From: ceicher@inav.net (Charles Eicher) Newsgroups: alt.folklore.computers Subject: "A Look Into The Future" from 1975 - by Adam Osborne Date: Mon, 02 Aug 1999 16:33:55 -0500 Organization: Crimson Permanent Assurance Lines: 151 Message-ID: NNTP-Posting-Host: p-058.newsdawg.com Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Newsreader: Yet Another NewsWatcher 2.4.0 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!isdnet!newsfeed.cwix.com!128.32.206.55!newsfeed.berkeley.edu!newsfeed.concentric.net!pln-w!spln!extra.newsguy.com!newsp.newsguy.com!ceicher This is an interesting document I found while scanning through some old manuals, so I thought I'd toss it out to a.f.c for your amusement. This is the final essay from Adam Osborne & Associates "An Introduction to Microcomputers Volume 2: Some Real Microprocessors - Sept 1978." It was published in 1978, but was written in 1975 (see the final line) so I'm giving it the original date. I love looking at old predictions and seeing how they work out.. ----- A LOOK AT THE FUTURE Let us take a moment to gaze into a crystal ball. What types of microcomputer products can we expect to see in the future, and what impact will they have on the minicomputer market? If there is one key aspect of microcomputer design which was not immediately apparent but is becoming more apparent every day, it is that the way logic is distributed among various devices of a microcomputer chip set is fundamentally the most important feature of any microcomputer product. Assembly language instruction sets, addressing modes and even instruction execution times are all of secondary importance in that they become inconsequential providing they meet modest criteria of sufficiency. The logic designer using microcomputers is likely to be far more influenced by control signals on the system bus and by the number of devices he has to work with, rather than by the complexity of the instruction set or its addressing modes. And this, we believe, is the key to a future drift into two types of microcomputer product: the logic device and the computer. If there will be a branch of the microcomputer industry which builds minicomputer look-alikes, what impact will this have on the microcomputer industry? In truth, most manufacturers of computers, mini or larger, are already scrambling to build their central processing units and support logic out of large scale integration devices; therefore, we may conclude that within ten years every computer will be a microcomputer in that every computer will be built out of large scale integrated logic. This does not mean that the microcomputer manufacturers of today will overwhelm the minicomputer and large computer manufacturers of yesterday. This is because programming expenses constitute an already expended front end fixed cost for most users of minicomputers and larger computers, the hardware savings that might be gained by switching from a minicomputer to a microcomputer are simply insignificant when compared to reprogramming expenses. Therefore, those minicomputer manufacturers who can defend their current sales with existing software are likely to be impacted very little by microcomputers. Those minicomputer manufacturers who are essentially selling components are likely to be eliminated from the component market entirely, unless they are able to scale down their minicomputers into microcomputers and survive as component suppliers at the new microcomputer price levels. It is this reduction in prices that opens a window for new products such as the National Semiconductor and Signetics microcomputers to attack markets that look characteristically like minicomputer markets. These are markets which were suited to minicomputer-type products, but in the past could not use minicomputers because of pricing considerations. Now that minicomputer-like devices are available for a few hundred dollars, a large number of new markets open up, none of which have used minicomputers before and none of which have invested in the front end program development fixed costs; the new markets are therefore equally likely candidates for the old minicomputer manufacturer's product or the new microcomputer manufacturer's product. This pseudo-minicomputer buyer will be interested in buying a great deal of support in addition to hardware and will not be quite so influenced by small dollar differences going from one product to another. It is in the area of discrete logic replacement that we may expect to see the greatest volatility among microcomputer manufacturers. The microcomputer user in this market will usually be buying in huge volumes with very little front end programming expense; therefore, this user has a much greater incentive to switch from one microcomputer to another, based solely on pricing considerations. This being the case, the logic device replacement market is the one which will be hardest for established microcomputer manufacturers to defend, and the most attractive to latecomers into the field. It is quite probable that a microcomputer manufacturer who has not established a market for mls on the system bus and by the number of devices he has to work with, rather than by the complexity of the instruction set or its addressing modes. And this we believe is the key to a future drift into two types of microcomputer product: the logic device and the computer. If there will be a branch of the microcomputer industry which builds minicomputer look-alikes, what impact will this have on the microcomputer industry? In truth, most manufacturers of computers, mini or larger, are already scrambling to build their central processing units and support logic out of large scale integration devices; therefore, we may conclude that within ten years every computer will be a microcomputer in that every computer will be built out of large scale integrated logic. This