Talk:IBM PC–compatible/Archive 1

"IBM PC Compatible" right name?
Aren't PCs that are compatible with IBM PCs called "IBM compatible PCs" more often than "IBM PC compatibles"? (Links are to Google tests to support my point...) -- Oliver P. 17:46 5 Jun 2003 (UTC)


 * That's a good question. I know that I've heard the term "IBM compatible PC" many times but I've never heard the term "IBM PC compatible" until now.  I will move the page in a few days if no-one objects.


 * Acegikmo1 08:35, 19 Aug 2004 (UTC)


 * You know, I think Oliver P. is right. Go for it. However... is it
 * IBM compatible PC (no hyphen)
 * or
 * IBM-compatible PC (with a hyphen?)
 * Doesn't matter to me as long as both are created and one redirects to the other.
 * Of course, other terms in common use. The "IBM" dropped out fairly soon. I remember being extremely annoyed when the phrase "PC" came to mean the specific IBM architecture. Other terms at various times were "PC-compatible" and "PC clone." Jerry Pournelle frequently used the neologism "PCompatible" (or was it "Pcompatible" in his column, but it never caught on&mdash;perhaps because nobody had a clue how to pronounced it. [[User:Dpbsmith|Dpbsmith (talk)]] 10:36, 19 Aug 2004 (UTC)
 * P.S. The misspelling "compatable" is extremely common and perhaps justifies creating another bunch of redirects if we don't have them already. [[User:Dpbsmith|Dpbsmith (talk)]] 10:37, 19 Aug 2004 (UTC)
 * PC-compatible might be a regional thing. I remember "IBM compatible" being equally common. --24.114.252.183 00:29, 3 Jan 2005 (UTC)
 * "IBM PC Compatible" isn't technically incorrect, and was the predominant slang in North America in the 1980s. It's not incorrect because the clones were compatible with the original IBM PC, making them "IBM PC Compatible".  So this page shouldn't be renamed.  In fact, it's important to *keep* it as "IBM PC Compatible", because "IBM Compatible PC" isn't specific enough. Trixter 18:35, 29 December 2005 (UTC)

Wintel
I think that the statement that WinHEC is giving marching orders to hardware vendors, is a little inaccurate, Microsoft steers the general direction of PCs. But I would hardly call it marching orders. Microsoft hasn't offical supported AMD64 (at least not yet), but AMD was free to make it's 64bit processor. PPGMD


 * I like the latest edit, it shows more properly the control MS has over hardware. They have control, but not total control. PPGMD


 * I was about to say... OK, maybe "marching orders" goes too far, but "guidance" doesn't go nearly far enough. To say that Microsoft only provides "guidance" is literally true, but it's sort of like saying that a traffic cop only provides "guidance" to motorists. How about an opportunity "to lobby for and in some cases dictate the pace and direction of the hardware side of the PC industry," which is language I got from Mark Hartman at ExtremeTech.


 * Thanks for your agreement. [[User:Dpbsmith|Dpbsmith (talk)]] 17:53, 27 Oct 2004 (UTC)

Vandalism
Any of you noticed that in the following sentence


 * Simultaneously, many manufacturers such as Xerox, Digital, Sanyo, and Wang introduced PCs that were...

Wtf? how did that make it there without anybody noticing?! I am pretty sure that Wang is not a brand name in any country, and certainly not a manufacturer of PCs 20 years ago. Therefore I'm removing that term. &mdash; LegolasGreenleaf 03:02, Nov 26, 2004 (UTC)


 * See http://www.digidome.nl/wanghall.htm for details of some Wang computers. HungryHorace 13:59, 3 Dec 2004 (UTC)
 * The youngins are taking over. Never heard of Wang computers... wow. If you'll excuse me, I have to yell at some damn kids to get off my lawn.

Apple
The article contains an unattributed claim that Apple had a "50% marketshare" and that IBM's PC was a response to the Apple II. My general feeling has always been that Apple has always had a fairly small market share, though it does have significant influence in the market. I'd like to know where that 50% figure came from, especially with the TRS-80, Commodore, etc. in the same market at the same time.


