The Rise and Fall of the IBM PC Part 3 The Cloners Strike Back

Show video

Welcome to Another Boring Topic. When we left off in part 2 of this series, IBM  had just realized that they were rapidly losing   control of the PC market, a market  that they themselves had created,   and were in serious danger of becoming just  one of many PC compatible makers, albeit   one with much higher costs and significantly  slower R&D processes than their competitors.  The gold rush mentality that was driving explosive  competitive growth in the IBM compatible market   was so strong that even companies who had built  their business on entirely separate computer   systems were starting lines of IBM compatibles,  such as Commodore with its unimaginatively   named PC series and Tandy with the equally  creatively named Tandy line of compatibles. 

Commodore mostly kept its focus, insomuch  as Commodore’s dysfunctional leadership was   capable of keeping a focus on anything,  on continuing to sell the Commodore 64,   its successors, and its shiny new line of Amiga  computers, but there just was too much money out   there in the PC compatible gold rush to pass  up the opportunity to at least tap into it.  But it wasn’t just large and well established  companies churning out IBM compatible clones, new   upstarts were also mushrooming up, such as Dell,  founded out of Michael Dell’s college dorm room.  The end result of this was that IBM’s market share  of the personal computer market contracted down   to 38 percent in 1987, down almost half from  its high of 63 percent in 1984. And this is an   even more significant figure in light of the fact  that the entire personal computer market of 1987   still included major players such as Apple that  were not selling PC compatibles. Six years after   introducing the original PC, IBM was reduced to  merely being a major player in the PC compatible   market. And there was no sign of the bleeding  stopping anytime soon, with a constant stream of   new compatibles coming to market that undersold  IBM’s offerings while offering more features. 

By the mid 1980s the barriers to entry into the  PC compatible market had been made even lower by   thanks to the proliferation of OEM Asian  companies that manufactured complete IBM   compatible white label systems, that could be  bought cheaply by a US garage based startup,   who would then slap their own logo on them  and undercut IBM’s offerings even further.  This cannibalization of IBM’s  market share stemmed from two major   weaknesses in IBM’s control of the PC  standard. The first one was the open,   off the shelf architecture of the PC.  With the exception of the BIOS chip,   essentially the entire architecture was available  to anybody to copy, specifically the ISA bus.  As a side note, for consistency sake I will be  referring to the standard PC bus as the ISA bus   although it was variously known as the PC/XT  bus in its original 8 bit configuration,   the I/O Channel once it was updated to 16 bit  with the release of the AT, although it was then   usually called the AT bus, and then was finally  retroactively named the ISA bus by a consortium of   the various major clone manufacturers of the late  80s, and that’s the name typically used today.  The bus is how the various pieces of a computer,  such as disk drives, memory, and the processor   communicate with each other. The IBM bus was  hardly the first bus to be used on a successful  

personal computer, as every personal computer  on the market had one, from the Altair bus,   more commonly known as the S-100 bus, to the  Expansion Interface box that the TRS-80 used.   However the ISA bus was the one to meet broad  market appeal once the PC clone market took off as   compatibility with it meant that a hardware plugin  board would work on every single PC compatible.  As a reminder, IBM hadn’t really expected this  to be a problem since it assumed its proprietary   BIOS chip would prevent any cloners  from legally making clones of the PC.  

And illegal clones could be easily handled  by IBM’s massive phalanx of lawyers.  Sadly, the BIOS chip had not proven the obstacle  to cloning that IBM had assumed it would be. It   was bad enough when Compaq reverse engineered it  for its own products, but at least they kept it   for their own clones and didn’t sell it to others.  The worst was when Phoenix Technologies reverse   engineered the BIOS for sale to anybody that  wanted to build a clone without the expensive and   tedious process of reverse engineering the BIOS  chip themselves. IBM had also painfully discovered   that both Compaq and Phoenix Technologies’ cloned  BIOS copies were one hundred percent legal proof.  But this was just half of IBM’s problem. The  second major problem was the deal IBM had made  

with a certain rapidly growing company based  out of Seattle. Microsoft’s youthful leader,   Bill Gates, had successfully licensed MS-DOS  to IBM for the original PC, without selling the   rights to it. This enabled Microsoft to sell  MS-DOS to any other computer company that was   interested in purchasing it, and as previously  mentioned, there were many of them. Gates and  

Microsoft were making a fortune licensing  MS-DOS to every company that came calling.  Furthermore, Microsoft had a very strict policy  that under no circumstances would they ever lose   out on a sale. They would do whatever it took to  get a would-be cloner to sign on the dotted line,   prices were always negotiable and one Microsoft  employee remembers that there really was never   even a price list published for MS-DOS  as every deal was separately negotiated. 

A further annoyance to IBM was the fact that  the DOS operating system they had licensed   almost always had the letters “MS”  prior to it when people talked about it,   and their attempts to get people to refer to  it as “PC-DOS” instead had been unsuccessful.  IBM had to solve both problems simultaneously  in order to regain control of the PC standard.   They needed a new hardware solution that was  custom to IBM and protected by as many patents,   contracts and drooling lawyers as it took to  fully lock it down from unauthorized cloning. And   second, they needed to come up with an operating  system replacement for MS-DOS, one that they fully   owned and controlled, not licensed. An IBM  owned operating system would cut Bill Gates   and Microsoft out of the picture entirely. If IBM could successfully pull off this   two-fold attack, they would ensure that  future PCs would be built off of hardware   they licensed and controlled, and would  run an operating system that IBM owned. 

