brooks@physics.llnl.gov (Eugene D. Brooks III) (10/04/90)
In article <2721@crdos1.crd.ge.COM> davidsen@crdos1.crd.ge.com (bill davidsen) writes: > I can't think of any, but back when 32 vs 36 was still being debated, >many minis were made with 16 or 18 bit words, so did anyone ever build a >9-bit byte system? I believe that the S1 built at LLNL had 9 bit bytes, what else do you do with 36 bit words? It had 18 bit floats, handled 36 bit complex values, 36 bit floats, and 72 bit "double precision" values. Are we having fun yet?
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (10/04/90)
There has been discussion of computer word size, does the number of bits have to be a power of two for new systems, etc. I was looking at a discussion in another group and saw a really nice way to solve the problem, but rejected it because it wasn't portable to any system. The question is, has anyone ever made a general purpose computer with and odd word size? No one doesn't count, thank you bit slicers. I can't think of any, but back when 32 vs 36 was still being debated, many minis were made with 16 or 18 bit words, so did anyone ever build a 9-bit byte system? -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) VMS is a text-only adventure game. If you win you can use unix.
jkenton@pinocchio.encore.com (Jeff Kenton) (10/04/90)
From article <2721@crdos1.crd.ge.COM>, by davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr): > > The question is, has anyone ever made a general purpose computer with and > odd word size? No one doesn't count, thank you bit slicers. > > I can't think of any, but back when 32 vs 36 was still being debated, > many minis were made with 16 or 18 bit words, so did anyone ever build a > 9-bit byte system? > Well (trick answer time) the PDP-10 let you use any size "bytes" you wanted, up to 36, and lots of people used 6, 7, 8 and 9 for different things. At Adage in the early '70s we had a 30-bit word machine which made serious use of the 15-bit halfwords. All in all, the answer to your question is probably "no". ----- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ----- ----- jeff kenton --- temporarily at jkenton@pinocchio.encore.com ----- ----- --- always at (617) 894-4508 --- ----- ----- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -----
jones@pyrite.cs.uiowa.edu (Douglas W. Jones,201H MLH,3193350740,3193382879) (10/04/90)
From article <2721@crdos1.crd.ge.COM>, by davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr): > > ... The > question is, has anyone ever made a general purpose computer with and > odd word size? No one doesn't count, thank you bit slicers. > Back in the days of the DEC-10, a 36 bit machine, the byte size was whatever you wanted (the DEC-10 had byte manipulation instructions that allowed any size, although 6 and 9 were the popular ones because they divided evenly into 36. The wierdest word-size I've seen was an early proposal for the ILLIAC II, written, I think, by Don Gillies. Someone with access to the tech reports from the U of Illinois back in the 1950's might be able to find it. The proposal called for a 53 bit word made of four 13 bit bytes and a sign bit. This is the only serious proposal I can remember seeing that had a prime number of bits per word (discounting oddities with one or two bits per word). ILLIAC II Instructions in this proposal were to be two bytes each, with a main memory address of one byte (8K 53 bit words is, after all, a reasonable memory size by the standards of the 1950's). One of the most interesting parts of the proposal was a loop control instruction that referenced a loop control word with something like the following fields (one byte each): loop counter (in the least significant byte) amount to increment counter by final value of loop counter location to branch to if loop counter not equal to final value Indexing was only supposed to use the least significant byte of the loop counter, so the loop control register could also be used as an index register. The loop control instruction could update the counter, compare it with the final value, and do the conditional branch in parallel, all in one instruction cycle. If memory serves me correctly, ILLIAC II was built as a 48 bit machine, a far less interesting word size. Doug Jones jones@herky.cs.uiowa.edu
doudna@nsc.nsc.com (David Doudna) (10/04/90)
From article <2721@crdos1.crd.ge.COM>, by davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr): > > There has been discussion of computer word size, does the number of > bits have to be a power of two for new systems, etc. I was looking at a > discussion in another group and saw a really nice way to solve the > problem, but rejected it because it wasn't portable to any system. The > question is, has anyone ever made a general purpose computer with and > odd word size? No one doesn't count, thank you bit slicers. > > I can't think of any, but back when 32 vs 36 was still being debated, > many minis were made with 16 or 18 bit words, so did anyone ever build a > 9-bit byte system? If I remember right, the PDP-8 I worked on used 12-bit words. Not numerically "odd", but certainly ODD! :-) -David Doudna doudna@nsc.nsc.com
dik@cwi.nl (Dik T. Winter) (10/04/90)
