good reasons for it. It was a superior OS on all levels. But the prices were absurdly higher than the competition. Consequently it only got picked u
I'll Tell You Why (Score:5, Interesting)
There were two primary reasons why it failed.
1. Price. It cost a frigging fortune!
2a. No off the shelf software. At its price point the only "people" buying these were businesses and granted researchers. But, there was no off the shelf or precompiled business software for it to be useful.
2b. You had to write your own software. Even if you bought software or had a developer write it for you, you then had to compile it to run on this box. There were only a select few with the knowledge, time, and patience to compile their own shit on that oversized calculator.
I enjoyed the video (Score:2)
I was around back in that era, but if I was aware of these the impression didn't stick.
Watching, I was struck by the (manufacturer's) decision to have the display physically affixed to (a raised post on?) the chassis of the computer. It seems like a potential point of failure, especially given the not-insignificant weight of even a smallish CRT - plus it would just make things more awkward when you needed to move the machine or get inside it. Anyone know if it was simple - or possible - to detach?
Re:I enjoyed the video (Score:4, Informative)
Up until the mid '90s, most computers were made to have a CRT monitor sit on top of it. I don't think that was a problem.
I used to lug CRT monitors to LAN parties back in the 90s... It was just the way we did it back then.
The mouse here was a common design from Mouse Systems, who also made the same model adapted to the IBM PC and for several other Unix vendors.
Re: (Score:2)
Right, I recall this as well - but in the video the monitor didn't appear to just be "sitting on top", it seemed to be rigidly attached to the case below it. Am I mistaken?
Re: (Score:3)
Re: (Score:2)
I think i mentioned in the vid that there are no external expansion things for this design. No external ports for SCSI or drives, and AFAIK, none of the boards presented that.
You did indeed! But I also have vague memories from that era of needing to open up electric / electronic devices occasionally, regardless of the manufacturers' intent (not that the general population would have done so!). I remember in the mid-late 1980s, at work I fairly regularly found myself replacing TTL chips that'd gone bad - back when it was realistic (and even straightforward!) to solder such things by hand...
Man now I've reminded myself how old I am, thanks a lot.
Re: (Score:2)
The short answer (Score:3)
Re: (Score:3)
No, not even close. 1984 was the height of the home computer boom, and businesses, by that point, had a personal computer on most white collar worker's desks. Games consoles had been pretty much killed at that point (https://en.wikipedia.org/wiki/Video_game_crash_of_1983), mostly by Jack Tramiel's war on Texas Instruments (!), and would take another half decade to start to become popular again. The Mac had just come out and was considered an impulse-buy personal computer.
Typewriters were already looking ant
Re: The short answer (Score:3)
Re: (Score:2)
Companies bought IBM PCs and XTs in jaw-dropping numbers while Home PC sold like "hotcakes" in the mid-80s.
I was watching a video last night, Radio Shack sold a quarter million TRS-80 Model Ones in the late 70s, by the time we got into the 80s home computers were all the rage from Apple, Atari, Commodore, and others (typically cheap clones of some of the more popular home computers), and thee were countless businesses/doctors offices, etc that were running their worlds on CP/M systems.
Related movie trivia (Score:5, Informative)
In the Sylvester Stallone movie "Cobra", there are two types of computers seen. In his apartment, there's a PC clone. In the police station, there are more PC clones but also a bunch of these AT&T Unix PC machines.
The only ones operating are the PC clones, whose vendor provided people. The provider of the AT&T machines provided nobody, and no information on how to run them. They looked super-cool for their day. During down-time, people switched them on, thought they were interesting, and then ignored them. Nobody has time on a movie set to figure stuff out just for some set dressing - every minute is money being burned. This was a very sad product placement marketing failure.
Re:Related movie trivia (Score:5, Funny)
Re: (Score:2)
Thank you! This is why I still come back to slashdot, those little tidbits are worth putting up with the slop.
