Ars Technica posted my plea for Amiga game developers this morning. By the time I got into work, this was the state of my inbox:

This is so awesome. My inbox now runneth over with people wanting to talk to me about what it was like in the world of Amiga game development. I can’t believe I didn’t think of this months ago!

Well, now at least I know what I’m doing for my Christmas break! I am so pumped for this!

In just a few short years, the iPhone has changed the mobile computing landscape. This cannot be denied. All other phone manufacturers are frantically trying to come up with the “iPhone killer”, usually by copying Apple’s basic design, and everyone is talking about how the iPhone apps are creating a new platform.

Some people take this idea and go overboard, claiming that smartphones are not just a new platform, but are going to be the dominant platform. They usually make a comparison to how mainframes and minicomputers were once the dominant force in computing, before personal computers overwhelmed them. The argument seems to be that since personal computers took over and they were smaller than mainframes, then a product that is smaller still—like mobile phones!—is destined to take over the PC’s place in the market.

It’s always tempting to make comparisons to earlier groundbreaking technology shifts when trying to predict new ones. However, it’s quite easy to miss the real reasons why these shifts happened. Let’s go back and look at why mainframes and minicomputers died in the first place.

There's a great line in the book DEC is Dead: Long Live DEC where founder Ken Olsen is being shown the samples of the new Alpha chip, and told that it is more powerful than the mainframe his company has just spent hundreds of millions of dollars designing.

He stares at it for a while, and shakes his head. "How can that be possible..." he says, completely unable to grok the impact of the moment.

When a tiny chip that can be mass-produced and sold for a hundred dollars has more power than a state-of-the-art mainframe that is sold for millions of dollars, that's a pretty powerful moment. There is no possible way that the story of the mainframe can end happily at that point. It was obvious to many PC people very early on, even in the 1980s, when personal computers had about the same computing power as an average steam engine.

Yet there is no similar "oh shit" moment when we move from personal computers to mobile phones. They both use exactly the same technology. Only in the mobile phone's case, because it has to be ultra-portable in order to be, you know, a phone, all these sacrifices have to be made in terms of power, speed, and storage. This is okay, because, you know, it's a phone. But there is no magical improvement in performance per unit volume, nor is there a magical improvement in price. The phone, if anything, costs slightly more than a laptop that is faster and has vastly greater storage and memory, not to mention a bigger screen and keyboard.

So the two situations (PCs versus mainframes and phones versus PCs) aren't really the same, as much as we might imagine them to be. Mainframes and PCs used completely different technology, with orders of magnitude differences in price/performance and size between them. PCs and phones use the same technology, so no such leap can ever be possible.

Note: This is my original StudlyOS article that I posted to Usenet some time around 1992 -- you can tell by the references to the beta of Windows NT 3.1 -- and I've reproduced it here unedited and unchanged, because really, StudlyOS 1.0 was absolute perfection and can never be improved upon. The screenshot was made some years later, but remember that StudlyOS looks exactly the way you want it to look, every time you boot it up, so your screens may differ.

StudlyOS(tm) Version 1.0

The Only Operating System You Will Ever Need!

Preface

Now that HappyNet is up and running, and Leader Kibo is ably directing the entire world with his custom Mondo Zeugma 6866688786/XA/sxe/IV (see HappyNet Manifesto), the fastest and best computer ever built in the history of time and space (second best was Deep Thought), we at Studly Research, Inc. have come up with an operating system that is simultaneously capable of keeping up with Kibo's needs and sufficing for general use by all the rest of the common and sometimes ignorant citizens of this planet, and of any other planets we can think of.

No doubt you have been endlessly entertained by the furious religious operating system wars now taking place on the PC hardware arena. Should you be content with DOS and Windows? No! Should you switch to OS/2? No! Should you try your luck running a buggy Windows NT beta? Never! How about NeXTstep/486, or the upcoming Pink, or maybe Apple's System 7/486, or Linux, or Cray XMP-OS/486?

None of these!!! We at Studly Research, Inc. have come up with a solution so superior that the entire industry will soon switch over to our operating system and accompanying software. Microsoft will fold and Bill Gates will get a job working at a 7-11, handing out coupons. Apple will also collapse and John Sculley will be found lying unconscious in a pool with a can of Pepsi and a hypodermic needle lying nearby. IBM will survive, but will be forced to lay off another 400,000,000 employees, and eventually end up as a subsidiary of the Moscow McDonalds. The only surviving companies will be the cheap clone manufacturers, producing faster and cheaper machines with the label "Studly-Compatible" and "SPC" proudly displayed on the front panel.

