The Worst-Selling Microsoft Software Product of All Time: OS/2 for the Mach 20 (microsoft.com) 127
Raymond Chen, writing for Microsoft DevBlogs: In the mid-1980's, Microsoft produced an expansion card for the IBM PC and PC XT, known as the Mach 10. In addition to occupying an expansion slot, it also replaced your CPU: You unplugged your old and busted 4.77 MHz 8088 CPU and plugged into the now-empty socket a special adapter that led via a ribbon cable back to the Mach 10 card. On the Mach 10 card was the new hotness: A 9.54 MHz 8086 CPU. This gave you a 2x performance upgrade for a lot less money than an IBM PC AT. The Mach 10 also came with a mouse port, so you could add a mouse without having to burn an additional expansion slot. Sidebar: The product name was stylized as MACH [PDF] in some product literature. The Mach 10 was a flop.
Undaunted, Microsoft partnered with a company called Portable Computer Support Group to produce the Mach 20, released in 1987. You probably remember the Portable Computer Support Group for their disk cache software called Lightning. The Mach 20 took the same basic idea as the Mach 10, but to the next level: As before, you unplugged your old 4.77 MHz 8088 CPU and replaced it with an adapter that led via ribbon cable to the Mach 20 card, which you plugged into an expansion slot. This time, the Mach 20 had an 8 MHz 80286 CPU, so you were really cooking with gas now. And, like the Mach 10, it had a mouse port built in. According to a review in Info World, it retailed for $495. The Mach 20 itself had room for expansion: it had an empty socket for an 80287 floating point coprocessor. One daughterboard was the Mach 20 Memory Plus Expanded Memory Option, which gave you an astonishing 3.5 megabytes of RAM, and it was high-speed RAM since it wasn't bottlenecked by the ISA bus on the main motherboard. The other daughterboard was the Mach 20 Disk Plus, which lets you connect 5 1/4 or 3 1/2 floppy drives.
A key detail is that all these expansions connected directly to the main Mach 20 board, so that they didn't consume a precious expansion slot. The IBM PC came with five expansion slots, and they were in high demand. You needed one for the hard drive controller, one for the floppy drive controller, one for the video card, one for the printer parallel port, one for the mouse. Oh no, you ran out of slots, and you haven't even gotten to installing a network card or expansion RAM yet! You could try to do some consolidation by buying so-called multifunction cards, but still, the expansion card crunch was real. But why go to all this trouble to upgrade your IBM PC to something roughly equivalent to an IBM PC AT? Why not just buy an IBM PC AT in the first place? Who would be interested in this niche upgrade product?
Undaunted, Microsoft partnered with a company called Portable Computer Support Group to produce the Mach 20, released in 1987. You probably remember the Portable Computer Support Group for their disk cache software called Lightning. The Mach 20 took the same basic idea as the Mach 10, but to the next level: As before, you unplugged your old 4.77 MHz 8088 CPU and replaced it with an adapter that led via ribbon cable to the Mach 20 card, which you plugged into an expansion slot. This time, the Mach 20 had an 8 MHz 80286 CPU, so you were really cooking with gas now. And, like the Mach 10, it had a mouse port built in. According to a review in Info World, it retailed for $495. The Mach 20 itself had room for expansion: it had an empty socket for an 80287 floating point coprocessor. One daughterboard was the Mach 20 Memory Plus Expanded Memory Option, which gave you an astonishing 3.5 megabytes of RAM, and it was high-speed RAM since it wasn't bottlenecked by the ISA bus on the main motherboard. The other daughterboard was the Mach 20 Disk Plus, which lets you connect 5 1/4 or 3 1/2 floppy drives.
A key detail is that all these expansions connected directly to the main Mach 20 board, so that they didn't consume a precious expansion slot. The IBM PC came with five expansion slots, and they were in high demand. You needed one for the hard drive controller, one for the floppy drive controller, one for the video card, one for the printer parallel port, one for the mouse. Oh no, you ran out of slots, and you haven't even gotten to installing a network card or expansion RAM yet! You could try to do some consolidation by buying so-called multifunction cards, but still, the expansion card crunch was real. But why go to all this trouble to upgrade your IBM PC to something roughly equivalent to an IBM PC AT? Why not just buy an IBM PC AT in the first place? Who would be interested in this niche upgrade product?