 * Probably best to remove it until the statistic can be verified.--24.114.252.183 21:30, 17 Jan 2005 (UTC)


 * It sounds about right to me, but I can't back it up with figures. I do remember one big statistical surprise. Circa 1982 or 1983, virtually every software bestsellers list claimed that Lotus 1-2-3 was the bestselling software package, by far. I saw a very credible article that indicated that this was not true at all. It was outsold by AppleWorks in dollars by a significant factor, maybe 2:1, and, of course, since AppleWorks was a cheaper package, the difference in unit sales was even higher. The problem with AppleWorks vs. 1-2-3 was that AppleWorks was sold directly by Apple and therefore was simply not counted at all in the "bestseller" figures, which only tracked sales by, IIRC, Corporate Software and Ingram Micro-D.


 * There has similarly always been a tendency to underestimate Apple's market share, because the estimates are usually made by outfits like Gartner who are only interested in sales to businesses, and do not track consumer purchases.


 * Finally, for what reasons I don't remember, but for about three years in a row, circa the mid-nineties, each year some much-quoted source would announce an Apple estimated market share, which always got a lot of press coverage and got cited a lot by MBA types, and each time about three months later they would announce a correction that amounted to doubling the number.


 * Of course one problem in this is that it's never clear how to define "the market." Apple has always had problems selling into businesses. Even in the Visicalc days they were practically smuggled in, and weren't usually bought on the IT budget.


 * In 1990, when I was embroiled in an internal hassle over market share numbers at a Fortune 500 company, which was debating whether to produce a Macintosh version of a software package, the truth, as nearly as I could determine it at the time, was that there were four computer companies whose market shares were each very close to 20%, so much so that you couldn't honestly say which was largest. IIRC those companies were IBM, Compaq, Apple, and... sorry, I forget the fourth. The reality at that time was that IBM _compatibles_ did, indeed, dominate the marketplace, but _no single company_ was doing much better than Apple. If Apple had about 20% in 1990, and was almost neck-and-neck with IBM, then I can easily believe that Apple might have had 50% circa 1980.


 * Apple, Commodore, and Radio Shack were the biggies before IBM entered the market and there's no way I can believe Apple had LESS than a third of the market. Don't forget, at that time Apple hard darn near 100% of the education market. Dpbsmith (talk) 00:03, 16 September 2005 (UTC)

What about IBM PC compatable servers?
I think it is worth noting that many non-PC computers today are IBM PC compatables. Just about any x86/AMD64 based server for example. WikianJim 12:47, 15 Mar 2005 (UTC)

Doom
Doom isn't actually a true 3D game, is it? Urbane legend 23:51, 9 August 2005 (UTC)


 * Not in a sense it is understood currently, but for practical purposes, yes.

":However, PC was already primary gaming platform when Doom came out. It was the games like Civilization, Wing Commander and Falcon 3 which estabilished dominance of PC as a dominant gaming machine. The biggest factor was that almost all PC's came with a hard drive, which were rare in other machines. The first gaming generation which grew up on C64's etc craved for more challenging and complex games, and these were more and more difficult to make in '80s home computers with mere disk drives. Article needs to be expanded. --Mikoyan21 18:53, 10 November 2005 (UTC)

This is rubbish
Quoting the article... "To make things worse, IBM, Intel and Microsoft introduced several design flaws which created hurdles for developing the PC compatible platform. One example of such a design flaw was the DOS 640k barrier (memory below 640k is known as conventional memory). This was partly to do with the way IBM mapped the memory of the PC (see the article on the Upper Memory Area for more), and partly because the memory-management of DOS (which was the most widely used operating system) had a way of dealing with it that made things worse."

The original IBM PC was based on an intel 80186 chip that had by design a 20 bit address bus, and thus a maximum memory addressing range of 1 megabyte. Reserving a quarter or this for hardware addresses and system roms is hardly a "design flaw". Given the hardware of the day the original design was entirely justifiable. Especially given that the maximum memory expansion available on the original IBM PC was 64mbyte (whoops make that kilobytes). 20/20 hindsite is wonderful. One could just as justifiably castigate the world for not desining DOS to be compatible with 64 bit processors. The above text is therefore POV as well as drivel.

