The complete history of the IBM PC

Bill Gates. Mysterious deaths. IBM trying to act like a nimble startup. This story has it all!

Read the whole story

 

[althaz]

Alright, I'm hooked. NOW GIVE ME PART TWO!

I can't believe the cliffhanger you have ended this story on you bloody evil genius.

If anybody needs me I'll be refreshing the Ars homepage non-stop for a while apparently.

 

[wigry]

One little gem found from the web giving a unique view to the creation of IBM PC. Why Intel? Why 640KB RAM?

http://tech-insider.org/personal-comput ... 90/09.html

 

[scoobie]

and all that was going on in my final years at school and soon after started working for Tandy/Radio Shack..... The UK101 and BBC Micro were in the ascendancy. Hindsight is a wonderful thing......

 

[Trippynet]

Great, detailed and interesting article! Reminds me of the fun I had reading all of Jeremy Reimer's Amiga articles a few years back. I'm very much looking forward to seeing part 2!

 

[SnakePlisskenUK]

The Sams/Kildall meeting (or non-meeting) is one of the great "what if?"s of tech history. Taking that to its extreme, no Microsoft behemoth, no IE, no Windows. The internet would still rise, but the way we access and use it would probably be quite different.

A true "Sliding Doors" moment.

 

[HellDiver]

UncleToad[/url]":2swdab1x]This is all very interesting, but it misses the parallel developments going on in the UK and Europe at that time. The history of computing of personal computing is not just USA based.

This is the story of the IBM PC. It is not the history of the Personal Computer.

 

[awelux]

"then IBM sent a couple of sales reps to discuss your needs."
I love that sentence. IBM personel was infamous for not arriving alone.
The "IBM bus" arrived was meant quite literally.

 

[TheGouldFish]

Hope this feature takes less then the Amiga one to push out each part

 

[Baron von Robber]

Autoexec.bat
config.sys

Copy - Del - Copy - Del.....what's this new command? MOVE?!


/endflashbacks

 

[Jim Z]

Unfortunately the IBM PC arrived too late to locate the golden tickets.

 

[Me Myself And I1]

In 1998 I acquired a 1987 version of the 5150, it was 20MB drive model and used to be a workhorse at the US energy dept.

It ended up behind what what used to be the Iron curtain courtesy of some US church supporting brethren in new European democracies.

It was my first computer and it essentially allowed me to experience the PC revolution from the very beginning albeit a decade or so later (it was a fast-track after that).

What a fun that machine was Single density 5 inch floppy, black-green screen that waved so much it would induce dizziness after prolonged use, 110/220 volts adapter was housed in a cocoa box, which wasn't washed properly so every time it heated, the room was filled with the smell of hot chocolate

I installed Volkov commander and wrote my thesis on that machine.

The only HW upgrade was a single density 3.5 inch drive which was a pain in the ass to fit because what was a standard size in 1998 was not what IBM used back then.

I missed those days

 

[kefkafloyd]

The IBM 5150 predated Van Halen's album 5150, and even then Van Halen's 5150 was referencing a section of California law.

Now you know.

 

[Justin Credible]

The IBM 5150 predated Van Halen's album 5150, and even then Van Halen's 5150 was referencing a section of California law.

Now you know.

Thanks for the little useless tidbit of information, i love that kind of stuff!

By the way they weren't Van Halen at the time, they were Van Hagar

 

[rorix]

"The machine that would become known as the real IBM PC begins, of all places, at Atari. Apparently feeling their oats in the wake of the Atari VCS's sudden Space Invaders-driven explosion in popularity and the release of its own first PCs, the Atari 400 and 800, they made a proposal to IBM's chairman Frank Cary in July of 1980: if IBM wished to have a PC of its own, Atari would deign to build it for them."

The machine... began?

Atari would design...

Deign to,

deɪn

verb

do something that one considers to be beneath one's dignity.
"she did not deign to answer the maid's question"

 

[masospaghetti]

Even now, the die shot of an ancient processor like the Z80 is amazing to look at.