Easy Answer (Score:5, Insightful)
But why go to all this trouble to upgrade your IBM PC to something roughly equivalent to an IBM PC AT? Why not just buy an IBM PC AT in the first place? Who would be interested in this niche upgrade product?
If the expansion card cost less than a new PC AT and you already had an XT hanging around, that's who.
Re: (Score:3)
It probably didn't though, because by that time there were already clones. They weren't as ludicrously cheap as PCs became later, but they offered substantial savings. I had a 286 clone (at all of 6 MHz) with 1MB RAM and a 40MB RLL disk, on which I ran SCO Xenix...
Re: (Score:2)
Clones were never under $500 during that time frame.
And what was up with the title of the article?
The Worst-Selling Microsoft Software Product of All Time: OS/2 for the Mach 20
Nobody was thinking of buying OS/2 back in 1987. OS/2 was only released in December, and it didn't bring anything new to the table - it was basically a DOS text-mode-only replacement. And if you waited to get the GUI version, you also had to get a new video card and monitor. For what? Editors/Word processors, spreadsheets, databases - they were all text-mode programs.
Attack of the clones (Score:2)
Around the same time I built a 80286 based system with a baby AT case ( remember those) 1 gig of ram and a 20 MB drive for around the mid &900's if memory serves .. plus it had an EGA card. The Mach Card was 495 just with nothing .. the 80286 cost like $400++
Re: (Score:3)
Around the same time I built a 80286 based system with a baby AT case ( remember those) 1 gig of ram and a 20 MB drive for around the mid &900's if memory serves .. plus it had an EGA card. The Mach Card was 495 just with nothing .. the 80286 cost like $400++
Sorry, but I'm calling BS. NOBODY could afford 1 gig of ram back then, and even if they could, no motherboard would take it, and a 286 couldn't address it. The hard limit was 16 mb. And a 256kb chip was $600 in 1985.
Of course nowadays you have to give away your old 4gig chips when you want to upgrade. Nobody wants them. It's a good time to be alive.
Re:Attack of the clones (Score:4, Informative)
I think the poster simply made an understandable slip up. Since everything memory nowadays is in gigabytes, it is easy to understand why "gig" slipped in.
He meant to say "meg". On the 286-AT c1987, with DOS 2.x, standard memory was 640K. Upgrading to 1 MB required a card, and then all of the arcane configuration that DOS demanded for expanded versus extended memory. Getting 2 MB to work, as I recall, was a pain.
But, had he really meant "gig" (which he didn't, he meant "meg"), then your comments were ready on the money.
Re: (Score:2)
Yeah meant to say 1 meg. it was a long time ago. After all who would ever need more than 640k according to Bill Gates.
Re: (Score:1)
Must be extremely difficult for you to grasp that he made typo and meant 1MB of RAM ...
Re: (Score:2)
Obviously he meant MEG not GIG. Also having one GB of ram with a 20MB hard drive would not make much sense
Re: (Score:2)
Thin/thick client. Kernel and basic init on teh 20mb drive, load into a ram disk and query a terminal server/start an X session and query,etc or start mounting remote file systems and kick off the rest of a "normal"-ish init
But yeah... he/she/it probably meant meg vs. gig. Happens to me all the time in reverse (constantly refer to megs instead of gigs), guess on us folk getting older it is showing when we got into computing. I imagine in 20 years there will be a /. posting about the "good old days" when
Re: (Score:2)
Probably. I want a box with 2tb, dual cpus, and the ability to switch to the alternate cpu and not execute the instructions that caused a crash. The first two are currently purchasable, but that last one is going to take some time to get down to the consumer level.