Any would-be cloner would have to come to IBM,  hat in hand, and license both the operating system   and the hardware it ran on from them, paying  substantial royalties for both. This would ensure   a steady stream of revenue flowing into IBMs  coffers, something that was increasingly needed as   the mainframe business they had built the company  off of was beginning to look...a little shaky.   Not that IBM had financial troubles yet, but it  didn’t take an accounting genius to see that it   would be difficult to sell mainframe computers  for 100 thousand dollars if their job could be   done just as well by a three thousand dollar PC.  Especially if that PC wasn’t even sold by IBM. 

Additionally, if IBM felt like it, they  could simply refuse to license the hardware   and software at all, ensuring that only  IBM computers were sold and completely   destroying the IBM compatible clone market. Steve Jobs did something similar to this ten   years later when he returned to Apple, destroying  the Macintosh clone market entirely and ensuring   Apple’s complete control of its hardware  and the software that ran on it. He did   this by refusing to issue new licenses to  cloners for Apple's new operating system,   Mac OS 8, forcing existing ones to keep using the  previous Mac OS 7, ceasing licensing hardware,   and buying back the clone license for  the largest Mac cloner, Power Computing. 

In the words of a 1997 Wired article: “The  other clonemakers...will be cut loose,   allowed to continue only with products for  which they currently hold licenses. They will   not be licensed to sell new Apple technology.  "We have no plans to give them Mac OS8 ... or   the new Apple-licensed proprietary designs". The end result of this was that Macintosh  

clone market ceased to exist in its  entirety within a very short span of time.   With the exception of a few rogue startups  like Psystar, that are usually sued out of   existence the moment they pop up on Apple’s  radar, there are no companies making Macintosh   computers apart from Apple. If you want to buy  a Macintosh computer running the latest version   of OSX, there is one company to get it from. So IBM’s strategy was not a bad one...in theory.  IBM’s solution to its first problem, replacing the  existing PC standard bus with a proprietary one,   was also a clever idea in theory. Since the  existing bus architecture was an open standard,   there would seem to be no way of collecting any  royalties whatsoever, nor of regaining control   of it. But IBM had a *insert Baldrick clip of  cunning plan* to fix this problem. They would  

create a brand new standard, called Micro-Channel,  that would be fully patented and owned by IBM. The   incentive to switch to it would be the fact that  it was far faster than the old ISA bus standard.   This plan was the brainchild of Bill Lowe, one of  the main figures responsible for the original IBM   PC who we met in Part 1 of this series. The flagship for this shiny hardware  

standard would be an entirely new line of  computers, split into four different models.  The first one would be an upgraded version of  the original IBM PC 5150, using the full 8086   processor instead of the somewhat crippled 8088  variant that the original PC had used. Apart from   that it would be fairly similar hardware wise, and  would still use the existing ISA bus, maintaining   compatibility with existing hardware add-ons. The next two models would be based around the   286 processor, a processor that Microsoft  chairman Bill Gates had called “brain dead”,   and the final one would be IBM’s first 386 based  personal computer. With the exception of the first  

model that was really just the original  5150 all over again and used the old bus,   the other three would all solely be available  with the new Micro-Channel standard. All four   models would also come with the new 3.5 inch  floppy drive as well, capable of holding far more   data than the older 5 ¼ inch floppy drives found  in the original PC and its AT and XT successors.  IBM also had replaced the BIOS chip that  Phoenix Technologies and Compaq had worked   so hard to clone, with a new BIOS called the  Advanced BIOS or ABIOS that would, in theory,   somehow be invulnerable to reverse-engineering. These new systems would be called the Personal   System/2s, or PS/2 for short, and IBM  spent years designing them and creating   and implementing the new Micro-Channel  specifications, as well as doing joint   development work with Microsoft as its junior  partner on the second piece of its strategy,   the new operating system that would replace  MS-DOS on the PS/2 line, the synergistically   named OS/2. More on OS/2 in a minute. In April of 1987, IBM publicly announced   and showed off the PS/2 family. In Blue Magic:  The People, Power and Politics Behind the IBM  

Personal Computer, a contemporary book released  in 1988, the authors give a good insight into   how the new systems were perceived at the  time, saying “For a change, the products   were truly impressive” but admitting several  paragraphs down that while the PS/2 line could   “use most software available for IBM-compatible  machines...it is not an industry-standard   computer. Its Micro Channel bus was not compatible  with add-in devices like the circuit cards   available for IBM-standard computers.” This was a major shift in the PC market,   and a very risky one. IBM was asking the entirety  of the PC marketplace to buy machines that,   for all their power, were completely  incompatible with any add-in cards   that the user had already purchased over the  previous six years. And this was no small ask. 