What Is Studly-OS?

Studly-OS is the result of over three decades of intense operating systems reasearch at Studly Labs, also known as the Studly Laboratories For User Triumph, or SLUT. Extensive research with actual humans at SLUT, instead of the trained chimps used in most useability labs such as Xerox PARC*, Microsoft Barf** and Borland Snooze***, has determined that people are less interested in operating systems that offer a wide selection of native programs, or have a pretty interface, or simply go 'bing', than they are in the concept of an operating system that will quite simply solve ALL of their problems for them.

Studly-OS is that operating system.

Not only will Studly-OS make any clone computer, from a ten year-old XT to a 486/330DX10, capable of doing more than all the former operating systems ever developed, it will also quickly, seamlessly, and invisibly solve all of their personal problems and make them happy, rich, sexually irresistible and permanently wonderful.

But What Is The Cost?

Nothing. We at Studly Research, Inc. have gained from our own inventions to the extent that we are already happy, rich, sexually irresistible and permanently wonderful. We are offering Studly-OS to the public free of charge. Every ftp site will soon be carrying, and running under, Studly-OS, and free diskette, CD-ROM and Braille copies will be available at bookstores, K-Marts and oil refineries worldwide.

How Is This Technological Miracle Accomplished?

Most of the developments at Studly Research, Inc. are of course patent-protected and highly secret, although we do not balk at hyping tantalizing tidbits of Studly technology, simply to gain free press coverage.

Studly-OS is built around a nanokernel, the advanced descendant of microkernel operating systems available today. Our crack team of coders, hackers and pizza enthusiasts took an early beta of Microsoft's Windows NT, completely disassembled and analyzed the code, and then built Studly-OS by doing everything completely differently. We'd like to thank Microsoft for $69 well spent as a helpful exercise on how not to design operating systems.

Whereas NT's microkernel is fat enough to tip over a bus, Studly-OS's nanokernel fits in under 1k. Instead of a multiple message queue, Studly-OS uses a method where messages are intercepted before the application in question has even sent them out. We redesigned the Windows interface to appeal less to schizophrenics and came up with a fully object-oriented system where the objects not only were oriented with respect to each other, but oriented themselves to best suit the individual computer user, including sexual and political preferences. No longer is the system merely user-friendly, it is positively user-worshiping!

The Operating System Respects The User

People work in different ways, and Studly-OS automatically adjusts to this, painlessly, seamlessly, smoothly and invisibly. We realize that most computer users want their operating systems to pretty much stay out of the way and run any application they choose to throw at it. Studly-OS handles all file manipulation. You will never touch another configuration file or menu again!

For example, when the user sticks the first disk of an application in a drive, or even in between the little air vents in the front of the case, Studly- OS automatically determines what the program is, where it should be installed and how it should be set up, and then proceeds to build the rest of the program based on the contents of the first disk, taking out features which you will never use and adding those which the software manufacturer blindly left out. It then opens the icon editor and lets you create the ultimate icon for that application, filling in any tedious or difficult painting bits along the way. It then adds sound and animation to the icon, and while you watch, loads the application in the background and does your work for you while you play a quick game of Wing Commander III.

Studly-OS Is Compatible

Not sure if your application will run under Studly-OS?

Studly-OS runs ALL software programs written for DOS (including those using VCPI, DPMI and Shmoodoo memory management, by rewriting and optimizing the code before installing) Windows (including Win32, Win32s, Win32c, Win32nb, Win32ack and Win32thbbth!), OS/2, GEM, AmigaDOS (including games which refuse to run on any Amiga past a 500), NeXTstep, Unix (Studly-OS maintains a daily-updated database on every Unix variation in existence, and automatically recompiles any Unix program in the background to work on your system) TRS-DOS, Apple ][ DOS and ProDOS, Macintosh Systems 1 through 9, Timex-Sinclair ZX81 programs, Atari 2600, Nintendo and Sega game cartridges, Heathkit HDOS, CP/M (including utilities that used weird Z80 opcodes), Epson's Q-DOS, Cromenco DOS, RISC OS, Commodore C=64, 128, VIC-20 and Plus/4 programs, and Coleco ADAM software. If Studly-OS encounters an application written for a platform it does not support, it rewrites it to conform to established standards. If Studly-OS senses that a particular application is not running at sufficient speed, it rewrites the code until it exceeds the performance on the best hardware available. For example, one user managed to get Studly-OS to run Strike Commander on his XT with 8-bit VGA, and noted that the game response was "very smooth, at least 60 frames per second with no flicker or pauses that I could find."