But when it does - just boot up once a year, updates get applied to one copy of the os, then the other after a suitable delay, and when not in use, just go sleepy-bye. The OS and programs are permanently resident in RAM between boots, because t
Re:Easy Answer (Score:4, Interesting)
Sure but you could run them all at the same time and switch between them (with some difficulty on the 286). DOS users used to dream of that kind of multitasking. Later on in 1994 I had a friend who ran OS/2 2.1 with a text-mode-only shell that he exclusively used to multitasks DOS programs. It was super slick and at the time that was all we needed and wanted since we'd not yet warmed to MS Windows. Even then most of the productivity software was in MS-DOS. Wordperfect 6 gave me everything Windows could have but in MS-DOS. Those were the days. I still fire these apps up in DosBox once in a while just for nostalgia reasons.
Re: (Score:2)
Re: (Score:2)
Various versions of DR-DOS had various task switcher tools as well. Most people didn't have enough RAM to make much use of them, though, and even if they did a lot of programs you'd really like to run in such a context (like games) didn't cooperate.
Re: (Score:3)
I feel like it was one of those things that when they started working on it, it made some economic sense. But with the clones and falling chip prices it stopped making sense. The computer market was really wild back in 80's and 90's. Even with hindsight it is difficult to explain what happened. I don't think there was really any genius back then that could predict what would happen. I'm convinced that everyone that succeeded was lucky.
Microsoft is especially lucky, they managed to produce several massive fl
Re:Easy Answer (Score:4, Interesting)
Microsoft had little competition. There were other DOS operating systems, there was the GEM graphical system, there was OS/2, but basically they were all worse than what Microsoft made.
There were superior computers - Amiga, Archimedes, PC98, X68000, FM Towns and many more, but they were all from one manufacturer which prevented them from getting really cheap and dominating the market.
Microsoft succeeded because everyone else was worse in some important way.
Re: (Score:2)
In the mid-80's the clones were still slowly arriving. And MS was trying to figure out how to sell DOS outside of bundles with IBM PC. Back in the early 80's they were still making custom OEM editions of DOS for the many not fully IBM PC compatible systems. I'd say the transition from IBM dominance to a vibrant clone market was over the course of a few years from 1983 to 1986. I doubt Microsoft caught on early but rather stumbled into the market.
Now assuming that it wasn't obvious in the early to mid 80's t
Re: (Score:2)
It's interesting that in the US you didn't have so much competition on the business side. Japan had a lot, and Europe had a few more. Well, there was the Spectrum that Clive really wanted to be taken seriously.
Re: (Score:2)
For most of the 20th century it felt like IBM had their finger in every government contract and a service agreement with every major corporation.
For small business Commodore, Apple, and Wang Laboratories were pretty well represented throughout most of the US. Late 70's to early 80's had a lot dedicated word processors and CP/M systems in small business roles. A word processor might not sound that interesting, but you could hook them together and share a printer and sometimes even share files between them. S
Re: (Score:2)
Competition was bad for the 'personal computer' industry back then. Incompatible operating systems forced software developers to, at best, compile for each platform. I/O presented problems. Not a good place. IBM, Microsoft, and PC/MS-DOS brought the interoperability and stability needed, and the architecture was IBM's fault for letting out of the copyright box, leading to a huge explosion of the market and ultimately more profit for everyone.
I bought a 'turbo XT' in the 80s, third* of my personal computers,
Re: (Score:2)
Interesting that it had to be taken from IBM. There was MSX which was a similar concept, but didn't last. Maybe because it never went beyond 8 bits.
Re: (Score:2)
MSX was just another platform. Interoperability would have doomed all these to focus on a manageable number of platforms, hardware and software, that could be supported.
Just software would have driven this. Networking was the logical and necessary next big thing in personal computing, and we saw IPX/SPX and NetBIOS being replaced by TCP/IP, a necessary change. But in hardware, even Apple did not want to support two software platforms for long, so AppleDOS gave way to the Macintosh system, which had to move
Re: (Score:2)
Re: (Score:2)
Microsoft had a strong arm in making sure that other competing operating systems didn't gain a foothold in the market. Not by making a better product, but by doing things like requiring retailers to drop competitors' OSes from their shelves or not allowing them to carry MS Office. I remember those days clearly and it colors my view of MS to this day. It's a driving reason I don't want them to ever gain monopoly status in any other market (gaming, etc...).