Today computers typically have only one or two  add-on cards installed, typically a video card.   Even further, the vast majority of computers  sold today have no add-on cards with video,   audio, and input/output options all integrated  directly into the motherboard. In the area of   laptops or tablets, it isn’t really even an  option to do further upgrades in many cases.  But the 1980s was a very different time for  personal computers, and motherboards of that   era frequently needed extensive plug-in cards  in order to do things that today we either have   on the motherboard or don’t even need. A few  examples of fairly common PC add-on cards would   be things like hard drive controller cards, audio  cards, Ethernet or token ring cards, modem cards,   MIDI cards, additional RAM cards, real time clock  cards, serial, joystick, or parallel port cards,   and many others. There were actually so many add  on cards needed for IBM PCs that some companies,  

like AST, got their start by designing and  selling multi-cards, a single card that integrated   multiple expansion cards into one drop in card. For example, AST’s popular SixPak Plus was an   expansion card that integrated six different  functions onto a single card, an RS-232 port,   a parallel printer port, a battery backed  calendar and clock, a game port, and an   additional 384 kilobytes of RAM in order to max  out the original IBM PC to its full 640k of RAM.   This early multi-function card catapulted  AST to 13 million dollars in sales in 1983,   and gave them the platform needed to expand into  other areas of personal computing, eventually   launching their own line of PC compatibles and  entering the Fortune 500 list in 1992 at spot 431.  As a side note, the 8 Bit Guy has a great  video on working tech support for AST that   I would highly recommend watching as AST was  a significant player in the 1980s and early   90s personal computer market, before they rapidly  went downhill and went defunct entirely in 2001.  With the PS/2 line and its Micro Channel  bus, IBM was asking people to throw away   years of investment and potentially thousands of  dollars, in many cases exceeding the original cost   of the actual computer itself. A person might  have bought an original IBM PC 5150 in 1981,   spent a bunch of money on add-on cards  and then upgraded to a Compaq Deskpro   386 in 1986, carrying forward all of his  expansion cards and adding some new ones. 

Additionally, to add insult to injury, the new  Micro Channel cards were not only incompatible   with the old standard, they also cost more. A  user would not only have to buy all new cards   to get most of the same features they had  in their non PS/2 machine, but they would   pay even more per card for the privilege. If it had been anybody other than IBM trying   to push this new incompatible standard  into the marketplace, it would have been,   in the words of a contemporary Byte magazine  article, “hailed as innovative, clever,   technically progressive, and destined to fail due  to lack of IBM compatibility.” But this was IBM,  

the creator of the PC standard in the first  place, and the largest and most influential   computer company in the world. If they could  successfully move everybody over to the PS/2 line,   from a hardware perspective there would be no more  clones cutting into their market share. The whole   PS/2 architecture and especially the Micro-Channel  bus were all fully locked down and proprietary.  Any company who tried to copy it without  paying IBM’s steep licensing fees would be   sued into oblivion, and the licensing fees  would be high enough that it would be very   difficult if not downright impossible for  a would-be cloner to undercut IBM on price.  

While IBM’s standard licensing agreement in the  past had been a mere one percent of the final   product’s selling price, the new licensing  agreement would require a royalty of up to   five percent of the final selling price, adding  about 150 dollars or so to every PS/2 compatible.  Furthermore, IBM also had what it considered to  be a clever trick up its sleeve. To date it had   been unable to match the frantic development  rate of the cloners, who brought out entirely   new generations of computers every 12-18 months.  Unable to match this speed thanks to corporate   inertia, IBM thus kept getting beat to market by  cloners such as Compaq, who beat IBM to market   with a 386 based system by almost seven months. Sure the new Micro Channel bus was locked down,   but IBM decided to give itself a leg up and  get twice the computer generations for the   same effort. And so, the first generation of PS/2  computers would ship with the Micro Channel bus   deliberately hobbled to only half speed.  This would give IBM an easy way to,  

in the words of Robert Cringely’s Accidental  Empires, “discover all sorts of neat additional   Micro Channel horsepower, which would be presented  in a whole new family of machines using what   would then be called Micro Channel 2.” They could produce a second generation   PS/2 line that was twice as fast as the  original, with essentially zero effort,   allowing IBM to compensate for far lengthier  development processes. In IBM’s mind, there was   no way this strategy could possibly backfire on  them. They were confident that the consumer market   would follow them to the brand new standard,  cutting the legs out from under every cloner   and putting them in a position to either go out of  business or come to IBM, cap in hand, and beg for   a license to use the Micro Channel architecture.  A license which may or may not be forthcoming. 

The 1987 PS/2 reveal was accompanied by the  reveal of OS/2, IBM’s next generation operating   system that would be replacing Microsoft’s DOS  operating system and the second part of their   plan to regain control of the PC market. But IBM  picked a funny way to create MS-DOS’s replacement,   as they partnered with none other  than Microsoft to develop it.   The reasons for this strange partnership are  somewhat unclear. One contemporary rumor was that  