(Cross-posted to alt.folklore.computers; followups go overthere.) In article <12857@encore.Encore.COM> jkenton@pinocchio.encore.com (Jeff Kenton) writes: > From article <2721@crdos1.crd.ge.COM>, by davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr): > > The question is, has anyone ever made a general purpose computer with and > > odd word size? No one doesn't count, thank you bit slicers. > All in all, the answer to your question is probably "no". > The answer is most certainly 'yes'. When I learned programming back in the sixties we had an Electrologica X8 (originally designed by the Mathematisch Centrum, now called Centrum voor Wiskunde en Informatica (CWI), later taken over by Philips). Word size was 27 bits (can you get odder). I still can quote from memory 67108864 (MAXINT+1). It was the successor of the X1, also with 27 bit words. There were quite a few X1's and X8's installed; not only in the Netherlands, but also in Germany for instance (Karlsruhe University if I remember right). The machine was ones complement. Floating point used two words. I still have the programmer's manual next to me here at home. And it was certainly general purpose. The machine we had was decommisioned in about 1976. The machine had very interesting features. I believe it was the first machine that had a separate I/O processor. And it was this machine and its precursor that got Dijkstra to his semaphore stuff. Further characteristics: 64 Kwords of memory of which 32 K directly addressable (when I started only 32 K was enabled). OS was simple: a job consisted of an Algol 60 program, possibly followed by its data. The system would fire up the compiler wich would compile the program and initiate execution. When done the next program was started. Later multi-tasking was implemented with four batch streams, each with its own characteristics. And again later a system was created that allowed interactive operations, although that was for special occasions only. There was another OS that would do Fortran programs but that was nearly never used at our institute. And initially job input was on 7-level papertape (a special coding derived from IBM 7-level). Later versions allowed also 5-level (Baudot/ALCOR) coding, 8-level (ASCII) coding and punched cards (IBM-029 modified). Mmm, that is some time ago. -- dik t. winter, cwi, amsterdam, nederland dik@cwi.nl
EGNILGES@pucc.Princeton.EDU (Ed Nilges) (10/04/90)
In article <2721@crdos1.crd.ge.COM>, davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) writes: > > There has been discussion of computer word size, does the number of >bits have to be a power of two for new systems, etc. I was looking at a >discussion in another group and saw a really nice way to solve the >problem, but rejected it because it wasn't portable to any system. The >question is, has anyone ever made a general purpose computer with and >odd word size? No one doesn't count, thank you bit slicers. On page 27 of Andrew S. Tanenbaum's book STRUCTURED COMPUTER ORGANIZA- TION (Prentice-Hall 1976), there is a list of computers that have been sold commercially and their word size. All are even numbers save for one. This is the "Electrologica X8", with "27 bits per cell." I have never heard anything else about this machine, which sounds like a vacuum cleaner.
meindert@inducom.UUCP (Meindert Kuipers) (10/04/90)
From article <2721@crdos1.crd.ge.COM>, by davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr): > > There has been discussion of computer word size, does the number of > bits have to be a power of two for new systems, etc. I was looking at a > discussion in another group and saw a really nice way to solve the > problem, but rejected it because it wasn't portable to any system. The > question is, has anyone ever made a general purpose computer with and > odd word size? No one doesn't count, thank you bit slicers. Although not odd word size, my good old HP-41 calculator has a little CPU with 12-bit instructions. The memory space for these "assembly" programs is completely 12 bits. The data space however has words of 56 bits, allowing 7 bytes of data, or 10 digit precision (with exponent). The CPU (and all the memory chips and ROMS) all have a 1-bit databus (serial bus) with a special protocol. I once built a ROM/RAM expansion box, and a lot of hardware went into decoding the serial bus. Do HP's new calculators use the same principle? I have understood that the 10-series (HP12, HP16) use the same processor. ------------------------------------------------------------------------------- +-----+ Inducom Systems B.V. | < Raadhuislaan 27 NL - 5341 GL Oss, Netherlands | o | P.O. Box 627 NL - 5340 AP Oss, Netherlands | INDUCOM SYSTEMS Phone: (31)-(0)4120-41922 +-----+ Fax: (31)-(0)4120-22640 Specialists in OS-9, VMEbus and G-64 Meindert Kuipers, Inducom Systems B.V. UUCP: meindert@inducom.UUCP ...!uunet!mcsun!hp4nl!inducom!meindert -- Coming soon to a VMEbus system near you: VMEtro BusBusters -- -------------------------------------------------------------------------------
ath@prosys.se (Anders Thulin) (10/04/90)
In article <11791@pucc.Princeton.EDU> EGNILGES@pucc.Princeton.EDU writes: >In article <2721@crdos1.crd.ge.COM>, davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) writes: >>[ ... ] The >>question is, has anyone ever made a general purpose computer with and >>odd word size? > > [ ... ] the "Electrologica X8", with "27 bits per cell." Aren't those 27 bits really 24 'real' bits + 3 bits for parity checking? -- Anders Thulin ath@prosys.se {uunet,mcsun}!sunic!prosys!ath Telesoft Europe AB, Teknikringen 2B, S-583 30 Linkoping, Sweden
tgg@otter.hpl.hp.com (Tom Gardner) (10/04/90)
The Elliott 803 had a 39 bit word which could contain two 19 bit instructions plus a modifier bit. Using modern terminology, the modifier allowed indexed addressing modes to be synthesised. Max 8K words of memory 0.576ms (not a typo) instruction cycle time.