Now can someone please start a Unix PC emulator and compile a media player with libcaca on it, so we can watch Cobra on one of those puppies?
Re: (Score:2)
This was a very sad product placement marketing failure.
I wonder if it was a "sad product placement...failure" or if it was simply a prop rented from a prop house to dress a set? Based on the number of times I've seen one of these in a movie or TV show I'm convinced they were owned by a studio or private prop house - when a company does a product placement they typically make sure the company logo is plainly visable. (Ever notice the strange places (wired) AT&T phones had logos on them when the camera pulls in on someone talking on a phone? Typically there's
Nope. (Score:2)
they were NOT rented props nor were they props from the studio warehouse. They were a product placement. Oh, and they were NEW and fully functional.
I was on set when they arrived, and played with them myself. I thought they were really cool (remember, this was the mid-80s and they were advanced machines for the time) and would have loved to get one. I've long wondered what happened to them after production wrapped, given that such items placed in such deals often do not go back to the suppliers. I do know t
For Comparison (Score:2)
Type IBM P/N Date ann Date w/dn Bus Slots Bays Processor MHz BaseRAM MaxRAM FDD HDD
XT 5160-087 March 1983 June 1984 ISA, 8-bit 8 2 Intel 8088 4.77 128 KB
Easy - Cost (Score:1)
Re: (Score:2)
Similarly expensive ... (Score:4, Interesting)
I never worked with one of those AT&T Unix PCs, but had a Sun386i on my desk for a while, on loan from a customer -- we offered a large discount on the software support contract if the customer loaned us one of their systems for support and development. We added an expansion storage case (that attached on top) with a (I think) 500MB SCSI disk, that was physically huge by today's standards - it had 8 platters and was almost the size of a shoe box. I later used that disk on my BSD/386 system, then disassembled it when it died and made the platter assembly into a paperweight that I still have.
My opinion (Score:3)
Found in oddest places (Score:2)
Around 1988 I found one of these in the back of the hall closet at a friends house. I asked him what it was and he said it was his dads work computer, but he hated it so he threw it in the closet.
However, his dad was a salesperson for a bearing company. Didn't seem odd to me as a kid, but I also didn't know those cost as much as mid-range sedan. I wonder what they expected him to do with it?
I briefly worked with one of these (Score:5, Informative)
I worked for a while in a shop that did server side Unix systems, and they had a ton of desktop computers for compatibility testing. In addition to the usual suspects, I remember testing on a Dec Rainbow, a Sun 386i, a Sirius/Victor 9000 (nice for an 8088 system), an HP/Apollo, a Lisa (slooooow), a Texas Instruments PC, and this one.
While there wasn't anything actually wrong with it, there wasn't anything particularly compelling about it, either. While it was nice to have a "pure" Unix (as opposed to Xenix, or rebranded versions like Esix), it didn't warrant the jaw-dropping cost. Sure, it was great hardware, and AT&T had the chops to fight IBM on the corporate desktop, but outside of the Fortune 500, most companies were buying clones rather than true IBM for a fraction of the price. The price of this was eye watering to begin with, and while it had great development tools, it had very little commercial software, and what it did have was also overpriced.
You could buy something like three actual IBM PCs with Lotus 1-2-3 for the cost of a single one of these. And when you factored in clones, it was closer to eight clones for one AT&T box.
The business case simply didn't make sense for the majority of the market.
Re: (Score:2)
It feels like that AT&T built it more for themselves than the market. Looking at the maintenance manual, except for memory, it had no way to upgrade the. The screen was straight bitmapped with no vertical scrolling, the hard drive is a custom controller, and the the modem was built right into the board. Its like its custom designed to just run X . With all that said, it was using all off the shelf parts it should of been much cheaper if you remove the UNIX licence.
It feels like they wanted a work
3B1... (Score:2)
I got my hands on a 3B1 around 1990. Linux wasn't available of course, and a used Sun was just out of my reach. It was a giant box of suck. The most corporate version of Unix I've ever used. It made me pine away for the dialup guest account on Compupro's Unix Version 7 system in the east SF Bay back in the 80's...