Bill Gates had heard that IBM was going to develop  a replacement for MS-DOS alone, and had threatened   to pull IBM’s license for MS-DOS if Microsoft  was not included in the development process.  Another possibility is that IBM was wary of  going at it alone in the first place, with   their confidence somewhat shaken by their failed  TopView system, an MS-DOS replacement developed   solely in-house that they had brought to market  in 1985, where it had promptly crashed and burned.  TopView is itself worthy of a full video, but for  the sake of time we will just hit its highlights.   It was a shell that ran on top of MSDOS, allowing  for limited cut and paste between programs,   as well as giving MSDOS a basic form of  multitasking. In theory, the user could  

open WordPerfect and work on a document while  waiting for Lotus 1-2-3 to finish recalculating   a spreadsheet. In practice however, TopView  required such a massive amount of resources   that multitasking really only worked on very light  applications, and even then it was very sluggish.  Additionally, since TopView was solely character  based, it was unable to do a true windowing based   GUI, since things like mixing graphics  and text together or displaying varying   fonts on screen inside of a document were not  possible with a solely character based system.   A quick note: although some of my sources  state that TopView was character only, the   Wikipedia entry for TopView says that it did have  a graphics mode, it just was almost never used.  So far as the user was concerned, they were still  using MSDOS, only it ran considerably slower   than standard MSDOS, required a more powerful  computer, and didn’t really add much to their   user experience. TopView also certainly didn’t  resemble the new hotness of Apple’s Macintosh   in any way whatsoever. Windows, although in an  embryonic form in 1985, with version 1.0 coming  

to market in November, at least sort of looked  like a Macintosh if you squinted hard enough,   and required 99 dollars to buy. Even still,  interest was very low and Microsoft sold   very few copies of Windows 1.0. What few people were interested   in a more advanced MSDOS bought Windows and  TopView was left to languish in obscurity.   A few companies, such as MicroPro, invested a lot  of time and resources into developing programs   that used TopView’s abilities in much better  ways than standard MSDOS programs, and they got   badly burned with its failure as well. The press  did not give TopView any good publicity either,   with reviewers such as John Dvorak labeling it  “TopDog” and not in a complimentary fashion.  In Merrill Chapman’s entertaining book In Search  of Stupidity, he recounts an entertaining personal   story from when he worked for MicroPro and  had to demonstrate a TopView version of   their word processor, Easy, at a conference. A bunch of Macophiles from the press had for  

some reason shown up and taken a great deal of  pleasure in torturing me during demos of Easy.   “Show us again how you cut and paste text,” they’d  say. “You call that easy?! Bwahahahahahahahaha!”   They brayed as I banged on a keyboard instead  of brandishing a mouse. “Now how do you display   a font onscreen? You can’t?! Wow! We can see  why TopView is so fabulous!” they howled as   they held their sides and laughed hysterically. By the end of 1985, it was clear that TopView   was a complete failure. As Chapman states  when recounting a conversation he had with   upper management at MicroPro, sometime  after his disastrous Easy demonstration:  “Customers don’t like it. They’ve seen the  Mac; that’s what they want on their PCs.  

The press hates TopView; they think the Mac  is where it’s at. Developers hate TopView;   they want to make cool Mac-like things  for PCs. Everyone hates TopView!”  Faced with such a monumental failure, IBM’s  confidence was probably at least slightly shaken.   No matter how much IBM loathed their dependency  on Microsoft, most likely they decided that they   would be setting themselves up for failure  if they tried to develop a DOS replacement   all on their own again. A better plan in IBM’s  mind, was to utilize Microsoft's expertise to  

co-develop a successor MSDOS, but ensure that  it was one that IBM owned in its entirety.  Whatever the reasons for their decision may  be, the fact was that on August 22, 1985,   IBM and Microsoft signed a “Joint Development  Agreement” that stated that the existing MS-DOS   operating system would continue to be supported  while its replacement was jointly developed,   a replacement that was called “CP-DOS” in  the JDA, but would eventually be named OS/2.  IBM may have been developing OS/2 jointly with  Microsoft, but they made it very clear that OS/2   was their product and Microsoft was at best a  junior partner. IBM would retain full ownership  

of OS/2, with the right to license it out if  they chose or freeze out the cloners entirely.  OS/2 was going to boot MSDOS out of the  game entirely, relegating it to obscurity,   hopefully along with its annoying founder and  company. In the words from a 1988 Byte article   called “OS/2 Dreams”, it would be “either  the operating system that will take us   into the twenty-first century, or the operating  system that will take Microsoft into Chapter 11.”  Byte missed the fact that as far as IBM  was concerned, they wanted OS/2 to do both,   and saw no reason why it wouldn’t. If Microsoft  couldn’t license out MS-DOS anymore as everyone  

switched to OS/2, then their biggest cash cow  would be dead, and they would be relegated   to competing with application software or  games, just like any other software company.  Fed up with seeing a steady stream of royalties  roll into Microsoft’s coffers from licensing   out MS-DOS to the very cloners that were killing  IBM’s market share, they were determined to not   only put an end to this torrent of cash, but  also to make sure that OS/2 corrected all of   MSDOS’s manifold weaknesses and was designed to  the highest IBM standard of software development.  Since IBM gauged its programmers productivity  by the KLOC, or per thousand lines of code, OS/2   was certainly going to have an enormous amount of  custom code and would be as far away from a lean,   mean operating system machine as possible. The full development and history of OS/2   warrants a completely separate video of its own,  so we will just focus on its early history here.  