Studly-OS Offers Superior Compression

Although the operating system itself, due to incredibly tight and sexy coding, fits into less than 32k of RAM and 500k of disk space, we realized that most user's applications are reaching such gargantuan sizes that we decided to include an advanced disk-compression package with the product.

16:1 lossless compression!! Yes, the reason this mythical product was never released to the marketplace was because we bought it out. Lock, stock, and barrel. You can compress a compressed file as many times as you like until all programs are down to the theoretical minimum of 1k! Yet still not lose any data. Of course, with all your programs at 1k, uncompression may take a little longer. However, we feel the extra disk space is truly worth it. Most graphics files, including .JPGs and .GIFs, can be safely compressed down to less than 32 bytes, especially the nudes, which all look pretty much the same anyway. Pictures of Madonna can be packed as small as 1 byte.

Studly-OS Is Here, Now!

No Microsoft FUD. No promises of shipments "when it's ready". Studly-OS is ready and available for you to install NOW! What are you waiting for?

Studly-OS Is Bug-Free!

Others promise, but we deliver. We don't have to name our product 3.1 just to fool people into thinking that it is a tested system. Studly-OS is, and will always be, version 1.0! There will never be a need for an upgrade!

And no, if you discover a bug, we don't send in the SWAT team to prove that you are an inconsiderate moron with the technical knowledge of a squashed gnat that can't even find his way out of the refrigerator. In fact, if you do find a bug, we are prepared to give you a $1 million prize, and an all-expenses paid tour to the fabulous Studly RESEARCH LABS in beautiful Barbados, where you will get to meet the Studly-OS design team and go for dinner and drinks! Then we will send out patches to everyone in the world free of charge.

Compare Studly-OS With Those "Other" Systems!

Feature Studly-OS! DOS/Windows OS/2 NT Unix
Nanokernel technology Yes! No No Hah! No
16:1 Lossless compress Yes! No No No No
Free Origin game (rewritten to actually handle memory the way sane people would) Yes! No No Heh!! Hah!
Automatically finishes important work for you Yes! No No No Work???
Free hyper-animator to make Babylon 5 look like Popeye cartoons Yes! No No No No
Ten million included .GIFs, .WAVs and .WOWs Yes! No No No WOWs?
Automatically optimizes application code Yes! No No Optimize? No!
Makes you feel sexy Yes! HAH!!! No No Sex???
Tastes good with ice-cream and chips Yes! No No No Food???
Makes Bill Gates seem like a weenie Yes! Yes Yes Yes Yes
Balances your checkbook Yes! No No No Money??
Washes your car Yes! No No No Automobiles?
Improves self-esteem Yes! No No Worsens Suicide
Makes you rich Yes! No No No sorry
Supports SMP Yes! No Soon Yes Sometimes
Requires SMP No No No Yes No
Message-passing Yes! No Yes Yes Yes
Message-losing No No No Yes core dumped
Message-SENSING Yes! No No No guru
Zen Yes! No No No flower
Software support Great! Lame OK Where? Software??
Technical assistance None needed! No No No ARMM
Documentation quality Great! Docs? OK Huh? grep
General studliness Super! Ouch! So-so ICK! alt.angst
RAM requirements 32k 640k 8 meg 16 meg How much???
Disk space required 500k 1 meg 30 meg 80 meg rm *
OSes supported 24 1 3 3 Support?
Price Free! $60 $99 $495 $0 [ n [ $oo

Here Are Some Real-Life Quotes Of People Who Have Used Studly-OS!

