Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!zaphod.mps.ohio-state.edu!mips!daver!ditka!zorch!amiga0!mykes From: mykes@amiga0.SF-Bay.ORG (Mike Schwartz) Newsgroups: comp.sys.amiga.advocacy Subject: Re: New Thread: What _REALLY_ makes a product successful? Message-ID: Date: 26 May 91 20:31:45 GMT References: <72306@microsoft.UUCP> <760@mixcom.COM> <1991May20.173553.11809@convex.com> <24947@well.sf.ca.us> Organization: Amiga makes it possible Lines: 97 In article <24947@well.sf.ca.us> farren@well.sf.ca.us (Mike Farren) writes: >swarren@convex.com (Steve Warren) writes: > >> Marketing was a factor, but not the "unknown" factor. The unknown factor was >>a synergistic combination of marketing and standards, and IBM stumbled into it >>by dumb luck, screaming and thrashing all the way. > >Agreed, but as for the rest of your post, well... > >>Here is how IBM "stumbled" onto standards: because they had larger systems >>that had extremely low performance/price ratios when compared to a >>state-of-the-art micro, IBM decided that they could not afford to sell a micro >>that encroached on their other, more costly systems. So they designed their >>micro systems to be slow and memory limited. They picked a microprocessor >>that they felt would not threaten their other product lines, and they put >>together a system that was altogether underwhelming. Then they marketed it as >>if it were an innovation, and it worked. > >Not at all the case. When the PC was introduced, it was the most powerful >micro system offered by anyone with any kind of mass marketing. The best >you could do at that time, unless you were willing to deal with a "fringe" >outfit like Godbout, was an S100 system, none of which approached the IBM >PC for performance. > >The reason for the 8088 processor was simply because at the time the PC was >designed, the biggest expense was in the support chips, NOT the processor. >The PC designers made reasonable choices in the technical environment of the >time. > Agreed. However, as I see it, the Apple II was gaining acceptance as a business machine (and it was more than competitive with S100) thanks to VisiCalc. IBM saw the future in micros for business and took the opportunity to gain entry with the PC. The 8088 itself was very easy for hardware designers to work with since it was similar enough to the Z80, but IBM did consider the 68000. The original PC was a 16K computer and used a cassette drive. It was definately designed to compete directly with the Apple II, which also came with 16K and cassette drive and had similar graphics standards. >>What happened next is that the free market kicked in. Once a few sharp >>engineers noticed how incredibly low-tech this box was, and how incredibly >>high IBM's margin was, they decided to knock-off some copies. > >Not for a number of years. IBM ruled the roost until Compaq came along >(their first *real* competitor - Eagle, Columbia, and such do *NOT* count, >as their "compatibility" was a joke, and they weren't really all *that* >less expensive). It wasn't until about a year or two after the introduction >of the PC-AT that clones started appearing in any volume. > IBM had only one thing that prevented clone manufacturers from thriving early on. It was their ROM BIOS. A company called Phoenix cloned the BIOS and won a lawsuit with IBM. This BIOS is what makes all the PC Clones "clones". Eagle, Columbia, and the rest didn't have the Pheonix BIOS. When the AT came out, there were a few choices in computers that already had the Pheonix BIOS, including the AT&T 6300. >>When everyone in the world began making and selling these boxes the >>competition forced prices down to the point that PC-compatibles became >>commodities. That was when everyday people began to purchase them. > >But that didn't happen until well into the cycle. 1985 was about the >earliest you could get "commodity" prices on *any* system. The AT originally ran at 6MHz, but you could put an 8MHz crystal in it and it would go 25% faster. IBM actually changed their BIOS ROMS to detect the clock rate and not work at 8MHz. They were trying to develop a market for higher priced faster ATs using this trick. People caught on real quick and offered faster/cheaper alternatives. > >>The magic happens when intensive marketing is combined with standards. In >>IBM's case the "standard" was de facto rather than formal. But nevertheless >>the standard was real. > >No, it was much more formal than most. In most cases, where there *was* >a standard to be followed, IBM followed it. Notice, for example, their >serial port connectors - the *only* ones, up until then, which adhered to >the RS-232 recommendations for DTE vs. DCE connector gender. Likewise, their >use of CR *and* LF for carriage return and line feed, respectively, when >almost every other system used either one, or the other, but not both. >They were also the *only* manufacturer of micros to pay attention to such >niceties as worst-case timing analysis, which accounts for some of the >extremely conservative nature of the IBM PC design. > By the time the AT came out, it was clear that 640K was going to be a problem, and IBM/MicroSoft could have done something about it then. Again, they took the safe way out, even if it did paint their future into an awkward corner. >-- >Mike Farren farren@well.sf.ca.us -- **************************************************** * I want games that look like Shadow of the Beast * * but play like Leisure Suit Larry. * ****************************************************