pcg@cs.aber.ac.uk (Piercarlo Grandi) (11/15/90)
On 14 Nov 90 20:46:56 GMT, chuck@trantor.harris-atd.com (Chuck Musciano) said: chuck> And Ed Borasky sends me: chuck> Ed Borasky's Top Ten Computer Architectures: chuck> 1. Babbage's Analytical Engine 2. The IAS machines (ILLIAC I, chuck> JOHHNIAC, etc. 3. The IBM 7030 (STRETCH) 4. System/360 commercial chuck> or scientific jobs) 5. The PDP-8 6. The FPS AP-120B LONG chuck> INSTRUCTION WORD machine 7. The Cray X-MP 8. The Connection chuck> machine machine 9. Intel's Touchstone Delta 10. The MOSTEK 6502 chuck> video games! chuck> Those are MY nominees; clearly others will have favorites. Conspicuosly chuck> missing are ... [ ... RISC machines, but they are post-1980 ... ] ... european machines. Isn't it appalling that the list above is so parochial? European computer industry may be an appalling failure and has gone down the drain, but european computer research does not deserve the same reputation. In particular I would add Zuse's machines from Germany and Kilburn's from Manchester's (to which the title of first supercomputer must surely go) and Wilkes' from Cambridge. In particular, as far as I know, Zuse's group and descendants at Siemens and Kilburn's group and descendants at Manchester are the only two architecture groups in the world to have an unbroken record of developing computer architectures for over 40 years consecutively (Amdahl and Cray are by comparison newcomers). -- Piercarlo Grandi | ARPA: pcg%uk.ac.aber.cs@nsfnet-relay.ac.uk Dept of CS, UCW Aberystwyth | UUCP: ...!mcsun!ukc!aber-cs!pcg Penglais, Aberystwyth SY23 3BZ, UK | INET: pcg@cs.aber.ac.uk
smith@sctc.com (Rick Smith) (11/16/90)
Everyone has their own lists, of course... Here are some of mine, fewer than ten, and in no particular order: The Manchester "baby," oft cited as the first one to run a stored program, and my nominee for the first RISC architecture :-) Everyone else was struggling to build multipliers while Williams and Kilburn put together something that could only subtract and zero the accumulator. The Whirlwind, whose design virtually set the standard for implementing IAS style machines (the IBM 701/4/9... series, everything from DEC, and almost every uniprocessor you see today). The Manchester Atlas, of course, the pioneer of multiprogramming and paging. The Burroughs B5000 and children, pioneering multiprocessing, and high level language support. Turing's Pilot ACE, for being so mind-bending. Seymour Cray's CDC 6600, for a similar reason. It was weird, but fast. The VAX, for burying so many misconceptions about the inherent differences between "little" (e.g. non-mainframe) and "big" computers. It was called a "supermini" computer when it was introduced. Remember? Rick. smith@sctc.com Arden Hills, Minnesota
bil@hq.demos.su (Igor' Bel'chinskiy) (11/19/90)
In article <4876@trantor.harris-atd.com> chuck@trantor.harris-atd.com (Chuck Musciano) writes: >Ed Borasky's Top Ten Computer Architectures: > >1. Babbage's Analytical Engine (first computer) >2. The IAS machines (ILLIAC I, JOHHNIAC, etc. -- the > Von Neumann/Goldstine/Burks designs) >3. The IBM 7030 (STRETCH) -- the first supercomputer >4. System/360 -- the first GENERAL PURPOSE COMPUTER (not specialized for > commercial or scientific jobs) >5. The PDP-8 -- the first commercially successful minicomputer >6. The FPS AP-120B -- the best array processor, first > LONG INSTRUCTION WORD machine >7. The Cray X-MP -- the most successful and best TRUE SUPERCOMPUTER >8. The Connection machine -- the most innovative massively parallel SIMD > machine >9. Intel's Touchstone Delta -- the best MIMD design to date. >10. The MOSTEK 6502 -- brains of the Apple 2, Commodore 64 and LOTS of > video games! > >Those are MY nominees; clearly others will have favorites. If you compare computer architectures from commercial success point of view than you skip electronic watches and HP calculators. :-) If you compare computer architectures from flames' density per instruction set point of view than uncoditionally SYSTEM/360 must be second - next after Intel 80x86 line. (sorry, no ':-)') IMHO, better define criteria at first. Igor.