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (10/04/90)
In article <11791@pucc.Princeton.EDU> EGNILGES@pucc.Princeton.EDU writes: | On page 27 of Andrew S. Tanenbaum's book STRUCTURED COMPUTER ORGANIZA- | TION (Prentice-Hall 1976), there is a list of computers that have | been sold commercially and their word size. All are even numbers | save for one. This is the "Electrologica X8", with "27 bits per cell." | I have never heard anything else about this machine, which sounds | like a vacuum cleaner. Thanks, someone who had actually programmed one mentioned this machine. The root of the question was an algorithm which is vastly faster than the one I have now, but requires an even number of bits in the size of an int. Obviously some machines have been created which will not run this code, so I can't use it. I probably wouldn't anyway, because someday someone will ask me to port it to somthing... I do appreciate all the people who posted and mailed answers to this. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) VMS is a text-only adventure game. If you win you can use unix.
lstowell@pyrnova.pyramid.com (Lon Stowell) (10/05/90)
Sure, lots of them. Examples are the old RCA 301 which used BCD code internally.....12 bits per character..... Also some of the early Univacs used a duodecimal based numbering scheme based on 12 rather than the 16 of "modern" computers....
news@haddock.ima.isc.com (overhead) (10/05/90)
In article <11791@pucc.Princeton.EDU> EGNILGES@pucc.Princeton.EDU writes: >...All are even numbers >save for one. This is the "Electrologica X8", with "27 bits per cell." >I have never heard anything else about this machine, which sounds >like a vacuum cleaner. From the Jan 18th, 1990 Boston Globe: Now, Lechmere is not only selling the incredible VAX, (R) they're giving them away too. Stephen Uitti suitti@ima.ima.isc.com
philip@beeblebrox.dle.dg.com (Philip Gladstone) (10/05/90)
In article <2721@crdos1.crd.ge.COM> davidsen@crdos1.crd.ge.com (bill davidsen) writes: > > There has been discussion of computer word size, does the number of >bits have to be a power of two for new systems, etc. I was looking at a >discussion in another group and saw a really nice way to solve the >problem, but rejected it because it wasn't portable to any system. The >question is, has anyone ever made a general purpose computer with and >odd word size? No one doesn't count, thank you bit slicers. I once owned a machine with a 7-bit byte size and it tended to use two bytes to make a 14-bit word. It was a Univac buffer processor built in the mid 1960s. I got it cheap from British European Airways who used it (and several like it) to front end their booking system. I never suceeded in doing much with it as I couldn't work out (a) how to do subroutine calls, and (b) how to do input. Getting the code in was no problem as it had a REALLY impressive set of neon lights and push buttons. Each light was slaved to a bit in a register and pressing the button would force the bit to a 1 EVEN if the machine was running. Thus you set up an auto-incrementing store as the current instruction, set the address register to where you wanted the code to go, and then just filled in the accumulator with the 7-bits you wanted to store; pressing the microcycle step button would then execute the store and away you went. Still, it did have a very fast interrupt context switch (0 microsecs), but the penalty it paid for that was a very slow everything else. The basic instruction rate was one per 28us (four memory cycles). Anything interesting took five or six cycles (42us). My workstation runs at 1 instruction per 40ns -- I suppose that is progess of sorts! Philip Gladstone philip@dle.dg.com Development Lab Europe C=gb/AD=gold 400/PR=dgc/O=dle Data General, Cambridge /SN=gladstone/GN=philip England. +44 223-67600
mash@mips.COM (John Mashey) (10/05/90)