OS/2 was planned to be a multithreaded,  multitasking operating system that could   handle up to a whopping 16 MB of RAM, a  number far in excess of MSDOS’s puny 640k.   And yes it's true that by the mid 1980s you could  have more than 640k of RAM in your PC compatible,   but then you had to struggle with software and  compatibility headaches as you used a clunky   workaround like expanded memory, which clunkily  allowed for access to extended memory above the   640k by means of bank switching RAM. A great  read on the details of the 640k problem and the   workarounds developed is over at filfre, check  it out from the link in the description below. 

IBM’s lofty ambitions for OS/2 immediately clashed  with the reality of two companies with wildly   different cultures struggling to mesh together  to jointly develop a massive piece of software.   When it came to the original MS-DOS, IBM had  simply licensed it from Microsoft, they did   no development on it themselves. But for OS/2,  with IBM and Microsoft both writing the code,   trying to coordinate who was working on what,  and keeping everybody focused and working in the   right direction was an almost impossible  task. When development of OS/2 began in   1986, immediate cracks in the partnership’s facade  began to show almost immediately. The corporate  

culture clash continually exacerbated enormous  tensions that frustrated everybody involved.  In the words of Neal Friedman, one of Microsoft’s  programmers: The project was extremely frustrating   for people at Microsoft and for people at IBM too.  It was a clash of two companies at opposite ends   of the spectrum. At IBM, things got done very  formally. Nobody did anything on their own.   You went high enough to find somebody who  could make a decision. You couldn’t change   things without getting approval from the joint  design-review committee. It took weeks even to fix   a tiny little bug, to get approval for anything. Development crawled along at a glacial pace,  

and matters were not helped along by the fact  that although IBM and Microsoft were partners,   the final word on all design decisions and  differences of opinion always came from IBM as   Microsoft was clearly viewed as the junior partner  in the relationship. And IBM was constitutionally   incapable of making rapid decisions, or reaching  any sort of consensus in a timely manner. Rather,   their byzantine policies and procedures and  plethora of sign-offs from multiple levels   of the bureaucratic hierarchy, ensured that  any decision consumed enormous amounts of   time. And when building an operating system,  there are many, many, many decisions to make.  

And every one ate up an inordinate amount of time. Matters were not helped by IBM’s decision to   incorporate what was known as SAA or “Systems  Application Architecture” into OS/2. The idea   behind SAA was that OS/2 would be somewhat  of a cross platform operating system,   capable of running on the full range of IBM  products, from personal computers all the way   up to mainframes with either no or only  minimal changes needed. This would allow   programs written for OS/2 to run on any IBM  machine large or small, giving IBM a further   lock on the market from mainframes to micros. There was some precedence for this thinking,   after all IBM had seen enormous success with  the 360 line of computers back in the 1960s,   a full range of computers, although obviously  not including PCs, that all ran the same software   and were compatible with each other, making  moving software from one 360 system to another   very straightforward. But times had changed  dramatically in the past couple decades,  

and the market was moving away from expensive  high end mainframes with expensive support   agreements and heavily marked up hardware,  towards fleets of cheaper PCs linked together   with network cards and a shared server. Trying to shoehorn SAA into OS/2 would   not only dramatically increase the development  time and workload, but would be somewhat similar   to designing a new car that was also capable  of being hooked up to a team of draft horses,   just in case you wanted it to be compatible with  older forms of transportation. Still, IBM insisted   on this and thus work on OS/2 dragged on for  years at an ever slower pace while the code base   for OS/2 expanded at an exponential rate thanks to  feature creep and IBM’s own development processes. 

In 1987 OS/2 finally limped onto the market. And  what had all of the development work produced   after years of effort? A massive resource hog  that, although under the surface contained many   improvements over MSDOS, presented the user with  the same cryptic command prompt that MSDOS did.   All of OS/2’s fancy new features and design  were not immediately evident as the project   had gotten so far behind schedule that the actual  graphical part of OS/2 was not ready to ship with   the rest of it, and thus the initial release  was command prompt only. But even though it   was command prompt only, without the overhead of  supporting a full bitmap graphical user interface,   OS/2 was extremely slow. And  not just slow, just to run it   at all required a massive increase in  RAM, hundreds of dollars worth of it.  For an additional 340 dollars, more than three  times the cost of MSDOS, which most likely came   bundled with your computer anyway, you got an  operating system that according to Byte’s tests,   required 5 to 8 megabytes of RAM to run  comfortably. This, in an era when most personal  

computers rarely had more than 1 to 2 megabytes  of RAM, and many Apple IIs and Commodore 64s were   still being sold with only a fraction of that. The  pricing also ran contrary to the expectations IBM   itself had set six years earlier, when it released  MSDOS 1.0 for a retail price of forty dollars.  Another source, Merrill Chapman’s In  Search of Stupidity says that OS/2 needed   “4MB of memory to be useful, 8MB to step  along smartly, and 16MB to really hum.”  And as a reminder, if you really wanted a GUI for  your compatible, you could always buy Windows for   100 dollars, version 2.0 of which was released in  December of 1987. And Windows 2.0 only required   a minimum of 512k of RAM to run, although  as usual, anybody who tried to run it with   the minimum specs quickly wished they weren’t. The Macintosh itself had been released with only  