does not mean that the microcomputer manufacturers of today will overwhelm the minicomputer and large computer manufacturers of yesterday. This is because programming expenses constitute an already expended front end fixed cost for most users of minicomputers and larger computers; the hardware savings that might be gained by switching from a minicomputer to a microcomputer are simply insignificant when compared to reprogramming expenses. Therefore, those minicomputer manufacturers who can defend their current sales with existing software are likely to be impacted very little by microcomputers. Those minicomputer manufacturers who are essentially selling components are likely to be eliminated from the component market entirely, unless they are able to scale down their minicomputers into microcomputers and survive as component suppliers at the new microcomputer price levels. It is this reduction in prices that opens a window for new products such as the National Semiconductor and Signetics microcomputers to attack markets that look characteristically like minicomputer markets. These are markets which were suited to minicomputer-type products, but in the past could not use minicomputers because of pricing considerations. Now that minicomputer-like devices are available for a few hundred dollars, a large number of new markets open up, none of which have used minicomputers before and none of which have invested in the front end program development fixed costs; the new markets are therefore equally likely candidates for the old minicomputer manufacturer's product or the new microcomputer manufacturer's product. This pseudo-minicomputer buyer will be interested in buying a great deal of support in addition to hardware and will not be quite so influenced by small dollar differences going from one product to another. It is in the area of discrete logic replacement that we may expect to see the greatest volatility among microcomputer manufacturers. The microcomputer user in this market will usually be buying in huge volumes with very little front end programming expense; therefore, this user has a much greater incentive to switch from one microcomputer to another, based solely on pricing considerations. This being the case, the logic device replacement market is the one which will be hardest for established microcomputer manufacturers to defend, and the most attractive to latecomers into the field. It is quite probable that a microcomputer manufacturer who has not established a market for minicomputer-like devices within the next two or three years will have no further opportunity to do so, however interesting the products he designs. No such window exists in the logic replacement market, where ten years from now a manufacturer who is able to sell microcomputer devices for 10 cents each, where the going rate has been 25 cents each, will be able to establish himself. In conclusion, we predict that microcomputer devices will separate into minicomputer look-alike and logic device replacements. The minicomputer look-alike market will become increasingly harder to break into and will stabilize fairly quickly. The logic device replacement market will continue to spawn products that look nothing like minicomputers and will continue to be extremely volatile until prices have been driven so low that there is simply no room left for further economies. (We have not changed a word of this prediction from the first edition of December, 1975.) ------------------ Charles Eicher -=- ceicher@inav.net ------------------ ###### From: "Jack Peacock" Newsgroups: alt.folklore.computers Subject: Re: "A Look Into The Future" from 1975 - by Adam Osborne Date: Tue, 3 Aug 1999 12:18:09 -0700 Organization: Simco Lines: 43 Message-ID: References: <7o75mp$hfq$1@mail.pl.unisys.com> X-Complaints-To: newsabuse@supernews.com MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Newsreader: Microsoft Outlook Express 5.00.2314.1300 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!skynet.be!news.algonet.se!algonet!newsfeed.icl.net!colt.net!newspeer.clara.net!news.clara.net!remarQ-uK!rQdQ!supernews.com!remarQ.com!corp.supernews.com!not-for-mail Tim McCaffrey wrote in message news:7o75mp$hfq$1@mail.pl.unisys.com... > In article , > ceicher@inav.net says... > > > >This is an interesting document I found while scanning through some old > >manuals, so I thought I'd toss it out to a.f.c for your amusement. This is > >the final essay from Adam Osborne & Associates "An Introduction to > >Microcomputers Volume 2: Some Real Microprocessors - Sept 1978." It was > >published in 1978, but was written in 1975 (see the final line) so I'm > >giving it the original date. I love looking at old predictions and seeing > >how they work out.. > > For some context here (since I read Osborne's books when they came out) the target audience for those books were EEs trying to learn all about the new-fangled microprocessors. IIRC Vol. 1 covered how to emulate TTL type logic functions in software, counters, flip-flops, latches, etc. The EE trade mag editorials were full of dire warnings of obsolescent EEs if they failed to learn how to use micros. At the time I was at Lockheed providing the software support for a group of EEs, their only prior experience being with rack sized minis. The idea that a small PC board with an 8080 could replace several wirewrap logic boards was quite appealing, far less work for them. The drawback was they had no idea how to change it, instead of schematics there was a listing of 8080 assembler code. Course the learning curve went both ways. Coming from Univac1100 and CDC 6000 mainframes, debugging a program embedded in a LIDAR receiver was a whole different world. Tools, emulators, were non-existent then. Even CRT terminals were scarce. Debugging was limited to O-scope probes triggered by pulses from the program. As