Re: (Score:2)
It sold poorly because there were better expansion boards available.
In 1987, I had an expansion board with a 386. I don't remember the price, but I do recall thinking it was a really good deal.
So why would anyone settle for an 8086 or 286?
Re: (Score:2)
Plus the upgrade would doubtless be worse than a real 8086 machine. The 8088 had an 8 bit bus, so all the RAM and ISA cards on the motherboard are going to be limited to 8 bit mode.
Re: (Score:2)
It would be worse but it's not a big problem because it's basically just using the PC as a backplane, and the only things on that bus will be HDD and graphics — if you install the RAM expansion, it gets attached to the CPU. Graphics rarely consumed a significant percentage of bus bandwidth back then (which BTW was actually better on the PC than most other computers in the same class, both for 8 bit and 16 bit ISA — in the really real world, 16-bit ISA with DMA is faster than Zorro II or NuBus) a
Re: (Score:2)
Interesting. I watch PC Retro Tech who does a lot of testing with video cards from that era, so I'll have to look to see if he has done any videos about the bandwidth. I got the impression that the main limitation was not the bus but the slow RAM on the card itself, being shared with the video generator.
Re: (Score:2)
Yes, the slow memory was a big part of it. I had one of the last ISA VGA cards in my last fully ISA PC (386DX25 running Slackware 2.0 with kernel 1.1.47 at the end, IIRC) and I had a Trident 8900D which was one of the fastest cards in part because it had interleaved memory.
Re: (Score:2)
Hot damn! DX25? Trident? But monochrome, right? Slackware 2.0? You were on easy street...
Re: (Score:2)
Hot damn! DX25? Trident? But monochrome, right? Slackware 2.0? You were on easy street...
No, I had 1MB video so I could do XGA@8bpp, or (narrowly) VGA@24bpp (which I never used, not even once, but the card would do it.) If it makes you feel any better, my prior machine was a 286-6 with 1MB RAM running SCO Xenix. Didn't even have room for the compiler, sadly. I got it free as I knew the lead dev.
Re: Easy Answer (Score:2)
Yeah, well I was trying to fix SCO Xenix apps back then, no fun. Ended up with Slackware 0.something in an sx25 myself. It was a distraction from NetWare, my day job, and getting enough RAM to keep it working as an internet router at home. Xenix was a challenge.
Re: (Score:2)
Such a rationale works now, but not in *that* era. That was the age of "Computerization", in which companies were putting computers on and network connections at the desks of people who'd never used a computer before.
This was an era in which, when a company got a shipment of computers, it was usually a whole truckload. It'd make more sense in that context to roll a slightly older computer down to someone who wasn't in line to get a computer yet than to upgrade it for someone who needed the latest and great
Re:Easy Answer (Score:4, Informative)
People seem to forget that the original xt with a couple of floppies, 256 k of ram, and NO video card or screen was $2,495. Add a 10 mb hard drive and you're north of $4k.
Mouse? You didn't need no stinking mouse. Or a light pen (anyone remember those)?
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
The standard actually allowed for four drives on a controller, but IBM screwed it up. Most or at least many drives don't have the jumpers for it anyway as a result. On some platforms (e.g. Amiga) you could have four floppy drives on one bus, because they didn't botch the same interface when they implemented it. A simple ID lead shift is all it takes for each external drive to automatically be assigned the next ID.
You could however easily have a second floppy controller, because some IBM machines had a secon
Re: (Score:2)
It was my experience that XT type 8088 computers were more often tossed than upgraded. VLB cards were not useful, needing the bus to work, and nothing overcame the graphics limitations.
Things moved quickly, back then, and machines got demoted. Some of my clients donated 'old' XT machines to schools etc.