 

[malor]

One thing that really gets me about the PC is just how bad it was. It was a terrible, terrible computer. Almost everything about it was kind of crippled, to the point that the 8088, running at 4.77MHz, was functionally just about the same speed as the 1MHz Apple II.

But: nobody really knew that. Nobody understood computers back then, or hardly anyone.
I was a teenager at the time, and while I could use a PC or an Apple perfectly well, and program them in BASIC and hook up peripherals and that sort of thing, I didn't really understand them at a deep level. I was pretty advanced as users went, and if I couldn't tell you why the PC kinda sucked, Lord knows not that many others could.

It was an abysmal base to build on, and yet, with all the money flowing into that architecture, they were able to bend it, and bend it, and bend it, until it actually got pretty darn good. It's kind of like a stunted, twisted little seedling that was gradually coaxed into a mighty tree, while much better, stronger seedlings died out due to neglect.

There's this tendency by later writers to sort of enshrine the PC, to make it better than it was, because it won the war. In almost every way, it was a horrible computer. But it had two enormous competitive advantages: it was fully documented and open, and IBM was selling it. Those two things are why it won, and almost everything else about the computer is why the war lasted as long as it did.

 

[Jim Z]

The IBM 5150 predated Van Halen's album 5150, and even then Van Halen's 5150 was referencing a section of California law.

Now you know.

Thanks for the little useless tidbit of information, i love that kind of stuff!

By the way they weren't Van Halen at the time, they were Van Hagar

 

[biscuitsandcookies]

"The machine that would become known as the real IBM PC begins, of all places, at Atari. Apparently feeling their oats in the wake of the Atari VCS's sudden Space Invaders-driven explosion in popularity and the release of its own first PCs, the Atari 400 and 800, they made a proposal to IBM's chairman Frank Cary in July of 1980: if IBM wished to have a PC of its own, Atari would deign to build it for them."

The machine... began?

Atari would design...

"Deign" means "to demean oneself" or to "stoop down", which I agree isn't quite appropriate here (who would find it beneath their dignity to work with IBM?)

It's hyperbole on the part of the author, casting Atari as brash and arrogant. As the article details, many of the early upstart PC vendors were simultaneously terrified of IBM's size and resources and condescendingly dismissive of its corporate culture.

 

[Old_Fogie_Late_Bloomer]

Bill Gates, mysterious deaths, and the business machine that sparked a home revolution.

Now that's what I call a lede!

 

[Jim Z]

One thing that really gets me about the PC is just how bad it was. It was a terrible, terrible computer. Almost everything about it was kind of crippled, to the point that the 8088, running at 4.77MHz, was functionally just about the same speed as the 1MHz Apple II.

depending on what you needed to do, popping an 8087 into the next socket could give it a real boot in the ass.

and if you had the scratch, you could get the "Professional Graphics Controller" which was pretty much another IBM PC on a card (three cards, actually)

 

[coniferae]

There's this tendency by later writers to sort of enshrine the PC, to make it better than it was, because it won the war. In almost every way, it was a horrible computer. But it had two enormous competitive advantages: it was fully documented and open, and IBM was selling it. Those two things are why it won, and almost everything else about the computer is why the war lasted as long as it did.

It would be really interesting to learn more about those that did not win the war. Do you know of any examples?

 

[yet_another_wumpus]

In 1980 Jay Miner (designer of the 400/800 and Amiga) was likely still working at Atari. If IBM would get a Jay Miner job, it certainly would be more impressive than copying the Intel applications notes (much of the "open system" was likely that IBM would have a hard time claiming that a system almost entirely based on published Intel systems could be proprietary [but then, look at those papers IBM wanted Gates and Kindall to sign. They might have gotten away with it].

The 6502 was certainly "based" on the 6800, but it was pretty much stripped down as far as it could do (mostly in the "hot rod" sense. The goal was to optimize the execution with the memory bus, and strip out anything that couldn't keep up. Having a low cost thanks to such simplicity was also critical). I'm not as familiar with the z-80.

 

[Khaaannn]

The really terrible thing about the IBM PC was the abysmal 808x architecture with it's "Calculator Chip" heritage still intact.