128k of RAM, which was barely sufficient  but at least its GUI operated in it,   and it's quickly released successor,  the Fat Mac, had a total of 512k   which allowed the Mac OS to run extremely well. Of course the Mac OS was not a multitasking OS   at that time, and it was only black and white,  however it was a fast and easy to use operating   system so long as you respected its limits. IBM  wanted users to pay 340 dollars for an operating   system that potentially required 1600 to 2800  dollars of further investment in additional RAM.  That was a lot of money, equivalent to 3400 to  6000 dollars in 2020. If Microsoft had decided   to release Windows 10 with the requirement that it  needed at least 128 GB of RAM to run comfortably   and really should have at least 256 GB, it  probably would have struggled in the marketplace.   Although one tangible benefit of running Windows  with that much RAM would be the ability to open at   least four concurrent tabs in Chrome, something  that many people would doubtless appreciate. 

IBM also did nothing to clear up the confusion  many people had with the OS/2 and PS/2 naming   scheme, where people assumed that OS/2 would  only run on the PS/2 line of computers.   Of course IBM had no interest in clearing  up that confusion when the double punch of   OS/2 and PS/2 meant that people would of course  only be buying PS/2 personal computers anyhow.  A number of companies did hop onto the OS/2  bandwagon however, primarily smaller companies   that were hoping to launch into a less crowded  marketplace than the MSDOS world, or companies   that were beginning to feel that dealing with  IBM was preferable to dealing with Microsoft.  There were a few heavy hitters in the  latter group, such as Ashton-Tate and Lotus,   who fully committed themselves to  OS/2 development, and ignored Windows.  This would wind up painfully  biting them in the long run.  The PS/2 line was launched into the marketplace  with a muddled ad campaign that was the exact   opposite of their extremely well received  ads for the original PC. Whereas the ads that  

introduced the original PC to the world featured  Charlie Chaplin’s famous Tramp character and were   extremely clear and easy to understand, the PS/2  line would be introduced by...a number of cast   members from the famous TV show MASH. Although  they did not appear in character, Harry Morgan,   Loretta Swit, Gary Burghoff, Larry Linville,  William Christopher, Wayne Rogers, and Jamie   Farr appeared in a series of TV commercials and  print ads extolling the wonders of the PS/2 line.  Although developed by the same marketing agency  that had developed the original Little Tramp ads,   Lord, Geller, Federico, Einstein, the contrast  between the work they had done years prior   with the original IBM PC commercials and  the new PS/2 commercials was striking.  The Tramp commercials had worked well due  to the fact that they featured a single   extremely recognizable character played by Charlie  Chaplin, they did not feature the actor himself. 

In an AP article on the new ads, the  following odd justification is given:   Richard Lord, chairman of the ad agency, said the  company and the agency were uncertain whether the   Chaplin character ″could carry all the products″  being introduced in the new line of computers.   He said the idea of computer ″connectivity,″  or the ability of computers to work together,   had also become a more important  consideration than it had been six years ago.  Using a group of actors in the ads, Lord said,  would give the admakers the opportunity to show   computers being used in a variety of situations. Having most of the primary cast of MASH   do advertising, but not as their famous  characters, was a very odd marketing choice,   and one that left many people unmoved.  At least it wasn’t actively insulting the   audience, as Apple’s infamous “Lemmings”  commercial had done several years prior.  Ultimately, IBM felt that they had been so  poorly served by their advertising agencies,   that they eventually terminated their relationship  with all of them, shutting down over 500 million   dollars a year in ad accounts. But that was  well after the damage had already been done. 

In fairness, bad marketing was undoubtedly the  least of IBM’s many problems with the PS/2 line.   Even good marketing would have struggled  to deal with driving consumer demand for   a product that was not only expensive,  but also was completely incompatible with   all of their expensive hardware add-ons. For some reason, IBM failed to think about   the problems that this would incur, especially  when the first generation Micro Channel bus was   actually slower in many cases than many clones  using the old ISA bus. Trying to make consumers   buy a new computer that required them to give  up their existing investment in add-on cards   and buy all new replacement add-on cards that  were not only more expensive than the old ones,   but were actually slower as well seems  more like a Dilbert cartoon than an actual   strategy but that was what IBM decided to do. In fairness, the PS/2 line did require somewhat   fewer add-on cards than its predecessors,  as PS/2 motherboards had a decent amount   of functionality built in that previously  required separate add-ons. Most significant  

was putting video out directly onto the  board, with the new Video Graphics Array port,   or VGA that was built into all PS/2 motherboards. However most users who had expanded their ATs   would still have needed to, for example,  replace their ISA sound cards with new   MCA ones. And some users would have just  been out of luck entirely, as some of the   more esoteric ISA add-on cards did not exist  in Micro-channel variants, and never would. 