Re: (Score:2)
Comment removed (Score:4, Informative)
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
The address space for the 8086 was 1 megabyte, not 640k. (actually you could address an additional 64k above the initial 1 - remember himem.sys? - meg by toggling the a21 line).
The problem is that space between 640k and 1 mb was reserved for other uses, so user space was 640k in DOS.
Re: (Score:2)
Re: (Score:2)
I'm not sure where you get it from that I'm claiming 640K was the address space of the 8088. I quite specifically mention applications, and in the 8088-based IBM PC and PC/XT architecture, 640K was the full memory space for applications.
Except it wasn't. You could always find spare space to stuff TSRs and stuff into. I wrote code to stuff commands into the monochrome video buffer (I had both colour and hercules video cards, and noticed that with a dual monitor setup, only the color video buffer was re-initialized), and on soft reboot execute the code in the mono video address space.
Re: (Score:2)
Re: (Score:2)
That was extremely rare for commercial software though, and for very good, very obvious,m reasons! In practice almost all applications, even SideKick, took up some of that 640k. It was the only memory they could guarantee was always there. Video RAM was video RAM, you didn't store anything there you wanted to keep.
The issue was so serious that memory expansion systems for 8088 machines became common, using the Expanded Memory APIs to make it available.
The IBM PC was one of the most horribly designed machines I've ever seen. It says something about those three letters that it succeeded anyway.
Here [wikipedia.org]
One technique used on early IBM XT computers was to install additional RAM into the video memory address range and push the limit up to the start of the Monochrome Display Adapter (MDA). Sometimes software or a custom address decoder was required for this to work. This moved the barrier to 704 KB (with MDA/HGC) or 736 KB (with CGA).
And a headless XT could use up to 952kb of contiguous ram for user programs.
I wrote TSRs that would only need a stub in conventional memory and stored the rest in "unconventional" areas. I should imagine others did to. It was simple enough in practice. TSR loads the stub, finds an address that's free, loads the rest into that address space, then patches that stub to jump to that address. It's not like memory had any sort of write protection, so why not?
Re: (Score:2)
I upgraded some systems at work back in the day similarly with NEC V20 chips. The users complained about speed and this was as much as the company would spend.
Re: (Score:2)
Re: (Score:2)
And of course MacCharlie
Re: (Score:2)
Or they could have just parked the older machines in a closet ... "just in case." They would still get their 7-year depreciation, while enjoying the benefits of faster, cheaper machines. Or they could have sold them at a discount (their real value, not the on-the-books value), and taken a terminal loss. Or "dropped them off the desk." Or donated them to get them off the books.
If an advancement will more than pay for itself, it makes sense to stop using the old stuff, even from an accounting point of vie
Wow that PDF link was so useful (Score:2)
Re:Wow that PDF link was so useful (Score:5, Informative)
I think the Microsoft article is no longer pointing to the right link. Doing a search on google leads to this:
https://thecomputerarchive.com... [thecomputerarchive.com]
Re: (Score:2)
you might try https://www.thecomputerarchive... [thecomputerarchive.com]
Challenge Accepted (Score:4, Informative)
I have a sealed copy of DOS 6.21 UPGRADE.
DOS 6.21 was only on the market for a month due to a lawsuit over the compression technology used, which prompted a very quick turn around to DOS 6.22 (quite possibly the most popular version of DOS outside of the "bundled" versions with Win9x)
I'm really curious to know the sales figures of DOS 6.21 INSTALL, DOS 6.21 UPGRADE, and this CPU upgrade mentioned in the article.
Re:Challenge Accepted (Score:4, Informative)
Wow, those were the days.
Re: (Score:2)
Microsoft also produced a customized version of OS/2 for the Mach 20. Despite being tailor-made for the Mach 20, it still had terrible performance problems.
One of my former colleagues spoke with the person who took over from him as the support specialist for OS/2 for Mach 20. According to that person’s memory (which given the amount time that has elapsed, means that we should basically be saying “according to legend” at this point), a total of eleven copies of “OS/2 for Mach 20” were ever sold, and eight of them were returned.