Eventually I picked up a Sun 3/50, and later a 3/60. Then some kid in Finland posted on Usenet and well...
Re: (Score:2)
Back before Solaris, SunOS was a truly great BSD. It really didn't cause you any problems. Sure, for early ones you needed to relink libc to get DNS resolution, but at least they gave you enough tools to do that with!
Even Solaris was pretty non-irritating up through about 7...
Robert Cringely of "Accidental Empires"... (Score:5, Interesting)
Robert Cringely of "Accidental Empires" wrote about how AT&T was run by a "3rd wave guy" and was shocked when Info World did actual scientific tests. He wrote in his book:
"Then Bob Kavner came to town, head of AT&T’s computer operation and the guy who invested $300 million of Ma Bell’s money in Sun Microsystems and then led AT&T’s hostile acquisition of NCR—yet another company that didn’t know its PC from a hole in the ground. Eating a cup of yogurt, Kavner asked why we gave his machines such bad scores in our product reviews. We’d tested the machines alongside competitors’ models and found that the Ma Bell units were poorly designed and badly built. They compared poorly, and we told him so. Kavner was amazed, both by the fact that his products were so bad and to learn that we ran scientific tests; he thought it was just an InfoWorld grudge against AT&T. Here’s a third-wave guy who was concentrating so hard on what was happening inside his own organization that he wasn’t even aware of how that organization fit into the real world or, for that matter, how the real world even worked. No wonder AT&T has done poorly as a personal computer company."
https://www.cringely.com/2013/... [cringely.com]
--JoshK.
Robert Cringely of "Accidental Empires"... (Score:2)
Robert Cringely of "Accidental Empires" wrote about how InfoWorld did scientific tests, and then the head of the AT&T PC division was shocked that they had done so. Cringely writes:
"Then Bob Kavner came to town, head of AT&T’s computer operation and the guy who invested $300 million of Ma Bell’s money in Sun Microsystems and then led AT&T’s hostile acquisition of NCR—yet another company that didn’t know its PC from a hole in the ground. Eating a cup of yogurt, Kavne
Hot as a griddle, loud as a vacuum cleaner. (Score:3)
August 1994, Madison WI, near West side close to the UW campus around University Ave, I found one of these sitting on the curb. Beige, a little stained, but completely intact.
I was staying nearby in a sublet room, basically homeless, but i'd gotten a temp job cleaning out nasty dormatories which gave me a tiny cash flow.
I already had experience with various 286/386 PCs, PS/2s and consumer all-in-ones like the TRS-80s, COCO2 and Commodore 64 so I lugged this lead anchor back to my room and plugged it in. It booted up and presented me with a Unix log-in prompt. I tried every "default" password I could think of, but none worked. I didn't know this Unix so I had no tricks to try that I didn't faintly remember from reading Cult of the Dead Cow newsletters.
I put it aside. When I tried turning it on later, it never booted up again.
I probably should have tried selling it on Usenet forsale groups but I may have been without any computer and living offline at that time.
Sublet came to an end, and I had to move on. It was too much to take with me on foot.
I did salvage the hard drive and I'm pretty sure I used it for something eventually. And the two cooling fans which were loud and powerful and still insufficient to keep that monster cool. They were proportioned just like ordinary fans except they were extra deep. Not server fans, though. I used these in projects for years afterward. The early Motorola 68010 CPU is still in my parts collection. The husk of the computer went into a dumpster so noone would see my desecration.
Re: (Score:2)
> When I tried turning it on later, it never booted up again.
I had one of these in the mid-to-late 90s - I forgot how much I paid for it - and as I recall you needed to run a command to park the HD heads before shutting down, especially if you were going to move it... and the 10meg drive was a Seagate, notorious for "stiction" in those days - sometimes you'd have to twist the whole unit abruptly to un-stick the platters (risky at best!) to get it going.