"I love it! It makes me want to eat!" - Rush Limbaugh
"Since it doesn't have the name Windows on it, it is an irrelevant platform." - Bill Gates
"We will develop applications for Studly-OS if they sell two million copies in the first year, but they won't sell more than 25, no matter how many they actually sell." - Bill Gates
"Would you like fries with that?" - Bill Gates
"It's a beautiful day in the Studlyhood" - Fred Rogers
"Pull the other one!" - Patsy
"This is the most impressive operating system I have ever seen in my entire life. It makes everything else seem like damp kleenex. However, it will surely fail and become a dead operating system and fail and fail fail fail it must fail!!!! Because it doesn't have the power of Microsoft's marketing behind it." - Paul Somerson
"I'm sure I've used Studly-OS before" - Shirley MacLaine
"I will be introducing the new Studly-OS-compliant retroactive moderation specifications directly to the Net" - Dick Depew
"I'm sorry, but I happen to own the copyrights to the letters O and S. Please send me all your money right away." - James "Kibo" Parry
"Ack! Pft!" - Bill The Ceo
"StudlyOS sucks!!!1111 Y00 think itz c00l but your rong!!!!!11111 I Cant run it on my Am1ga so what yoos is it????/ My Am1ga beats yor peecee anyday!!!!!! !!!11111111 Peecee even with StudlyOS cant beet Amiga because Amiga rules!!!! Amiga iz better because it is Amiga!!!1111 Nothing else is Amiga!!!11111" - B1FF

Notes

* Palo Alto Research Center
** Boring Applications Rarely Fascinate
*** Stupid Nonsensical Object-Oriented Ziff-Davis Enema
HappyNet, Mondo Zeugma, and O and S are trademarks of Kibo.
Windows is NOT a trademark of Microsoft.
B1FF is a trademark of himself.

Google gave the world a first look at their upcoming ChromeOS yesterday, and the technology press is buzzing with excitement. I can’t quite get on the ChromeOS cheerleading bandwagon though, which is weird, because I love Chrome the browser and use it constantly at work. Maybe that’s why I can’t get on the ChromeOS bandwagon.

ChromeOS is an open-sourced, Linux-based operating system from Google that is intended for netbooks and other lightweight browsing devices. Instead of a standard Gnome or KDE desktop, however, ChromeOS boots into the Chrome browser, with a couple of extra tabs on the left that work as a sort of Start Menu for launching applications. The big difference is that all the applications are web-based: there will be no “desktop” applications as such.

Anyone else remember when the iPhone launched and Steve Jobs said “hey, you don’t need native applications, just use web-based ones” and everyone gasped and Apple had to immediately relent and release an SDK for developing real native apps? Now Apple proudly touts the number of apps available for their phone and these apps are a major selling point over all other phones.

Google, apparently, gets no such criticism. “Just use web-based apps!” they say, and the crowd cheers. “Google is leading us into a web-based revolution!” come the cries of the pundits. “The road ahead will be paved with web-based gold! No more operating system maintenance! No more updates! No more virus scanners or tedious file manipulation!”

Okay, but here’s the thing. You could do all of this right now, with the computer you already have. Just try it. Uninstall all your applications except for Chrome, make sure auto-updates are on (Windows and Chrome will do this by default) and put Chrome in your Startup folder. Now, when you boot your computer, you’ll be in Chrome, running web-based apps all the time, and you will be living in the future with flying cars and vacations on the moon.

Have you done it yet?

No?

Why not?

The answer is that people don’t like taking something, no matter how complicated and difficult to use, and reducing its functionality. This tale is as old as the PC. Joel Spolsky wrote about it way back in 2001:

A lot of software developers are seduced by the old "80/20" rule. It seems to make a lot of sense: 80% of the people use 20% of the features. So you convince yourself that you only need to implement 20% of the features, and you can still sell 80% as many copies. 

Unfortunately, it's never the same 20%. Everybody uses a different set of features. In the last 10 years I have probably heard of dozens of companies who, determined not to learn from each other, tried to release "lite" word processors that only implement 20% of the features. This story is as old as the PC. Most of the time, what happens is that they give their program to a journalist to review, and the journalist reviews it by writing their review using the new word processor, and then the journalist tries to find the "word count" feature which they need because most journalists have precise word count requirements, and it's not there, because it's in the "80% that nobody uses," and the journalist ends up writing a story that attempts to claim simultaneously that lite programs are good, bloat is bad, and I can't use this damn thing 'cause it won't count my words. If I had a dollar for every time this has happened I would be very happy.