Re: (Score:2)
OS/2 (Score:2)
The good part about OS/2, is you could own an IBM machine and be as smug as an Amiga or Macintosh person.
Re: (Score:2)
My web guru still raves over Presentation Manager. I used to send him screen shots of local ATM machines hung on the OS/2 boot screen for fun. He was not amused, but then he wasn't standing in the snow wanting beer money.
And yes, I had a Sony/Ericsson T632 that was da bomb.
Re: (Score:2)
Re: (Score:2)
closest i came to os/2 was a coffee cup when os/2 java was released.
Wost selling? (Score:1)
Re: (Score:2)
How about Microsoft Bob!
Re: (Score:3)
Re: (Score:2)
Dear summary writter (Score:5, Insightful)
If you're going to run a headline, don't forget to make the summary relevant. The headline: OS/2 for the Mach 20. Copy the part of the article about OS/2 rather than a history of the hardware.
Re:Dear summary writter (Score:4, Informative)
One of my former colleagues spoke with the person who took over from him as the support specialist for OS/2 for Mach 20. According to that person's memory (which given the amount time that has elapsed, means that we should basically be saying "according to legend" at this point), a total of eleven copies of "OS/2 for Mach 20" were ever sold, and eight of them were returned.
That leaves three customers who purchased a copy and didn't return it. And the support specialist had personally spoken with two of them.
Slots (Score:2)
You would usually run out of IRQs before you ran out of slots. An AST Six-Pack gave you more memory, a clock, two serial ports, a parallel port and a game port. That would eat up four IRQs if you used all the ports. The original PC/XT only had one IRQ controller, so that one card alone could eat up half your available IRQs, not including a floppy controller and two reserved IRQs for system devices.
Re: (Score:3)
Re: (Score:3, Interesting)
The serial po
Re: (Score:2)
Network cards broke IRQ sharing as I recall, too demanding.
Try running an IBM 4MB Token-Ring card alongside a Thomas-Conrad ARCNet card on the ISA bus. Never happening. Sharing is caring, and neither company cared.
With Microsoft, worst selling isn't worst product (Score:1)
WIth most companies, the worst-selling product is often the worst, the msot detested, the one that makes the least sense and misses its intended market.
WIth Microsoft, a sizeable portion of their products across all divisions and across version numbers is at best tolerated, and at worst properly hated by Microsoft customers. But the customers are buying them anyway because they have no choice: new PCs with the newest version of Windows pre-installed, Windows update forcibly shoved down the users' throats, o
Didn't even know it existed (Score:2)
Re: (Score:2)
I had never heard of it until today myself. Really peeves me to find about this now, since I went to the trouble of buying a NEC V20 chip just to tweak my old 8088 PC clone at the time! And at the time, I was happy with the ~40% maximal, paltry increase. I would compile useful programs (for me) with the V20 CPU instructions. The biggest performance "boost" came from uuencode/uudecode. Of course, soon after that time, I eventually upgraded to a 386 PC clone (and that would have made the MACH 10/20 moot
I had one (Score:2)
Re: (Score:2)
Wish OS2 Source was Open (Score:2)
Oh no, no room for a network card? (Score:2)
How many people were even using network cards circa 1987? IIRC, my dial-up modem connected to a serial port.
1987 Re:Oh no, no room for a network card? (Score:1)
How many people were even using network cards circa 1987?"
In large companies and universities, plenty. IBM Token-Ring, ARCNet, and Apple's Macintalk were just a few of the hardware networks besides the various flavors of Ethernet that were around at the time. Not to mention the many other physical-layer network interfaces that fell by the wayside.
As far as using networks in the 1980s, there were PCs that used NICs to connect to minicomputers and mainframes, PCs that used NICs to connect to Novel servers a
Re: (Score:2)
I guess I was under the impression these "Mach" products were being marketed to home users, not to businesses.