The segmented memory architecture, the IRQ-hell, and masked interrupts, it was crap from the start and hamstrung the "PC" growth terribly.

It wasn't until the 80286 came out that some of these issues were "fixed" and even then it was hell to program around all the limitations and gave rise to the AHCI interface that hides those architectural flaws.

The 680x series on the other hand was a breeze to program for. Flat memory architecture, no IRQs, etc. (everything is just a memory address) but Apple, typically, sat on their laurels and Steve Jobs eschewed anything "Business" and they were seen as a "toy computer" by the suits.

Oh what could have been...........

 

[Khaaannn]

Dammit! Cool articles like this keep me coming back to Ars in spite of the insufferable climate change sermons and politics mudslinging.

It's almost as if a site centered on technology is really, really good at its core competency.

Hey, you can always go to Slashdot if you want conservative rants.

 

[pokrface]

The really terrible thing about the IBM PC was the abysmal 808x architecture with it's "Calculator Chip" heritage still intact.

The segmented memory architecture, the IRQ-hell, and masked interrupts, it was crap from the start and hamstrung the "PC" growth terribly.

I miss those days sometimes—even though I recognize that my memories of them are very rose-tinted by the redshift of the past.

Still, I learned so much as a kid trying to make the computer do what I wanted—what IRQs are and why they work the way they do; how to edit config.sys and autoexec.bat, and from that, my first intro into how scripting and programming works; how and why a floppy and a hard drive work. More importantly, having to get elbows-deep into the computer to do anything made me comfortable with computers—and that put me on a life path that I suspect will be deeply intertwined with technology until the day I die.

 

[ubercurmudgeon]

Surely no article on this subject is complete without the classic Bill Gates and IBM PC image...

 

[rkhalloran]

There's this tendency by later writers to sort of enshrine the PC, to make it better than it was, because it won the war. In almost every way, it was a horrible computer. But it had two enormous competitive advantages: it was fully documented and open, and IBM was selling it. Those two things are why it won, and almost everything else about the computer is why the war lasted as long as it did.

It would be really interesting to learn more about those that did not win the war. Do you know of any examples?

Late 70's/early 80's, this would mostly be the S-100 based 8080/Z-80 systems of the day; the big players in that space were MITS with the original S-100 Altair 8800, IMSAI with their 8800 clone, Cromemco with their Z-80 systems, then smaller terminal-styled systems such as the 6502-based Commodore PET & VIC-20, and the Z-80 based Radio Shack TRS-80, Processor Technology Sol-20 and Exidy Sorcerer.

My first home computer build out of college was a Cromemco Z-2 box, which I still remember fondly.

 

[operagost]

In 1998 I acquired a 1987 version of the 5150, it was 20MB drive model and used to be a workhorse at the US energy dept.

It ended up behind what what used to be the Iron curtain courtesy of some US church supporting brethren in new European democracies.

It was my first computer and it essentially allowed me to experience the PC revolution from the very beginning albeit a decade or so later (it was a fast-track after that).

What a fun that machine was Single density 5 inch floppy, black-green screen that waved so much it would induce dizziness after prolonged use, 110/220 volts adapter was housed in a cocoa box, which wasn't washed properly so every time it heated, the room was filled with the smell of hot chocolate

I installed Volkov commander and wrote my thesis on that machine.

The only HW upgrade was a single density 3.5 inch drive which was a pain in the ass to fit because what was a standard size in 1998 was not what IBM used back then.

I missed those days

Your floppies were probably 5 1/4" and 3.5" double density. Early DOS versions supported single density drives and media, but I'm nearly certain that the 5150 came with a DS/DD drive. Single density was a home computer thing. The Apple II and C64 drives had something on the order of 160KB capacity. The cool thing is that you could usually hook up a 3.5" floppy to the original controller as long as you had the right cable and DOS version-- but only for 720K double density, which I imagine is what you ended up with. Also, the fact yours had a hard drive suggests it may have been an XT rather than an original PC.

