16

IBM famously created the PC standard with the introduction of the first IBM PC in 1981. IBM also extended the standard with the introduction of the PC/AT (1984) and the PS/2 line (1987). However, by 1987, IBM had lost control of their market and PC clone manufacturers were responsible for most of the new system sales.

I know that it was IBM that initiated most of the standards, such as ST-506 HDDs, the various graphics adapters/modes, BIOS interfaces, connectors, keyboard layout, standard peripheral chips, etc. So, I suspect there are a very few standard technologies first brought to market by one or more clone manufacturers. The well-known example is the EISA bus standard, which was simply to negate the effect of IBM's proprietary Micro-Channel Architecture.

Besides EISA, what eventual PC clone standards were first introduced by manufacturers of clone computers, and who introduced them?

NOTE 1: I'm not looking for a list answer per se, since my assumption is there are very few, if any technological innovations that can be traced to R&D efforts by a PC clone manufacturer (besides EISA).

NOTE 2: I'd greatly prefer an answer where the innovation really came from a clone computer manufacturer, not someone in the PC peripheral or semiconductor business. Companies such as Dell, Gateway, Compaq, Wyse, NEC, Olivetti, Toshiba would be potential examples of companies that manufactured clones.

Brian H
  • 60,767
  • 20
  • 200
  • 362
  • @Tommy I think AdLib, later SoundBlaster, became standards because other companies (besides Creative Labs) promoted their adoption. I think it is worth including in the answer, if a clone manufacturer did something "significant" that aided that adoption. I'd like the answer to challenge my "thesis" that the clone manufacturers DIDN'T contribute much/any tech innovation. – Brian H May 23 '17 at 17:11
  • 1
    There is no standard based around the sound card. There is a convention of similar uses but there was no standard that was agreed on. Saying something was compatible was martketing speak for "buy our cheaper solution not theirs." – Rowan Hawkins May 23 '17 at 18:43
  • @RowanHawkins I think the only distinction there is between an agreed standard and being a de facto standard. And compatible wasn't just "we're cheaper", it was also "we're better but backwards compatible" (ala the Pro Audio Spectrum, or via software the Gravis Ultrasound, or the Sound Blaster itself respective to the Adlib). – Tommy May 23 '17 at 19:11
  • This question feels broad to me. As it stands, it seems to be asking for list answers, which is something the Q&A format in general doesn't handle very well. I'm flagging this for review, and personally would vote to close outright as too broad if I had the reputation to do so here. – user May 23 '17 at 19:12
  • ST-506 was the interface used on a hard disk with the same name introduced by Shugart (new Seagate) in 1980. IBM chose to use it in the PC/XT by selecting controllers made by a third party. – Blrfl May 23 '17 at 21:52
  • @Blrfl they did introduce it into the PC schema by outfitting the XT with an ST506 capable disk controller :) ... before things get too broad, it would be helpful to introduce a deadline year, even if IBM have been manufacturing PCs (Aptiva... Thinkpad...) much much later... maybe introduction of PS2 1987, maybe end of ISA in the mainstream (~2005?), maybe end of the 5V CPU era (~1995?) – rackandboneman May 23 '17 at 23:20
  • Toshiba T1100 is often credited as "first laptop", which isn't a standard but a form factor. – pjc50 May 24 '17 at 08:20
  • @rackandboneman Right, but IBM didn't invent it. – Blrfl May 24 '17 at 11:08

6 Answers6

24

Here are a few innovations that are not attributable to IBM, but also are not attributable to PC clone manufacturers, either:

  • LIM EMS, Lotus / Intel / Microsoft Expanded Memory Specification;
  • IDE, designed by Western Digital (and later expanded into various ATA standards)
  • ATX, designed by Intel
  • USB, designed by a large number of companies including IBM and Intel, based on Atari’s SIO
  • AC’97, designed by Intel
  • Multimedia PC, largely driven by Microsoft
  • PCI and its successors, designed by Intel

There’s an obvious constant here, and it’s not IBM... As far as answering the question goes, Intel isn’t exactly a clone manufacturer in the same way as Compaq etc., but it did produce PC clones for a while. Compaq was involved in the design of a number of the above standards too, if you’re looking for a “real” clone manufacturer. But it’s really Intel (and Microsoft) who have been driving the platform.

Brian H
  • 60,767
  • 20
  • 200
  • 362
Stephen Kitt
  • 121,835
  • 17
  • 505
  • 462
  • I think only LIM/EMS and IDE are old enough to have coincided with IBM's innovations. The rest seem like post-IBM, "Wintel" adaptations. So, effectively, the torch was passed from IBM to Microsoft/Intel and clone makers contributed very little in terms of technology innovations (though they contributed much in lowering prices). – Brian H May 23 '17 at 15:30
  • 3
    "based on Atari's SIO" — citation severely needed. Sounds like revisionism. It's not more similar to Atari's serial interface than it is to Commodore's, or to Apple's desktop bus, and none of those three companies were amongst the seven that developed the initial spec. – Tommy May 23 '17 at 15:57
  • 6
    @Tommy I’ll look up some references; the common link isn’t a company, it’s Joe Decuir, who designed SIO at Atari and then USB at Microsoft (he holds a number of related patents). – Stephen Kitt May 23 '17 at 16:00
  • 2
    That said it might well be revisionism and good PR on Joe’s part... – Stephen Kitt May 23 '17 at 16:15
  • 3
    "Technology X was designed by a person who also designed Technology Y" does not imply "Technology X is based on Technology Y". Perhaps thankfully. – user May 23 '17 at 19:15
  • 1
    IDE could be a false positive here ... it is a simplified ISA bus in its original form :) – rackandboneman May 23 '17 at 19:57
  • Most of the items you mention already post-date the times of IBM leadership in PC standards development. – tofro May 24 '17 at 06:56
  • Ethernet (DEC-Intel-Xerox in its 10Mb/s incarnation) - which is an example of IBM trying to set a standard (Token Ring) and failing. – dave Jul 12 '19 at 16:57
17