In article <18429@haddock.ima.isc.com> suitti@anchovy.UUCP (Stephen Uitti) writes: >In article <11791@pucc.Princeton.EDU> EGNILGES@pucc.Princeton.EDU writes: >>...All are even numbers >>save for one. This is the "Electrologica X8", with "27 bits per cell." >>I have never heard anything else about this machine, which sounds >>like a vacuum cleaner. > >From the Jan 18th, 1990 Boston Globe: > > Now, Lechmere is > not only selling > the incredible VAX, (R) > they're giving them > away too. The VAX is indeed a vacuuum cleaner, in fact "3 vacuum cleaners in one". I'd heard of it before, but a few months ago, I was in a hotel in Australia, and saw a commercial, and even got a photo of it... I'm not sure if this is just an apocryphal story, but I'd swear that I heard there was once a lawsuit against DEC regarding this name, where the judge finally threw it by saying "you say there's going to be confusion in the market. Are you really telling me people can't tell the difference between a vacuum cleaner and a computer, and buy one when they think they're getting the other? be serious." or words to that effect. (Maybe somebody who really knows can shed more light, although this really belongs in alt....) -- -john mashey DISCLAIMER: <generic disclaimer, I speak for me only, etc> UUCP: mash@mips.com OR {ames,decwrl,prls,pyramid}!mips!mash DDD: 408-524-7015, 524-8253 or (main number) 408-720-1700 USPS: MIPS Computer Systems, 930 E. Arques, Sunnyvale, CA 94086
wunder@orac.HP.COM (Walter Underwood) (10/05/90)
Somewhere, I have a photocopy of the reference card for an Arcturus A-17, which has, of course, a 17-bit word. Sixteen bits plus an indirect bit. I think it was a British mini, of the PDP-8 persuasion (like the HP1000 and DG Nova). Perhaps this is the largest prime word length? wunder
jim@garland.UUCP (Jim Darby) (10/05/90)
In article <11791@pucc.Princeton.EDU> EGNILGES@pucc.Princeton.EDU writes: >On page 27 of Andrew S. Tanenbaum's book STRUCTURED COMPUTER ORGANIZA- >TION (Prentice-Hall 1976), there is a list of computers that have >been sold commercially and their word size. All are even numbers >save for one. This is the "Electrologica X8", with "27 bits per cell." >I have never heard anything else about this machine, which sounds >like a vacuum cleaner. Oh ye modern hackers! <Yorkshire accent on> Back the good old days I used to use an Elliott 803B. This fine beast had a 39 bit word, and up to 8K words. Now, you're wondering `How did this work then?'. Well, I'll tell you.... The words could either be considered as a 39 bit word for numerical stuff (boring) or as *two* instructions. `But how do you split up 39 bits into two?' you cry. Well, barf bags out at this point, it worked as follows: +----------+------------------------+-+----------+------------------------+ | F1 | ADDR 1 |B| F 2 | ADDR 2 | | 6 bits | 13 bits | | 6 bits | 13 bits | +----------+------------------------+-+----------+------------------------+ So, we have two instructions (F1 and F2 (F = function)), and two operand addresses (ADDR 1 and ADDR 2). This meant that the PC had a notional half value when executing the second (F2) instruction. Because there wasn't a half bit on the PC there were two versions of all jumps, one to jump to the first instruction and another to jump to the second. All rather silly, but not too grotesque. BUT WHAT ABOUT THIS MYSTERIOUS `B' THING IN THE MIDDLE? Well, this is where it gets *fun*. If the `B' bit was clear, then the machine would execute the left hand then the right hand instruction. However, if the `B' bit was SET, then the value in the address ADDR 1 would be added to F2 and ADDR 2 *before* F2 with its ADDR 2 operand were executed, allowing address modification. So, to load the accumulator indirect through location 100 you would do the following. NOP 100 ! LDA 0 The NOP 100 would do nothing with address 100, however the `!' set the `B' bit so the contents of location 100 would be added to the LDA address. Good eh? Now, all the hackers amongst you will have noticed that I said that the `B' bit caused the F2 as well as the ADDR 2 field to be added to. This is all OK if the modifier was still in the address range, but if you went over 8K words then you modified the instruction field. This is how the loader worked. I took me ages to figure out that the way it entered your program was to modify the instruction that stored the data into a jump to the start of the code! Boffo, a good one! Makes the Sparc seem sane, doesn't it? Jim.