Don't really see how to rescue it, but somthing along the lines of "the unexpected durability of PC based software, in the face of massive improvements in the performance and scale of hardware, lead to a series of unhappy compromises in the archetecure of PC memory managemnt, as software developers attempted to maintain backwoard compatability with existing software, but adapt to the enhanced power offered by later hardware. A durable memeory management scheme was not achived until the wide deployment of 80386 class and better processor hardware which offered a mode providing a flat 32 bit addressing space, and even then the compromises necessary to maintain backward comapatability with DOS remained inelegent. Full resolution was not possible until DOS was replaced by OS/2, NT and Linux, with DOS programs relegated to running under emulation on more sophisticated operating system designs capable of using the full power of the updated hardware." is truer to the real situation Shoka 22:16, 10 August 2005 (UTC)


 * I completely agree with you. I think it was not a design flaw. Maybe it was best to rename this section as "Early Limitations", "Hardware Limitations", or something like. In my opinion the hardware wasn´t "limited" in the sense expressed in the article. When using a 80186 CPU, the 1 MB limitation is applied. When using a 80386, this limitation is not present. It is a CPU limitation, not a hardware design limitation as I understand it. If the 80186 allowed for a 32-bit (or greater) flat memory addressing, and the IBM-PC hardware didn´t provided the possibility of using all this extra memory then I would call it a design flaw. DOS on the other hand should have been more flexible and elegant in dealing with memory and possible future expansions. It was much too hard-coded/tied to the hardware of the original IBM-PC. It should have taken some inspirations from previous OS (unix, vms), all of them very flexible in the memory managment area. I think the DOS and the very basic BIOS (hum? doesn´t BIOS must be BASIC in the first place?, so better call it "the very rudimentary BIOS"...) was the real limitations of the original IBM-PC, an thus it was not a limitation of the IBM-PC, but rather a software limitation, but a limitation from a software that was bundled with every IBM-PC, thus plaguing the machine itself, not the vendor/provider of the OS and BIOS. Loudenvier 13:23, 19 October 2005 (UTC)


 * I agree with the above, so I've greatly clarified the work. Original author was not familiar with exactly what the limitations of the hardware and software were; my revised paragraph should clarify everything. Shoka, thanks for pointing this out.  Trixter 18:31, 29 December 2005 (UTC)


 * Since this has been incorporated into the article, I've re checked. Though I have always believed (from information in an Intel seminar announcing the 80186 and 80286 in around 1982) that the IBM PC was based on the Intel 80188 (not 80186, my bad), all the indications I can find, including a picture of the original prototype motherboard:

http://www.answers.com/topic/motherboard
 * indicate that the chip used was a vanilla 8088, with separate interupt controller and memory latch. So I'm removing the reference to 80186 in the original article. The address limitation and thus the design limitation are still entirely correct. Moral NEVER believe an Intel salesman.... --Shoka 22:51, 29 December 2005 (UTC)

Removed Albert Clark paragraph
While I appreciate Mr. Clark's influence on setting US Federal Government IT standards, I am unconvinced that the paragraph (which I commented out, not deleted, for now) belongs in a Wikipedia article on the history of PC-compatible computers. (And certainly not in such a rambling form!) Yes, the US government is a huge influence, but not the only one (otherwise Americans would all be using 24-hour time and developing software in Ada). In actuality, the financial sector has historically been the most-important player in IT spending. For example, financial-services companies accounted for 26% of IBM's 3Q sales, small-to-medium businesses at 19%, and all public spending worldwide (of which a large portion, but not all, would be US Federal) in third place at 15%. Yeechang Lee 22:03, 24 October 2005 (UTC)

PC's were not game machines until doom was released?
''One thing that PCs did have in their favour was raw processing power. This made them suitable for 3D games. The PC made a breakthrough as a games machine when Doom was released in 1993 thanks to its outstanding graphics and gameplay. Because networking hardware was widespread on PCs, Doom also offered multiplayer support across a network. Few games offered that at the time. Doom finally established the PC as a games-machine.'' - quote from the article.