The first PC compatible 80386 machine was introduced by Compaq.

The VESA Local Bus, and similar short-lived/proprietary local bus standards, were also not introduced by IBM.

Most graphics cards that had capabilities beyond what the original VGA (IBM specification) offered were by third party manufacturers, and not based on IBM standards (like PGA or 8514a).

Also: dedicated, fan/heatsink-based CPU cooling in a PC (it likely existed in specialized workstation systems), which only became popular when the 486DX50 (not the 486DX2/50, different beast altogether - appeared later, and into an ecosystem where CPU coolers had become common) appeared on the enthusiast's map - while not designed to be run with a cooler, practice quickly proved it needed one :)

psmears
  • 173
  • 4
rackandboneman
  • 5,710
  • 18
  • 23
17

Just two examples of the most important "standards" that came up during the PC's first years:

One of the first and most common PC "standard" graphics card was the Hercules Graphics card that was never part of any IBM "standard". Cards to that de-facto standard were never produced by IBM, although they were very common in the early days and although there were quite some "Hercules clones" after a while.

The computer mouse was introduced to the PC world by Logitech, Mouse Systems and Microsoft in the early 80ies - IBM introduced their first mouse much later, in 1987, with their PS/2 systems.

tofro
  • 34,832
  • 4
  • 89
  • 170
  • ... and the Hercules was widely cloned. So it wasn't just a particular card that sold well. – Tommy May 23 '17 at 17:57
6

The Compaq Portable

Now, not the first portable personal computer, but the first one which was "100% IBM PC compatible". It shipped first in March 1983 (or January, according to some other source). The IBM Portable Personal Computer appeared in February 1984, the IBM PC Convertible followed in 1986.

Now, portable computer were not really an innovation. Then again, the IBM PC, introduced in August 1981, wasn't an innovation either. The first personal computer (which was also named as such) was the Xerox PARC's Alto from 1972. The Xerox Alto had a mouse-driven GUI (yep, a real bitmap graphics display), Ethernet, featured WYSIWYG (they specifically invented the laser printer for that feature) and an object-oriented operating system/development environment. The IBM PC, introduced almost ten years later had none of this.

That the IBM PC was so primitive was a key to its success, as everybody could build a cheap clone. However, if a clone manufacturer wanted to be "100% IBM PC compatible", innovation was severely limited. Several clone manufacturers "improved" the BIOS, as no one needed a computer which boots into a BASIC interpreter (as the original IBM PC did).

When IBM decided to make their IBM PCs incompatible by introducing the Microchannel Architecture (MCA), they lost their market dominance (the competing EISA, developed mainly by Compaq, IIRC, was more successful).

Also probably not invented by a clone manufacturer, but possibly first used for an IBM PC compatible: the tower enclosure.

Not an IBM Clone, but nevertheless the inspiration for the extension card slots of the IBM PC: the Apple II.

So, basically, when looking for innovation, don't look at the IBM PC market, but everywhere else. Some of the innovations in non-PC-compatibles were later incorporated into the PC design.

Anecdotal evidence:

Imagine that Cray computer decides to make a personal computer. It has a 100 MHz processor, 20 megabytes of RAM, 500 megabytes of disk storage, a screen resolution of 1024 X 1024 pixels, relies entirely on voice recognition for input, fits in your shirt pocket and costs $3,000. What's the first question that the computer community asks? "Is it PC compatible?" — InfoWorld, February 1984

You don't ask whether a new machine is fast or slow, new technology or old. The first question is, "Is it PC compatible?" — Creative Computing, November 1984

Klaws
  • 521
  • 3
  • 5
  • Welcome to the site, and thanks for the insightful answer! – Brian H May 24 '17 at 14:25
  • 1
    I dont have rep to comment but @Klaws: the first tower cases were just desktop cases turned on their side. There used to be stands to put next to desks. The motivation was to get the bulky CPU case off your desk not any cooling or accessibility benefits that turned up later. – FlapjackJoe Jul 12 '19 at 13:56
3

The "Turbo XT" 4.77/8mhz clock switching first appeared on no-name clones, and was carried over into the 386 era with 16/33mhz switching.

peterh
  • 1,749
  • 1
  • 14
  • 26
FlapjackJoe
  • 31
  • 1
3

Hal Prewitt patented Hot Plug in 1987, and his company Core was the first to offer hot plug systems disks, controllers etc in 1990, they filed for trademark in 1991.

chx
  • 931
  • 1
  • 8
  • 21