chuck@trantor.harris-atd.com (Chuck Musciano) (10/05/90)
In article <41932@mips.mips.COM>, mash@mips.COM (John Mashey) writes: > The VAX is indeed a vacuuum cleaner, in fact "3 vacuum cleaners in one". > I'd heard of it before, but a few months ago, I was in a hotel in > Australia, and saw a commercial, and even got a photo of it... The local Sears in these parts sells the entire VAX line. In fact, there is a sign advertising the VAX right next to the big display for the PS/1. I always find this amusing, but you know, most of the people I point this out to don't. -- Chuck Musciano ARPA : chuck@trantor.harris-atd.com Harris Corporation Usenet: ...!uunet!x102a!trantor!chuck PO Box 37, MS 3A/1912 AT&T : (407) 727-6131 Melbourne, FL 32902 FAX : (407) 729-2537 A good newspaper is never good enough, but a lousy newspaper is a joy forever. -- Garrison Keillor
EGNILGES@pucc.Princeton.EDU (Ed Nilges) (10/05/90)
In article <818@garland.UUCP>, jim@garland.UUCP (Jim Darby) writes: >In article <11791@pucc.Princeton.EDU> EGNILGES@pucc.Princeton.EDU writes: > >>On page 27 of Andrew S. Tanenbaum's book STRUCTURED COMPUTER ORGANIZA- >>TION (Prentice-Hall 1976), there is a list of computers that have >>been sold commercially and their word size. All are even numbers >>save for one. This is the "Electrologica X8", with "27 bits per cell." >>I have never heard anything else about this machine, which sounds >>like a vacuum cleaner. > >Oh ye modern hackers! <Yorkshire accent on> Back the good old days I used to The IBM 1401 was a 1959 computer that many middle-aged posters will remember. Although in the terminology of that distant era the 1401 was a character rather than a word machine it can be said to have had a 7-bit word. This was because its 6-bit characters had an extra bit, called a word mark. Word marks delimited fields and (as an unsung side benefit) allowed character arithmetic of completely unlimited precision. This was how I floored my math professor by calculating the precise value of 100 factorial in 1972, years before Mathematica and REXX, two systems which can calculate the exact value of 1000 factorial in seconds. We also managed to compile and execute an interpretive FORTRAN in only 8K of memory...completely on (gack! neep!) punch cards. We're more productive now, but computing sure was fun then. "The dark chasm and abyss of time"
sritacco@hpdmd48.boi.hp.com (Steve Ritacco) (10/05/90)
It is my understanding that DEC actually made a few 18 bit word machines. It was the PDP-? (I can't remember)! The reason I know this is that triple-i (a company that makes typesetters) latched on to the architecture for their equipment, and as far as I know still manufacture a clone of it today (pitiful no?). I know this because I worked for them for a short while.
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (10/05/90)
In article <129258@pyramid.pyramid.com> lstowell@pyrnova.pyramid.com (Lon Stowell) writes: | Sure, lots of them. Examples are the old RCA 301 which used | BCD code internally.....12 bits per character..... | | Also some of the early Univacs used a duodecimal based numbering | scheme based on 12 rather than the 16 of "modern" computers.... Ummm, most people would consider 12 and 16 "even" rather than odd. I sure am learning a lot about "strange" computer words, though. -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) VMS is a text-only adventure game. If you win you can use unix.
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (10/05/90)
In article <41932@mips.mips.COM> mash@mips.COM (John Mashey) writes: | The VAX is indeed a vacuuum cleaner, in fact "3 vacuum cleaners in one". Our local Sears is selling those, and their TV ads all were using the catch phrase "nothing sucks like a VAX" (I am *not* making that up). I know people who have been telling me that for a decade now ;-) -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) VMS is a text-only adventure game. If you win you can use unix.
gideony@microsoft.UUCP (Gideon YUVAL) (10/06/90)
In article <12857@encore.Encore.COM> jkenton@pinocchio.encore.com (Jeff Kenton) writes: >> >> The question is, has anyone ever made a general purpose computer with and >> odd word size? No one doesn't count, thank you bit slicers. >All in all, the answer to your question is probably "no". I think the AN/UYK-21 has a 21-bit word; I also think the Russians made a 37-bit machine (?Ural or Minsk?) -- they had a parity bit, and decided at the last moment to make it user-visible. -- Gideon Yuval, gideony@microsof.UUCP, 206-882-8080 (fax:206-883-8101;TWX:160520)
gillies@m.cs.uiuc.edu (10/06/90)
I am surprised DEC has not sued over the reuse of the trademark "VAX". After all, the beatles were nasty enough to sue Apple when they wanted to offer a MIDI interface. Kudos to DEC! Of course, it could be that Sears was once a DEC VAX shop, got fed up, switched to Big Blue, and decided to drag the name through the mud with their marketing folks!