Cool machines for their time, certainly, but hard as
We used them at USPS (Score:1)
Constrained size. (Score:2)
A lot of the compromises that put microcomputer OSes on the 'how about you get a real computer?' list just don't hit as hard on very small systems. Oh boy; the multitasking is nonexistent or one of the nasty kludges like TSR or 'coo
QNX (2.x) in mid-'80s was great! (Score:1)
My dad was running a 16-modem BBS system (running an interactive multi-player text game, the Scepter of Goth) in the mid-80s (around '85) running QNX. It was running an 8086 with a 10MB full-height hard drive and 640KB RAM and it was rock solid and ran 24/7 for years. Maybe some of these *nix systems had messed up default configs, but QNX was pretty slick.
DEC (Score:2)
I played around with DEC Unix on a MicroVAX around the same time. It worked pretty well, but you were still constained to the DEC environment.
Yeah, Amiga (Score:1)
Return vs Enter (Score:2)
Back in the "old days" computing power was mainly available on mainframes and miniframes. Applications were built where displays were just forms rendered off terminals. The terminals had some smarts and after each form was filled in the Enter (Submit, etc.) key was pressed so the values were sent to the "real" computer so either the next forms could be rendered or the output could be calculated/printed. This form was rendered in what was typically called block mode. There is no communication with the comput
We had some alternatives atm (Score:1)
2. SCO
3. DELL Unix
# Not unix per se
4. 386BSD
5. Freebsd 1993+
6. Slackware Linux ~1993 too
used ISC, SCO & BSD a lot
Re: (Score:2)
Unix on 68000-based "servers" (not Sun, et al, but no-name 68000 boxes, more like an S100 bus system than, say, a Sun server) was a thing back in the late 1980s, I remember my small college in central PA had one that ran a dozen or so serial terminals for CS students to develop code on at the time.
I think it was marketing power and marketing cash! (Score:2)
But the money and marketing of the I
Re: (Score:2)
How reliable were hard drives back then? (Score:1)
2027 will finally be the year of the UNIX desktop! (Score:2)
PC/IX (Score:2)
Ran in 256K, but only on a real IBM PC with the original 10MB disk and controller. No GUI, just text. My brother-in-law bought it for his brand new XT, but basically scrapped it when he realized he could only run that and not switch to DOS when he needed to.
Unreliable, or just expensive (Score:2)
Funny, I had a 10M h/d in my first PC machine at home (a 286). The drives were solid. Then Coherent dropped, and I played with it when I wasn't busy at hom.
AT&T 3B1 Was My First PC (Score:2)
I remember unboxing this AT&T 3B1 with utter excitement. It ran UNIX, real AT&T UNIX (which we later learned was riddled with bugs lol). I could compile GCC overnight LOL. There were times I had no f'ing idea what I was doing, but I kept learning. I had the modem and the voice card, which gave me tons of hours of exploring/learning, dial-up BBS via ProComm. I connected a DEC VT330 to this and logged in -- total excitment (LOL!). Hey, it was my geeky time.
I agree with other sentiment he
I bought one, then sold it a few months laterâ (Score:2)
PC-IX (Score:2)
Xenix required me to waste memory for a card (Score:2)
Xenix required me to waste memory for a serial concentrator card, but it wasn't the operating system's fault, it was the computer architecture's fault.
Back then, I worked on Xenix machines with serial concentrator cards, which were ISA bus cards that provided a dozen or more serial terminal interfaces. That way each workstation could just be cheap $300 terminals (compared to $3,000+ computers). It worked well at many doctors' and dentists' offices, and even at least one TV station.
The way memory worked on
AT&T's CPU and Unix (Score:2)
While we're on this topic of AT&T's Unix PC, I do wonder what happened to AT&T's own CPU - the WE 32000 microprocessor? There were 2 computers that used that CPU - the 3B5 and the 3B15. AT&T could/should have made that the main platform for Unix, then they could have bundled it for free w/ those computers and sold it that way