sysmgr@KING.ENG.UMD.EDU (Doug Mohney) (11/20/90)
In article <1990Nov19.123415.22488@hq.demos.su>, bil@hq.demos.su (Igor' Bel'chinskiy) writes: > >If you compare computer architectures from commercial success point of view >than you skip electronic watches and HP calculators. :-) > >If you compare computer architectures from flames' density per >instruction set point of view >than uncoditionally SYSTEM/360 must be second - next after >Intel 80x86 line. (sorry, no ':-)') > >IMHO, better define criteria at first. > Igor. Maybe we could do the Top 10 cloned-architectures in Eastern Europe? :-) Doug Mohney, Operations Manager, CAD Lab/ME, Univ. of Maryland College Park * Why do VMS system managers get more sleep and less ulcers than their * * UNIX(TM) counterparts, despite the sophistication of UNIX? *
sxr@cs.purdue.EDU (Saul Rosen) (11/21/90)
In article <1990Nov19.123415.22488@hq.demos.su> bil@hq.demos.su (Igor L. Bel'chinskiy) writes: >In article <4876@trantor.harris-atd.com> chuck@trantor.harris-atd.com (Chuck Musciano) writes: >>Ed Borasky's Top Ten Computer Architectures: >> >>1. Babbage's Analytical Engine (first computer) followed by 2 to 10 >IMHO, better define criteria at first. > Igor. Agreed. But I would expect that a list of the ten most important computer architectures, by almost any reasonable criteria would include the architecture of the EDVAC in which the idea of the modern stored program computer was formulated. I would also expect such a list to include the CDC 6600 which was certainly a super computer in its time. Saul Rosen
davidb@brac.inmos.co.uk (David Boreham) (11/21/90)
In article <1990Nov19.123415.22488@hq.demos.su> bil@hq.demos.su (Igor L. Bel'chinskiy) writes: >>9. Intel's Touchstone Delta -- the best MIMD design to date. >>10. The MOSTEK 6502 -- brains of the Apple 2, Commodore 64 and LOTS of ^^^^^^ I think that was MOS Technology and NOT MOSTEK. Two different companies. I am guessing that MOS Tech is now part of Rockwell and I know that MOSTEK is now part of SGS-Thomson. >> video games! >> >>Those are MY nominees; clearly others will have favorites. > David Boreham, INMOS Limited | mail(uk): davidb@inmos.co.uk or ukc!inmos!davidb Bristol, England | (us): uunet!inmos.com!davidb +44 454 616616 ex 547 | Internet: davidb@inmos.com
skipper@motaus.sps.mot.com (Skipper Smith) (11/22/90)
In article <12670@ganymede.inmos.co.uk> davidb@inmos.co.uk (David Boreham) writes: >In article <1990Nov19.123415.22488@hq.demos.su> bil@hq.demos.su (Igor L. Bel'chinskiy) writes: >>>9. Intel's Touchstone Delta -- the best MIMD design to date. >>>10. The MOSTEK 6502 -- brains of the Apple 2, Commodore 64 and LOTS of > ^^^^^^ > >I think that was MOS Technology and NOT MOSTEK. >Two different companies. I am guessing that >MOS Tech is now part of Rockwell and I know >that MOSTEK is now part of SGS-Thomson. > >David Boreham, INMOS Limited | mail(uk): davidb@inmos.co.uk or ukc!inmos!davidb >Bristol, England | (us): uunet!inmos.com!davidb >+44 454 616616 ex 547 | Internet: davidb@inmos.com MOS Technologies is owned by Commodore Business Machin~es, Inc. Skipper Smith Motorola Technical Training All opinion [etc...]
mbutts@mentor.com (Mike Butts) (11/22/90)
The control processor onboard the Voyager spacecraft certainly deserves mention, for greatness of result, if nothing else. -- Michael Butts, Research Engineer KC7IT 503-626-1302(fax:1282) Mentor Graphics Corporation, 8500 SW Creekside Place, Beaverton, Oregon 97005 ...tektronix!sequent!mntgfx!mbutts uunet!mntgfx!mbutts mbutts@mentor.com Any opinions are my own, and aren't necessarily shared by Mentor Graphics Corp.
andras@alzabo.uucp (Andras Kovacs) (11/22/90)
In article <0093FFDB.70843660@KING.ENG.UMD.EDU> sysmgr@KING.ENG.UMD.EDU (Doug Mohney) writes: >Maybe we could do the Top 10 cloned-architectures in Eastern Europe? :-) Most certainly. Without trying to be too exhaustive, the whole so-called 'R' series in the East Block are 360/370 clones (I was an operator on a R-35 which was supposedly the equivalent of an IBM 370/145 - is there such IBM model? - and on a R-40 which was some System 360 equivalent). In Hungary we had the TPAs which were PDP clones (I believe PDP-8). The Soviets had some decent PDP-11 clones (can't recall the name). But the best were the East Germans They were making the Zilog 80 chip (I don't know under which arrangement). So they had this Z-80 @ 2.5MHz - it was HOT! I mean, you had to divert some cool air onto it; it run very hot (at least in that machine I was working with). But they had a machine which had this Z-80 int it and emulated (I believe) the PDP-11 instruction set! I recall that the programmer ladies(!) were afraid to change anything in their COBOL programs 'cause a recompilation on a ~200 line source took ~30 minutes... (Do you know that I can pick up a 40MB hard disk at Budapest for less than here in Ottawa? Hungarians are hot for high tech and we started to make IBM PC clones around 1985...) Egeszsegetekre -- Andras Kovacs andras@alzabo.UUCP
avg@hq.demos.su (Vadim G. Antonov) (11/23/90)
In article <0093FFDB.70843660@KING.ENG.UMD.EDU> >sysmgr@KING.ENG.UMD.EDU (Doug Mohney) writes: > >Maybe we could do the Top 10 cloned-architectures in Eastern Europe? :-) IMHO we have cloned THE WORST architectures (hm, especially IBM mainframes). (Strange, but I had to write a program in PL/I today... I didn't touched it for seven years... :-) Vadim Antonov DEMOS, Moscow, USSR
rwt@ohm.york.ac.uk (Richard Taylor) (11/23/90)
looking at the progression of suggested ideas, I would suggest that Jacquards Loom should really head the list.. Richard -- ****************************************************************** * Dr R W Taylor tel : (44) 904 432351 fax : 432353 * * Adaptive Systems Group email : rwt1@uk.ac.york * * University of York, England *
sms@lonex.radc.af.mil (Steven M. Schultz) (11/24/90)
In article <1990Nov23.080156.11942@ohm.york.ac.uk> rwt@ohm.york.ac.uk (Richard Taylor) writes: >looking at the progression of suggested ideas, I would suggest that >Jacquards Loom should really head the list.. i'm a bit surprised that the pdp-11 didn't make the list. considering that this thread started off in a Unix vein i thought for sure the 11 would at least make the list on that account alone (the pdp-7 doesn't seem to have any systems still running). Steven Schultz sms@lonex.radc.af.mil