rpw3@rigden.wpd.sgi.com (Rob Warnock) (10/08/90)
In article <2515@ns-mx.uiowa.edu> jones@pyrite.cs.uiowa.edu (Douglas W. Jones,201H MLH,3193350740,3193382879) writes: +--------------- | Back in the days of the DEC-10, a 36 bit machine, the byte size was | whatever you wanted (the DEC-10 had byte manipulation instructions that | allowed any size, although 6 and 9 were the popular ones because they | divided evenly into 36. +--------------- 6 & 9 were used for some things, as was even 8, occasionally, but the overwhelming majority of all byte operations were on *7* bit bytes, which is what the PDP-10 used for ordinary ASCII text files. Yes, packing five 7-bit bytes into a 36-bit word wastes a bit. There were early PDP-10 editors that tried to make use of this bit to flag a word as containing a 5-digit "line number" -- five decimal digits in ASCII -- but that hack died out... eventually. The 36th bit was just left fallow. The second most common size was 6, which is how file names were encoded in directories. (Which explains why DEC file names have 6 chars and a 3-char "extension": FOOBAR.BAZ) There was lots of 6 <-> 7 conversion going on, but the hardware byte load/store were indirect through "byte pointers" which contained the byte size, so conversions were cheap (load/store loop). 8-bit bytes were used when writing "industry-compatible" magtapes, and when talking to the PDP-11 communications front-end. I don't recall any production software that used 9-bit bytes, though at least one attempt at an experimental C compiler used them. -Rob ----- Rob Warnock, MS-9U/510 rpw3@sgi.com rpw3@pei.com Silicon Graphics, Inc. (415)335-1673 Protocol Engines, Inc. 2011 N. Shoreline Blvd. Mountain View, CA 94039-7311
rpw3@rigden.wpd.sgi.com (Rob Warnock) (10/08/90)
In article <2721@crdos1.crd.ge.COM> davidsen@crdos1.crd.ge.com (bill davidsen) writes: +--------------- | The question is, has anyone ever made a general purpose computer with an | odd word size? +--------------- I almost forgot! The ancient & venerable LGP-30 (Royal Precision, Librascope Division, circa 1959) was a drum machine with 4096 words of *31* bits each. The accumulator, which recirculated continuously, was actually 32 bits, but when you did a store to memory only the upper 31 bits got stored. The "32nd" bit was a separator between words in memory, giving needed time to turn the write gate on and off, so it was forced to zero. The instructions (all 16 of them!) did not fill the word length, and were not aligned with anything in particular. As I recall, it was: 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |T| |op code| | track | sector | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ <---- memory address ---> The "T" ("transfer") bit was an extension to the opcode that affected only the branch and halt instructions, making them conditional. Data input from the Flexowriter in "4-bit mode" or "6-bit mode" was shifted in from the right. In either mode, the low-order 4 or 6 bits of whatever character you typed went in. "Bit 31" of the accumulator (the "32nd" bit) was included in the I/O shift-in, so the low order bit of the last byte input was dropped when you stored a word (unless you multiplied the accumulator by two first, which was needed for single-character input of the 6-bit set). The character code assignments for the opcode mnemonics were cleverly (?) arranged so that when typing in the bootstrap (which was actually called a "shoelace", there being a larger "bootstrap" which came next), one typed assembly language with absolute addresses. A load -- or as the LGP-30 called it, "bring" -- instruction was opcode 1, which value was also produced by "b". Thus "b1234" became 0001 0001 0010 0011 0100, or opcode 1, track 18, sector 13. Because bit "32" wasn't stored, and because the sector address was one bit up from the bottom of the word, memory addresses incremented in 4's. That is, one counted memory as 0, 4, 8, j, 10, 14, 18, 1j, 20, 24... Oh yes, since "a" was already used for "add", "b" for "bring", and "d" for "divide", to avoid confusion (!?!), they had to use other letters for hexidecimal numbers: 0-9, f, g, j, k, q, w. [Seemed natural enough at the time; it was the first hex I learned!] -Rob ----- Rob Warnock, MS-9U/510 rpw3@sgi.com rpw3@pei.com Silicon Graphics, Inc. (415)335-1673 Protocol Engines, Inc. 2011 N. Shoreline Blvd. Mountain View, CA 94039-7311
dolf@idca.tds.PHILIPS.nl (Dolf Grunbauer) (10/08/90)
In article <3300188@m.cs.uiuc.edu> gillies@m.cs.uiuc.edu writes: >I am surprised DEC has not sued over the reuse of the trademark "VAX". >After all, the beatles were nasty enough to sue Apple when they wanted >to offer a MIDI interface. It is a little more subtle than this. When Apple (= the computer company) started they contacted the Beatles. They (= The Beatles Apple) said that is was ok to use Apple as name as long as they (= the computer company) would not get involved with music. The Apple computer company agreed. Now they (= apple computers) started MIDI and that is when the trouble started. So I am only expecting trouble with VAX when it is going to implement user programmable interfaces in their vacuum cleaners or when DEC builds vacuum cleaners into their machines. -- _ _ / U | Dolf Grunbauer Tel: +31 55 433233 Internet dolf@idca.tds.philips.nl /__'< Philips Information Systems UUCP ...!mcsun!philapd!dolf 88 |_\ Some kind of happiness is measured out in miles
davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr) (10/08/90)
In article <71390@sgi.sgi.com> rpw3@sgi.com (Rob Warnock) writes: | The ancient & venerable LGP-30 (Royal Precision, Librascope Division, circa | 1959) was a drum machine with 4096 words of *31* bits each. The accumulator, | The instructions (all 16 of them!) did not fill the word length, and | were not aligned with anything in particular. As I recall, it was: What a delightful machine. Forgive me if I don't offer to do a C compiler for it ;-) -- bill davidsen (davidsen@crdos1.crd.GE.COM -or- uunet!crdgw1!crdos1!davidsen) VMS is a text-only adventure game. If you win you can use unix.