Re: (Score:2)
Cars existed by the Great Depression, so personal computers may have been the first time under standardized accounting practices that a new product (with the same name as an existing product!) appeared at lower c
Re: (Score:2)
Yes, Token-Ring was IBM's solution to some interesting use cases with mid-range systems and common in banks, ARCNet was a working industrial control system method, and Ethernet was getting a lot of use from DEC. All this on coax PHY. Appletalk of course on twisted pair was bastardized regularly, leading to heated discussions on the actual designed topology, and why it often didn't work for people who never had the documentation.
Novell and Banyan were two of the early PC networking successes, then the compet
Re: (Score:2)
Re: (Score:2)
'later'.
For instance, around 1985. SDLC was an IBM mainframe protocol in the mid-late 70s, of course needed with the S/360 etc. 5250 terminals were common in their midrange systems, and twinax the normal PHY. Early Days with emulation cards is more latter days, and had to move fast to beat TCP/IP and telnet/5250 to the punch. I recall helping out with AS/400 and RS/6000 systems getting them running TCP/IP properly and using the gateways available to them. Or not, as some systems had to be isolated, so teach
Re: (Score:2)
Re: (Score:2)
Astoundingly surprising. (Score:2)
Re: (Score:2)
Since OS/2 was largely an IBM product, Microsoft had few reasons to love it. And IBM was mostly left holding the bag with OS/2 when Microsoft abandoned it.
Heck, Apple should have adopted NeXT as its future, but it all worked out for the good of the corporations, and we do things unimaginable in 1986.
Re: (Score:2)
Since Microsoft wanted OS/2 to fail, not selling many copies seems like success.
I think Microsoft wanted OS/2 to fail because IBM insisted that OS/2 1.x would run on a 16-bit 286. Microsoft OTOH saw the 286 as a dead end, and wanted to target the 386 instead.
3 copies (Score:4, Informative)
Card required was not a great seller, and the Microsoft OS/2 operating system version that required this rare combo of hardware did not sell well.
They sold 11, 8 were returned for refund meaning only 3 copies were kept.
I spent a year in the Microsoft Hardware Support g (Score:1)
That was one of the products I supported.
The primary thing we supported was the Microsoft Mouse. You have to remember that a mouse was a strange accessory at the time. Lots of people bought the mouse as part of a bundle with PC Paint, and then wanted support for the paint program.
Re: (Score:2)
My first personal on-my-own-desk PC was an 8 MHz 286-AT in the late 1980's.
It had 1 MB memory and a mouse. It came with PC Paint, and not long after I upgraded to Z-Soft's PC Paintbrush.
And yes, at the time, that was about the only thing the mouse was much meant for.
Before that, my graphics and art were with easel, pens, paints, and drafting tools. Now, my diverse graphics activities use many modern digital tools. But, for me, and I would guess quite a few others, those early PC art programs were a cruc
Amiga Bridgeboard (Score:2)
I had the Commodore version - the 80286 with the 287 add-on chip. Awesome toy.
Might be collectible, what happened to the rest? (Score:2)
Were they binned or where does MSFT surplus go?
Re: (Score:2)
Adrian Black has a video of the boards themselves (Score:4, Informative)
0020 A rare Microsoft MACH-20 286 accelerator and more! [youtu.be]
cued up to the 18 minute mark
I never knew they made this.... (Score:2)
Interesting... I would have been in the target market for one, at one point in time. I was on a budget as a teenager and was pretty excited when I finally got my first IBM PC XT clone (after having used a Tandy Color Computer for years and a Timex Sinclair 1000 before that).
I really liked OS/2 when I first got my hands on a copy of Warp, and was a big advocate for it for a long time. I ran the less known "Merlin 4.0" edition of OS/2 as well. But eventually, IBM just withdrew support for the product to the
Sneakernet pro-tip (Score:1)
3 1/2" floppies were a lot easier to fit into a sneaker than the big 5 1/4" ones.
They weren't around in '87, but the micro-SD cards are so small they can fit between your toes.
Re: (Score:2)
Not new.