There’s no reason to expect that things have changed in 2009, and suddenly people will prefer having less functionality on their notebooks and laptops. This has already happened with Linux-based notebooks: when they first came out, 100 percent of notebooks ran some form of Linux with a simplified shell, because “people just use notebooks for web browsing, and this way you’ll save money!” What happened? There was such a demand for putting XP on these notebooks that the companies relented and started selling both Linux and XP versions, with the latter outselling the former. The market share of Linux on netbooks went from 100 percent to anywhere between 4 percent (according to Microsoft) and 32 percent (according to ABI Research). As time goes on and netbooks naturally get more powerful, there’s no reason to expect the demand for Linux on netbooks to go anywhere but down. Going from 100 percent to 32 percent in a couple of years is not a great sign.

But at least with Linux on a netbook you can still run Linux applications! You can run OpenOffice.org, which is a passable simulation of Microsoft Office, and you can run Linux games and graphics editors and sound editors and all manner of other things. Sure, most people just use their netbooks for surfing the net, but people still like to have the option of doing other things. ChromeOS takes away this option, and people don’t like it when their options are removed. People will get all excited about a ChromeOS netbook, right up until the point where they are told they can’t run any non-web application. No Office, not even OpenOffice. Just Google Docs and Google Spreadsheet, which (quite frankly) suck.

This is why ChromeOS will fail.

Personal Computer Market Share: 1975-2005

stats and graphs compiled by Jeremy Reimer

UPDATE!!

I have written a full article based on my Total Share numbers and it has been published at Ars Technica!

Please click here to read the article!

The first personal computer, the MITS Altair 8080, was released in 1975 and changed the world forever. A handful of geeks (Bill Gates included) saw this humble $395 kit as the beginning of something big... but nobody knew how big!

The Altair sold a few thousand units in 1975. Today, more than 130 million personal computers are sold each year! How did we get here, and which computer platforms were around for the journey? A lot of people who have come into personal computing recently do not know that there were once many different platforms-- a glance at a 1980 issue of Popular Computing revealed over 100 different manufacturers of incompatible brands!

The following graphs reveal some of the story, and show the incredible growth of the industry. They should also spark some memories of platforms gone by.

All figures in 1,000's of units

Notes on sources:

http://www.etforecasts.com/products/ES_cinuse.htm :

Table 1.1 US and Worldwide Computers-in-Use Growth

1980 1985 1990 1995 1999 2000 2002 2005

Worldwide Computers-in-Use (#M) 5.0 35.2 105.6 237.6 459.5 543.2 721.6 1,044

An estimated 835 million PCs were sold worldwide between 1981 and 2000. (Source: International Data Corp.)

IDC forecasts that in 2001, 138 million PCs will be sold worldwide, which is more than:

In 1980, Apple Computer Inc.?s sales totaled about $200 million, compared with Tandy Corp.?s approximately $175 million and Commodore Business Machines Inc.?s approximately $40 million. (Source: Robert Metz, I.B.M. Threat to Apple, The New York Times, Sept. 2, 1981)
Worldwide, some 500,000 computers priced less than $5,000 were sold in 1980 at a total value of $730 million, according to Dataquest Inc. In 1985, the firm estimated that the total value would grow at least 40 percent annually, to 3.7 million units, valued at $3.9 billion.
By August 1981, Apple had sold more than 250,000 personal computers (since 1977) and was selling 15,000 to 20,000 a month. In 1981, it had more than $300 million in sales. Sales for the quarter that ended June 30, 1981, rose to $90.7 million, a 179 percent increase from the second quarter of 1980, while net income increased 334 percent, to $11.9 million. (Source: Andrew Pollack, Next, a Computer on Every Desk,The New York Times, Aug. 23, 1981)

By August 1981, Apple had sold more than 250,000 personal computers (since 1977) and was selling 15,000 to 20,000 a month. In 1981, it had more than $300 million in sales. Sales for the quarter that ended June 30, 1981, rose to $90.7 million, a 179 percent increase from the second quarter of 1980, while net income increased 334 percent, to $11.9 million. (Source: Andrew Pollack, Next, a Computer on Every Desk, The New York Times, Aug. 23, 1981)

In 1980, IDC estimated that 327,000 desktop computers, ranging in price from several hundred dollars to $20,000, were sold in the United States. It projected that this total would increase to
1.3 million by 1985.