much as I was showing the EEs how software worked, they were showing me how to set up scope controls to catch transients.and generate synchs to time events. Osborne's books proved invaluable even though I wasn't the target audience, because I could look at the software examples and figure out what J-K flip flops and clock generators did. Jack Peacock ###### From: timothy.mccaffrey@spam2filter.unisys.com.takethisoff (Tim McCaffrey) Newsgroups: alt.folklore.computers Subject: Re: "A Look Into The Future" from 1975 - by Adam Osborne Date: 3 Aug 1999 16:34:33 GMT Organization: A series networking Lines: 25 Message-ID: <7o75mp$hfq$1@mail.pl.unisys.com> References: NNTP-Posting-Host: mccafftm.tr.unisys.com X-Newsreader: WinVN 0.99.9 (Released Version) (x86 32bit) Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!uunet!ams.uu.net!ffx.uu.net!eanews1!eanews1.unisys.com!plnews.pl.unisys.com!not-for-mail In article , ceicher@inav.net says... > >This is an interesting document I found while scanning through some old >manuals, so I thought I'd toss it out to a.f.c for your amusement. This is >the final essay from Adam Osborne & Associates "An Introduction to >Microcomputers Volume 2: Some Real Microprocessors - Sept 1978." It was >published in 1978, but was written in 1975 (see the final line) so I'm >giving it the original date. I love looking at old predictions and seeing >how they work out.. > >----- > >A LOOK AT THE FUTURE > > [Snip] To put it in modern terms, they appear to be talking about the PC market and the embedded market, and I would have to say that the predicition is essentially correct (although some of the specifics are a bit off). Tim McCaffrey ###### From: jsavard@tenMAPSONeerf.edmonton.ab.ca (John Savard) Newsgroups: alt.folklore.computers Subject: Re: "A Look Into The Future" from 1975 - by Adam Osborne Date: Tue, 03 Aug 1999 23:11:11 GMT Organization: PowerSurfr - High Speed Internet Lines: 40 Message-ID: <37a774e2.18193446@news.prosurfr.com> References: NNTP-Posting-Host: c9169-003.v-wave.com X-Trace: dagger.videotron.ab.ca 933721790 24018 24.108.21.103 (3 Aug 1999 23:09:50 GMT) X-Complaints-To: abuse@powersurfr.com NNTP-Posting-Date: 3 Aug 1999 23:09:50 GMT X-Newsreader: Forte Free Agent 1.11/32.235 Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!newsfeed-zh.ip-plus.net!news.ip-plus.net!News.Amsterdam.UnisourceCS!newshunter!cosy.sbg.ac.at!news-feed.inet.tele.dk!bofh.vszbr.cz!hermes.visi.com!news-out.visi.com!newsfeed.direct.ca!news.tac.net!news.videotron.ab.ca!not-for-mail ceicher@inav.net (Charles Eicher) wrote, in part: >This is an interesting document ...and the prediction it made was partly right and partly wrong. He predicted that microcomputers used for embedded systems purposes would keep going down in price, and this happened. He also predicted that, due to lack of software, microcomputers wouldn't displace existing minicomputer manufacturers. Instead, the minicomputer companies would transfer their architectures to silicon, and competition would be locked out. This did not happen - we aren't using PDP-11 computers on a chip nowadays. The dominant architectures - the PC and the Mac - don't correspond to a previous minicomputer software base or architecture. But the prediction was half correct: although minicomputer makers failed to leverage their existing software base to dominate the emerging market, a limited number of microcomputer companies did wind up having an inordinate share of existing software. Thus, the Atari ST and the Amiga fell by the wayside, and even Apple has a hard time competing. And the increase in chip power has meant that a Pentium is very similar to (although not quite as powerful as) an IBM 360/195 (both cache and pipelining) ... and the new IA-64 chips to be out next year are edging into Cray territory. Basically, the main slip he made was that he didn't realize that a large dollar savings would powerfully entice people who could not have a computer otherwise, and this volume allowed microcomputer makers to eventually become able to compete with the makers of other types of computer. Of course, he could not be blamed for failing to predict the GUI. John Savard ( teneerf<- ) http://www.ecn.ab.ca/~jsavard/crypto.htm ###### Sender: eric@ruckus.brouhaha.com From: Eric Smith Newsgroups: alt.folklore.computers Subject: Re: "A Look Into The Future" from 1975 - by Adam Osborne References: <37a774e2.18193446@news.prosurfr.com> X-Disclaimer: Everything I write is false. Organization: Eric Conspiracy Secret Labs X-Eric-Conspiracy: There is no conspiracy. Date: 04 Aug 1999 00:22:30 -0700 Message-ID: Lines: 19 X-Newsreader: Gnus v5.5/Emacs 20.3 NNTP-Posting-Host: ruckus.brouhaha.com X-Trace: 4 Aug 1999 00:38:11 -0800, ruckus.brouhaha.com Path: chonsp.franklin.ch!pfaff.ethz.ch!news-zh.switch.ch!news-ge.switch.ch!enews.sgi.com!news.sgi.com!news.spies.com!ruckus.brouhaha.com jsavard@tenMAPSONeerf.edmonton.ab.ca (John Savard) writes: > And the increase in chip power has meant that a Pentium is very > similar to (although not quite as powerful as) an IBM 360/195 (both > cache and pipelining) ... To which Pentium are referring? If you mean one of the original Pentiums, I wouldn't be surprised. But if you mean that a current Pentium III in a well-configured system can't outperform a 360/195, I'd have to be very skeptical. The modern system executes many more instructions per second, has much more cache, and has a much higher bus bandwidth. With suitable disk controllers and peripheral cards, it also has higher-performance I/O, although not by as large a margin. If you're talking end-user application-level performance, modern systems will lose, but that's a software issue. All the extra MIPS have been thrown at making things pretty. But if you configure the modern machine with a reasonable OS and use it as (for example) a database engine, it will win.