I think that entire paragraph is wrong. Though Doom is a great game, the PC had been a huge force in the game industry for years at the time of Doom's release. That section should be removed.--The Goat 20:01, 23 January 2006 (UTC)


 * I agree. The remark reflects a third-millennium understanding of the word "game." Microsoft Flight Simulator was among the very, very first pieces of software ever released for a PC. It's true that IBM was confused enough to make the standard display option the non-graphics-capable Monochrome Display Adapter&mdash;thinking most users would not want or need graphics&mdash;while making the fairly crude Color Graphics Adapter an option. However, I don't know that I ever in my life saw a PC equipped with the MDA... because games were driving the PC market from the very, very beginning. The presence of a joystick adapter from day one is also indicative. Dpbsmith (talk) 20:05, 23 January 2006 (UTC)


 * P.S. http://fshistory.simflight.com/fsh/fs2.htm says:
 * Thus in November 1982, Microsoft Flight Simulator 1.00 hit the shelves as one of the pioneering entertainment titles. Soon it became one of the most sought after software for the PC and after nearly 20 years it still is. It even made the Guinness Book of Records. Over 20 million copies of FS 2000 alone are reported to be sold. It is also safe to say that the application’s demanding hardware needs actually influenced the landscape of computer development and sales. A kind of race started between FS and the PC: every new version of FS demanding more than the current PC-model could deliver and every new PC-model and graphics card offering room for enhancements of FS.
 * IMHO this confirms that games drove PC sales and development right from the very beginning. Dpbsmith (talk) 20:10, 23 January 2006 (UTC)


 * Having had more than 2 decades of experience playing and creating PC games, I agree completely -- although I noticed nobody actually corrected the article. I will do so this afternoon. --Trixter 16:14, 6 April 2006 (UTC)

Pictures
These pictures are AWFUL. First of all, what the hell is a coke can doing next to a laptop... second of all, "a modern PC" is a bit misleading.. first of all, I've never heard of that PC-maker, and if I've never even *heard* of it, it's either incredibly obscure, or it simply isn't made in the US. Either way, modern PCs do NOT look like that. That's one example of a very, very ugly PC... maybe get a new Dell?


 * First of all don't be so THOUGHTLESS (repeating your silly capitals) in being unkind to the people who spend time in getting pics prepared, uploaded and labelled. It's more work than you seem to realise. I uploaded the "modern PC tower" pic, it's my computer, it's a very well known (still existing) UK brand called Evesham and it's two and a half years old. Why does it have to be known in the US? The US is not the centre of the universe for me!!! Other countries do exist outside of the US, did you know that?! The coke can immediately gives scale to the other pic so seems necessary to me - Adrian Pingstone 09:56, 14 February 2006 (UTC)


 * IMHO the "modern PC" and Libretto are darn good pictures. The backgrounds are uncluttered (showing that either the photographer is tidier than 99.999% of PC owners or that a lot of time and trouble was taken with the picture). The Coke can for scale is appropriate. The article has pictures of typical, PC's of varying sizes and it's helpful to have some way of showing that it's smaller than a laptop.


 * Who cares what brands they are? The article is about generic PC compatibles and there's a genuine IBM at the top. As for it being "typical," it's a very typical tower configuration and whereas ten years ago you could have said that there was a standard, rectangular, beige box appearance, nowadays all of the vendors trick out their cases with decoration. They all look different. Most of them IMHO look rather tacky, like 1950s American cars with tailfins, but the picture is very typical of the genre. It's not supposed to be a picture of a PC that's notable for excellent industrial design, it's supposed to be a picture of a typical modern PC, and it is. I live in the U. S., by the way. And I don't believe the PC's the Chinese build for us are substantially better or worse in appearance than the ones they build for Europeans.


 * Anyway, anon, if you have a better picture, please feel free to upload it. Dpbsmith (talk) 12:18, 14 February 2006 (UTC)

Extreme POV
Come on, guys. This is worse than the Mainframe article. 71.116.217.242 20:46, 4 April 2006 (UTC)