steve@Pkg.Mcc.COM (Steve Madere) (10/09/90)
In article <41932@mips.mips.COM>, mash@mips.COM (John Mashey) writes: | | The VAX is indeed a vacuuum cleaner, in fact "3 vacuum cleaners in one". | I'd heard of it before, but a few months ago, I was in a hotel in | Australia, and saw a commercial, and even got a photo of it... | | I'm not sure if this is just an apocryphal story, but I'd swear that | I heard there was once a lawsuit against DEC regarding this name, | where the judge finally threw it by saying "you say there's going to be | confusion in the market. Are you really telling me people can't | tell the difference between a vacuum cleaner and a computer, and buy | one when they think they're getting the other? be serious." | or words to that effect. (Maybe somebody who really knows can shed | more light, although this really belongs in alt....) | -- The only thing they might confuse is the applicability of the company slogan: Nothing sucks like a VAX. :-) Steve Madere
tropp@ce.chalmers.se (Ulf Tropp) (10/09/90)
In article <41932@mips.mips.COM> mash@mips.COM (John Mashey) writes: >In article <18429@haddock.ima.isc.com> suitti@anchovy.UUCP (Stephen Uitti) writes: <>In article <11791@pucc.Princeton.EDU> EGNILGES@pucc.Princeton.EDU writes: <>>...All are even numbers <>>save for one. This is the "Electrologica X8", with "27 bits per cell." <>>I have never heard anything else about this machine, which sounds <>>like a vacuum cleaner. <> <>From the Jan 18th, 1990 Boston Globe: <> <> Now, Lechmere is <> not only selling <> the incredible VAX, (R) <> they're giving them <> away too. < <The VAX is indeed a vacuuum cleaner, in fact "3 vacuum cleaners in one". <I'd heard of it before, but a few months ago, I was in a hotel in <Australia, and saw a commercial, and even got a photo of it... < In Sweden, at a household fair, they advertised a vacuum cleaner called UNIVAC. It was, I think, a centralized system with the machinery stashed away in the basement and tubing inside the walls which you would connect the hose to. I'm not joking! Is there something to learn from this? By the way, at one of the Canary Islands, we were supposed to put out any hotel fire with a fire extinguisher called UNIX. I wonder what AT&T in Spain has to say? Ulf Tropp, tropp@ce.chalmers.se
jkenton@pinocchio.encore.com (Jeff Kenton) (10/09/90)
From article <14900017@hpdmd48.boi.hp.com>, by sritacco@hpdmd48.boi.hp.com (Steve Ritacco): > It is my understanding that DEC actually made a few 18 bit word machines. > It was the PDP-? (I can't remember)! The PDP-1 had 18 bits. Nice machine. ----- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ----- ----- jeff kenton: consulting at jkenton@pinocchio.encore.com ----- ----- always at (617) 894-4508 ----- ----- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -----
jones@pyrite.cs.uiowa.edu (Douglas W. Jones,201H MLH,3193350740,3193382879) (10/09/90)
From article <14900017@hpdmd48.boi.hp.com>,
by sritacco@hpdmd48.boi.hp.com (Steve Ritacco):
> It is my understanding that DEC actually made a few 18 bit word machines.
Check out the PDP-9 and PDP-18. The latter was a UNIBUS based 18 bit machine
with fairly nice support for both two's complement arithmetic and one's
complement arithmetic (don't ask why, I don't know). PDP-18 machines were
fairly widely used for real-time control in some areas. Back in 1974, I had
a friend who used one regularly, I've never had the pleasure.