Furthermore, IBM decided that not only would PS/2  cloners have to pay license fees of 5 percent on   every Micro Channel clone they sold, they would  also be required to pay a retroactive 5 percent   license fee for every single PC clone they had  previously built using the ISA bus. Cloners would   have to cough up large sums of money to not only  use the new architecture, but would also have to   retroactively pay IBM millions for computers  they had already built and sold using the open   PC bus that IBM did not own the rights to. At  this point, even Dilbert’s Pointy Haired Boss   might be saying that this sounded like a bad idea. Bill Lowe, head of IBM’s Entry Systems Division,   started 1987 by confidently waiting for cloners to  approach him about licensing. Publicly, he stated  

that any cloner who tried to reverse engineer  Micro Channel would need to either have great   engineers or very good lawyers. He was confident  that it would be neither feasible nor legal to   clone it, and thus the would-be cloner would be at  his door trying to get a license, which IBM would   probably be unwilling to give, at least initially.  As the months went by and no cloner made   any overtures, Lowe started publicly saying that  IBM might be willing to license the design out to   cloners for the right terms. Still nobody came. By 1988, Lowe was, in the words of one author   “practically begging other companies to  adopt the Micro Channel. How could IBM claim  

to own the new standard if nobody else used it?” It wasn’t that the cloners were ignoring the PS/2   line and Micro Channel, Compaq even quietly did a  decent amount of work towards engineering a clone,   but without wide market adoption, there was little  incentive to quit producing machines using the old   ISA bus. And those machines were what was selling,  with the broader consumer market treating the PS/2   line and its complete incompatibility with their  existing hardware with complete indifference.  Still, if Micro Channel had been running at its  full speed, it would have been at least twice as   fast as the older bus, and that might have been  enough to overcome IBM’s poor decision making   by making it attractive enough for consumers and  businesses in broad numbers to stand for the extra   expense and buy the new machines and rebuy all  their hardware expansion. But hobbled as it was,   plus the massive fees to use it and the  forcible extraction of royalties that IBM had   no legal right to...the existing cloners were  up in arms and they had the powerful force of  

the broader consumer market behind them. The  consumer market wanted “PCs based on the powerful   new Intel 80386 chip or they wanted good old  reliable machines along the lines of IBM’s AT.”   And IBM was offering neither option in a format  the market wanted as the PS/2 386 line was   incompatible with consumers existing hardware  and the popular AT had been killed off by Bill   Lowe. Yes, IBM actually killed off their top  selling computer, the AT, in an attempt to force  

market adoption of its incompatible replacement. Shockingly this bold strategy didn’t work out,   especially because the market could still buy  plenty of AT clones from Compaq, Dell, or any   of the plethora of other cloners. Fortunately  for IBM, Lowe’s cancelation of the AT in North   America did not affect IBM’s operations in  Europe, where Bill Lowe’s previous boss,   Mike Armstrong, had recently moved to take  over IBM’s European branch. Armstrong wisely  

kept the AT in production, and it continued to be  highly successful in Europe for quite some time.  IBM was correct in diagnosing the problem with  the ISA bus, with new faster processors like the   386 and its 32 bit architecture, a new bus was  definitely needed in order for that computers   equipped with the 386 and future 32 bit processors  would not find themselves hobbled by a 16 bit bus.   IBM had failed to make Micro-Channel that new  standard, but there was another company with a   replacement bus that was ready to take its own  stab at the problem. And its success would see   IBM’s chances of regaining control of the PC  standard permanently slip away from its grasp.  Unsurprisingly, the company that would lead  the rebellion against Micro Channel was none   other than Compaq. Compaq had been quietly  working on a replacement bus that was not only  

as fast as Micro Channel, but was also fully  backwards compatible with the legacy ISA bus,   the very thing that IBM had said was  impossible. Using this proof of concept,   Compaq reached out eight other significant  cloners, creating an uneasy alliance that the Wall   Street Journal called The Revolt of the Clones.  This alliance’s goal was to take on IBM and force   the adoption of a Micro Channel alternative. This consortium was comprised of Compaq, AST,  

Epson, HP, NEC, Olivetti, Tandy, Wyse Technology,  and Zenith Data Systems and had a combined 1987   market share of 33.5 percent of the PC market.  Although led by Compaq, this was definitely a   fractious alliance of fierce competitors,  brought together solely due to the need   to present a united front against IBM and avoid  being put out of business by IBM’s new standard.  Out of this unusual alliance of bitter competitors  emerged a new bus standard, the Extended Industry   Standard Architecture, a full 32 bit bus that did  what IBM claimed was impossible...it was fully   backwards compatible with the old AT ISA bus. IBM had now definitely lost control of the PC   hardware standard and they would never regain  it. IBM’s attempt to break the cloners had   backfired badly, as one source bluntly put it:  “(Bill) Lowe’s PC strategy had been a torpedo   below the waterline at IBM. No one knew it  by the end of 1988, but IBM was sinking.” 