Dataquest predicted that by 1985, 3.7 million personal computers would be sold annually, worth
$3.9 billion.

http://www.pcworld.com/news/article/0,aid,102386,00.asp

In 2000, the industry's best year to date, around 132 million PCs were shipped worldwide, and the industry is expecting a similar figure in 2002 as the market rebounds from a weak 2001.

The $1,295 Apple II (1977) and the $2,880 IBM PC Model 5150 (1981) started the rapid growth of PC sales, but it took the market almost 15 years to cross the 50 million units per year mark (in 1995) and just another four to cross the 100 million per year mark.

http://www.time.com/time/special/moy/1982.html

The sales figures are awesome and will become more so. In 1980 some two dozen firms sold 724,000 personal computers for $1.8 billion. The following year 20 more companies joined the stampede, including giant IBM, and sales doubled to 1.4 million units at just under $3 billion. When the final figures are in for 1982, according to Dataquest, a California research firm, more than 100 companies will probably have sold 2.8 million units for $4.9 billion.

At the moment, Dataquest estimates that Texas Instruments leads the low-price parade with a 35% share of the market in computers selling for less than $1,000. Next come Timex (26%), Commodore (15%) and Atari (13%). In the race among machines priced between $1,000 and $5,000, Apple still commands 26% followed by IBM (17% and Tandy/Radio Shack (10%). But IBM, which has dominated the mainframe computer market for decades, is coming on very strong. Apple, fighting back, will unveil its new Lisa model in January, putting great emphasis on user friendliness. The user will be able to carry out many functions simply by pointing to a picture of what he wants done rather than typing instructions. IBM is also reported to be planning to introduce new machines in 1983, as are Osborne and others.

http://timeline.99er.net/article-time-hottest_selling_hardware.html

Timex Sinclair 1000 ($99). This tiny toy is good for dipping one�s toes into the micro revolution and not much more. It will play video games with boxy, black-and-white graphics and speaks only one language: BASIC, A buttonless �membrane� keyboard is well designed for learning the fundamentals of computer programming, but for written work it is a step down from the old typewriter. With 600,000 sold in 1982 alone, there is sure to be more software on the shelves soon. A more powerful model that speaks child-oriented Logo is expected out this spring.

Commodore VIC�20 ($299). Skillful packaging and aggressive marketing helped make this machine the surprise bestseller of 1982: between 600,00 and 1 million sold. The VIC has the only cut-rate keyboard suitable to touch typing, and when hooked up to a $110 telephone model, it becomes an inexpensive electronic mail terminal. There have been software shortages, but more programs are being written to meet the new demand. The Commodore 64, a $595 version that packs the memory capacity of some machines three times its price, arrived late in 1982 and could be a big seller in 1983.

Atari 400 and 800 ($299 and $899). With 256 colorss, four separate sound generators and built-in �missile qraphics,� the Ataris are the machines of choice for game players and games writers. The 800 has a keyboard suitable to touch typing, but writers would do well to look elsewhere for a first-rate word processor. Nearly 200,000 Atari 8OOs were shipped in 1982 and some 400,000 model 400s.

Texas Instruments 99/4A ($450). The sleeper of the year, In 1978, when it retailed for $1,100, it was a market failure of historic proportions. The company upgraded the keyboard, hired Bill Cosby to do its commercials and drastically cut prices. Sales exploded, T.I. shipped 530,000 in 1982, and at year�s end was selling nearly 150,000 a month. Software has been slow to come, but now there is a generous supply of high-quality educational programs, and, for $380 extra, owners can get a spritely version of LOGO. More will follow. With sales spurting, program writers say they are giving the machine a second look.

Epson HX�20 ($795). Although this book-size portable computer arrives late in the year, it is probably the hottest new machine in its class, shipping 7,500 copies in its first month on the market. It packs into one handy 4�lb. package a full--size keyboard, a screen that displays four lines of text, a cash�register�type printer, a microcassette tape drive and more built�in memory than any comparably priced machine. Its Japanese manufacturers say their intention was to "stand America on its ear." U.S. experts say they may have done just that.

TRS-80 Model III ($999). Back in 1978 Radio Shack, Commodore and Apple had the field to themselves, and Tandy-Radio Shack, with its nationwide chain of retail outlets, had more of the field than anyone else. A sturdy word� and number�crunching machine, the "Trash-80", as it is affectionately known, seemed to have a lock on the corner computer market. By year�s end there were 300,000 Model IIIs in place. But the company has been overtaken by less stodgy competitors, and last year Tandy�s share of the mid-range market fell from 13% to 10%.