 

[operagost]

One thing that really gets me about the PC is just how bad it was. It was a terrible, terrible computer. Almost everything about it was kind of crippled, to the point that the 8088, running at 4.77MHz, was functionally just about the same speed as the 1MHz Apple II.

It most certainly was not. The 8088 flat out ran calculations an order of magnitude faster than a 1 MHz 6502 if you took advantage of its 16 bit architecture. I guess that if you tried to port a 6502 program to it directly, manipulating data in bytes and not taking advantage of its 8 registers over the 6502's three registers, it would be a lot slower.

But: nobody really knew that. Nobody understood computers back then, or hardly anyone.
I was a teenager at the time, and while I could use a PC or an Apple perfectly well, and program them in BASIC and hook up peripherals and that sort of thing, I didn't really understand them at a deep level. I was pretty advanced as users went, and if I couldn't tell you why the PC kinda sucked, Lord knows not that many others could.

I understood them, because in the pre-internet days I WENT TO THE LIBRARY. And talked to older folks who worked with computers. You didn't have to know the guts of the computer to use it, but it helped. I guess that is still true today.

 

[Stern]

The 386 system we had in the 90s still responded with a "press play on tape" message if you called the right BIOS interrupt. I wonder when that cruft was finally removed?

For a very opinionated view on the beginning of the PC industry, see the old issues of DTACK Grounded. It offers some in retrospect interesting tidbits on technical factors that steered early PC development, like chip manufacturer delays and RAM pricing.

 

[krebizfan]

One thing that really gets me about the PC is just how bad it was. It was a terrible, terrible computer. Almost everything about it was kind of crippled, to the point that the 8088, running at 4.77MHz, was functionally just about the same speed as the 1MHz Apple II.

But: nobody really knew that. Nobody understood computers back then, or hardly anyone.
I was a teenager at the time, and while I could use a PC or an Apple perfectly well, and program them in BASIC and hook up peripherals and that sort of thing, I didn't really understand them at a deep level. I was pretty advanced as users went, and if I couldn't tell you why the PC kinda sucked, Lord knows not that many others could.

It was an abysmal base to build on, and yet, with all the money flowing into that architecture, they were able to bend it, and bend it, and bend it, until it actually got pretty darn good. It's kind of like a stunted, twisted little seedling that was gradually coaxed into a mighty tree, while much better, stronger seedlings died out due to neglect.

There's this tendency by later writers to sort of enshrine the PC, to make it better than it was, because it won the war. In almost every way, it was a horrible computer. But it had two enormous competitive advantages: it was fully documented and open, and IBM was selling it. Those two things are why it won, and almost everything else about the computer is why the war lasted as long as it did.

At roughly the same time as the IBM PC, Apple was pushing the Apple III with the famous tech support suggestion of dropping the computer to correct errors. Instead, business dumped Apple. Over on the S100 side, it was tricky getting groups of cards working together and any change in hardware tended to require recompiling the OS. The IBM PC was a major jump in reliability compared to the average competitor at the time.

 

[JustMe74]

When I was a little kid in elementary school, I went to ComputerLand with my parents and we bought an original, 4.77mhz IBM PC with a green screen and two floppy drives. That thing was just fucking magical, and it changed my life.

 

[Ezzy Black]

This is all very interesting, but it misses the parallel developments going on in the UK and Europe at that time. The history of computing of personal computing is not just USA based.

Paul

This is not a history of personal computing. It's a history of the IBM PC.

 

[rmaine]

Comparing speeds based on clock frequency was a classic marketing ploy of the day, but had little grounding in reality. The article repeats that marketing line in claiming that the 8088 was much faster than the 6502 because the 8088 could run at 5 MHz compared to the 1 MHz of the 6502. This neglects the fact that the 6502 needed far fewer cycles to do comparable operations.

I had, well, quite a bunch of various machines in the day: an Apple 2+, an Apple 2e (which I still have in operating condition), a Heath Z100 (CP/M), several IBM PC clones of various vintages, and a wire-wrapped machine that I threw together myself just for the fun of it (mostly a CPU card plus a video card, with no peripherals other than the keyboard and monitor).