IBM had lost its battle to force a new industry  hardware standard and regain control of its   hardware. However, if it could succeed with  OS/2, it could at least regain control of   the operating system that every PC ran. Given  the rapidly decreasing margins on hardware,   controlling the software the hardware ran on would  be a great consolation prize, and would allow IBM   to replace Microsoft. Software had extremely high  margins, and OS/2’s success would firmly relegate  

Microsoft to merely being an application  company with its productivity applications   such as Word and Excel, probably a large and  influential one, but not the dominant one.  Although interest was initially high for OS/2,  its repeated delays prior to its first command   line only release, plus even more delays in the  release of its graphical user interface, plus the   massive cost of the memory required to run it,  meant that people kept right on using MS-DOS,   and many started to pay more attention to Windows. And IBM further shot itself in the foot with   its stubborn refusal to leverage its vast RAM  production capacity to help further the installed   base of OS/2 and get it to a critical mass. Given  the massively prohibitive cost of the RAM required   to run OS/2, IBM could have started bundling OS/2  for free or massively discounted with RAM sticks.   In the words of a product manager for  an OS/2 word processor called DeScribe,   “OS/2 without memory was a one thousand dollar  upgrade. Bundled with a handsomely discounted   4MB memory stick, it was a million copy seller” Because the price of RAM had quadrupled in the   late 80s, with 1 MB sticks going from 100 dollars  to 400 dollars, IBM would have profited handsomely   from giving OS/2 away with RAM and driven market  adoption of OS/2. Sadly for IBM, nobody at the  

top ever thought about adopting this strategy. On May 27th, 1988 Microsoft even released a   special version of Windows 2 that took advantage  of the powerful new 386 processor to allow for   multi-tasking multiple DOS programs, if you had  the RAM for it. It also gave the user many of the   features of OS/2, although it was still behind the  Mac, Amiga, or even the Atari ST in many areas.   Although still three years away from the 3.0  release that would rocket it to the moon,   Windows 2 gave the user a GUI that ran well  in a single megabyte of memory and only   required 512 kilobytes as minimum specs. For users who wanted a GUI for their DOS machines,  

they could pick OS/2 which cost a fortune and  required a larger fortune in RAM to run well. Or   they could go with Windows, which was reasonably  priced, pirated enough that many got it for free,   and ran acceptibly, especially if you had a  blazing fast 386 machine. Microsoft didn’t   really care too much about Windows being pirated  back then, the more people using it the more   market share it got, and Microsoft made most  of its Windows money licensing it out to OEMs   for preinstall on new systems anyhow. OS/2 development was also problematic,   as the pricing for the software development kit or  SDK had been set at 3000 dollars, a price greater   than even Apple’s SDK’s and one that stood in  stark contrast to the Window SDKs, which Microsoft   practically gave away. For large companies  this was not a huge problem, but for small  

companies or individual programmers, the initial  investment required was a substantial barrier.  IBM also failed to set up any programs to help  promote OS/2 products, making no attempt to   secure shelf space on retailer’s shelves or help  software vendors market directly to OS/2 users.  One other problem with OS/2 is worth mentioning.  When IBM finally released OS/2 1.1 in October   of 1988, with the Presentation Manager  included, finally giving OS/2 its graphical   user interface and making it look far more  polished...it couldn’t print. More specifically  

it couldn’t print to any non-IBM printer. IBM  had made no effort to get drivers for any other   printers besides the ones it sold, and most  people who had printers didn’t have an IBM one.  Printers in the 1980s were far more expensive than  today, and represented a significant investment   for the user, especially if they had one of the  new laser printers. This meant that a new user  

of OS/2 was not only confronted with needing to  replace all of their hardware, buy OS/2 for 340   dollars, pay a small fortune for enough RAM to  make it usable, and deal with a pronounced dearth   of OS/2 specific software...but they also  needed to buy a brand new printer from IBM.  Windows was looking more and more attractive  by the day, Windows 2.0 started selling 50,000   copies a month. And Windows 2.0 could actually  print to non-IBM printers of course. In October   of 1989, IBM released OS/2 1.2 which still  struggled to print to non-IBM printers.   Sales of Windows 2.0 rose even further thanks  to this, and then in 1990, Microsoft released   Windows 3.0 which promptly sold two million  units in six months and pretty much sealed  

OS/2 into a permanent vault of obscurity. Although the PS/2 line of computers was a   market failure, there are a couple hardware  holdovers from it that continue to the present   day in some form. First of all was the  PS/2 serial port for a mouse or keyboard,   this small analog port, typically purple for  keyboard or green for mouse connection is still   found on some towers even today. Second was the 15  pin VGA monitor connector, which became a standard  

for connecting a monitor to a laptop or tower,  and is also still found on some towers today,   although it has essentially been entirely  superseded by HDMI or the somewhat older DVI.  OS/2 actually soldiered on for well over a decade  after its initial release, its final release   was in December of 2001, with IBM’s support of  existing installs continuing until December 31,   2006. Never quite managing to break through IBM’s  incredible ability to shoot itself in the foot,   it still attracted a dedicated user base that was  legendarily devoted to it and its stability and   reliability. Scott Adams even refers to OS/2 users  in one of his Dilbert comic strips from 1995,   poking fun at the fact that OS/2 die-hards  were even more fanatical about their   operating system than Macintosh users were. That wraps up part three of the rise and fall   of the IBM PC. I had a lot of fun researching  each part and writing the scripts, and I hope   you enjoyed it. The next computer history series I  will be starting sometime in the next year will be  

an in-depth look at the rise and fall of OS/2,  possibly interwoven with the rise of Windows.   Both subjects have been touched on in this series,  but both deserve their own stand alone videos.  If you want to see more videos along  these lines, drop a comment below   with what computer history subject you would  most be interested in seeing. And as always,   thanks for watching! If you enjoyed this video,  please leave a like and consider subscribing.

2021-01-05

Show video