$1,000 to $2,000

Apple II Plus ($1,330). The hardy bestseller of the late �70s is also the hardy bestseller of the early �80s: 700,000 have been sold: 270,000 in 1982 alone. With so many cheaper and more sophisticated machines available, why does the Apple II still hold the biggest slice of the $1,000-to-$2,000 pie? Software. More programs are available for this six-year-old machine than for any other single computer-, some 16,000 in all. Also more user groups, more space in the computer magazines, more plug-in expansion units, more peripheral devices. It used to be that when something was done a microcomputer, it was done first on an Apple II. Today IBM, Commodore and Atari are changing that.

iBM Personal Computer ($1,565). The top executive�s Apple, this is the machine that put the stamp of corporate legitimacy on the computer revolution, and it quickly set industry standards in everything from operating systems to its no-nonsense instruction manuals. Although other machines have sold in greater quantities, the IBM is the computer of the year. Introduced in August 1981 nearly 200,000 were shipped in the past twelve months, winning it 17% of the market for mid-range machines. Already 1,000 programs are available for the PC. including games. Though IBM discourages using its machine as a toy (it charges $30 extra for color graphics), software programmers are busily translating all manner of playtime activities to the run on the IBM PC.

Osborne 1 ($1,795). The first of the sewing machine-size portable computers, the Osborne 1 squeezes into a 24-lb. package a video monitor, a pair of disc drives and the two programs indispensable to businessmen: financial forecasting and word processing. Despite its eyestraining 5-in, screen, 55,000 Osborne 1 models were sold in 1982, bringing the total number shipped to nearly 100,000.

Over $2,000

Only growing small businesses and big corporate clients are likely to go after these computers, known in the trade as �professional work stations� and designed to hang at the branches of a network of similar machines. Price tags range as high as $10,000: Altos, Corvus, Control Data, Cromemco, Digital Equipment, Fortune, Hewlett- Packard, Nippon Electric, North Star, Olivetti, TeleVideo, Toshiba, Vector, Victor, Xerox and Zenith are among the biggest names in this upscale but increasingly crowded field. Even proletarian Apple is joining the crowd with its long-awaited Apple IV (code-named Lisa), due to be unveiled in mid-January. Lisa�s probable price range: somewhere between $7,000 and $lO,000. The Apple V (code�named Mackintosh), on the other hand, due out in mid-1983 and priced around $2,000, could be a true mass�market machine.

http://groups.google.com/groups?q=commodore+64+best+selling+year&hl=en&l...

In one year, Commodore has sold more than 100,000 Amigas world wide, McIntyre
says. He notes that since it was only recently launched in Europe, most
of that figure is North American. In Canada more than 5000 units have been
sold, he said.

Far outselling the Amiga, in terms of numbers, are the Commodore 128 and
Commodore 64, now rereleased as the Commodore 64C. McIntyre boasts that
the C-128 is "still the fourth best selling microcomputer in the world."
North American sales in the three-month period ending in September were
about 50,000

http://groups.google.com/groups?q=neXT+computer+sales+figures+1987&hl=en...

The market share figures above do not agree with reality.
In 1992, the Apple world-wide share was 11.8% according to Dataquest.
Their unit laptop share last quarter was 22%. The IBM ThinkPad was 18%
and closing. (CRN audits & Surveys)
Sabu

http://groups.google.com/groups?q=commodore+annual+report+1986&start=10&...

C64 Sales:

1982: 150K - 300K

1983: 2M

1984: 2M-3M

1985: 2M -3M

1986: 2M -3M

1987: 1M - 2M

1988: 1M - 1.5M

1989: 1M - 1.5M

The following figures are from Commodore annual reports:

Fiscal 1990: 700K - 800K (decline begins),

fiscal 1991: 800K ( non official 1M)

fiscal 1992: 650K

fiscal 1993: 150K - 200K

1993 Annual Report: 17M total C64, 4.5M C128

The popular myth that we are all told repeatedly is that Apple designs their products from a human-centric behavior, whereas dumb old Microsoft designs from a cold, machine-centric standpoint. Apple's operating system is, therefore, much easier and friendlier to use, whereas Microsoft's OS is user-hostile.