Doug Jones
jones@herky.cs.uiowa.edu
sandee@sun16.scri.fsu.edu (Daan Sandee) (10/09/90)
In article <14900017@hpdmd48.boi.hp.com> sritacco@hpdmd48.boi.hp.com (Steve Ritacco) writes: >It is my understanding that DEC actually made a few 18 bit word machines. >It was the PDP-? (I can't remember)! >The reason I know this is that triple-i (a company that makes typesetters) >latched on to the architecture for their equipment, and as far as I know >still manufacture a clone of it today (pitiful no?). PDP-9. I did a lot of assembly code programming on it in 1971. 18-bit word, 15-bit addresses, allowing for 32K words of directly addressable memory. The oddest thing I remember about it is how it used ASCII. Apart from its own 6-bit character set, it also stored 7-bit ASCII characters, five characters in 2 consecutive words, with one bit unused. Called "5/7 ASCII". (ASCII had only codes 0-127 defined. The 8th bit was used by I/O devices for parity). Daan Sandee sandee@sun16.scri.fsu.edu Supercomputer Computations Research Institute Florida State University, Tallahassee, FL 32306-4052 (904) 644-7045
simon@fuquad.ccur.com (Simon Rosenthal) (10/09/90)
In article <12894@encore.Encore.COM> jkenton@pinocchio.encore.com (Jeff Kenton) writes: >From article <14900017@hpdmd48.boi.hp.com>, by sritacco@hpdmd48.boi.hp.com (Steve Ritacco): >> It is my understanding that DEC actually made a few 18 bit word machines. >> It was the PDP-? (I can't remember)! > >The PDP-1 had 18 bits. Nice machine. > Not to mention its descendants, the PDP-9 and PDP-15. I remember there being a whole roomful of these in use at Massachusetts General Hospital (running MUMPS), as late as 1983/4. - Simon _______________________________________________________________________________ Simon Rosenthal: ___________ Concurrent Computer Corporation / _________/_ Westford, MA 01886 /_/________/ / Internet: simon@westford.ccur.com
jkenton@pinocchio.encore.com (Jeff Kenton) (10/09/90)
From article <61237@fuquad.ccur.com>, by simon@fuquad.ccur.com (Simon Rosenthal): >> >>The PDP-1 had 18 bits. Nice machine. >> > Not to mention its descendants, the PDP-9 and PDP-15. I remember there being > a whole roomful of these in use at Massachusetts General Hospital (running > MUMPS), as late as 1983/4. > And the PDP-4 and PDP-7 in between. ----- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ----- ----- jeff kenton: consulting at jkenton@pinocchio.encore.com ----- ----- always at (617) 894-4508 ----- ----- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -----
kym@bingvaxu.cc.binghamton.edu (R. Kym Horsell) (10/10/90)
In article <14900017@hpdmd48.boi.hp.com> sritacco@hpdmd48.boi.hp.com (Steve Ritacco) writes: >It is my understanding that DEC actually made a few 18 bit word machines. >It was the PDP-? (I can't remember)! \\\ PDP-9 & PDP-15. I used them for a while. -Kym Horsell
fwebb@bbn.com (Fred Webb) (10/10/90)
In article <14900017@hpdmd48.boi.hp.com> sritacco@hpdmd48.boi.hp.com (Steve Ritacco) writes: >It is my understanding that DEC actually made a few 18 bit word machines. >It was the PDP-? (I can't remember)! The following DEC machines were 18 bits: PDP-1 PDP-4 PDP-7 PDP-9 PDP-15 The following were all 12 bits: PDP-5 PDP-8 PDP-12 Note that these numbers are both divisors of 36 bits, which was the "standard" IBM word size of the time (701/704/709/7090/7094). Until the PDP-8, essentially all of DEC's machines were 18 bits (the PDP-5 wasn't a very large seller). -- Fred
esmith@goofy.apple.com (Eric Smith) (10/10/90)
In article <2018@inducom.UUCP> meindert@inducom.UUCP (Meindert Kuipers) writes: > Although not odd word size, my good old HP-41 calculator has a little > CPU with 12-bit instructions. The memory space for these "assembly" > programs is completely 12 bits. The data space however has words of > 56 bits, allowing 7 bytes of data, or 10 digit precision (with exponent). Actually, it has 10-bit instructions and ROM. The instruction and data space are almost completely separate, with the exception of an instruction to allowing reading ROM words as data. That instruction was the primary architectural advance of the HP-41's processor compared to their older ones. Do HP's new calculators use the same principle? I have understood that the 10-series (HP12, HP16) use the same processor. The 10-series apparently used the same architecture, possibly with minor changes (I'm not sure). The new machines use the same basic architecture as the Saturn processor that was used in the HP-71B. Saturn had a single address space using nibble (4-bit) addressing and 5-nibble (20-bit) addressing, for a total 1 Meganibble (512K byte) address space. The processors in the current HP calculator line use essentially the same architecture with a few additional instructions. The following is a (possibly incomplete) list of the HP products that use this architecture, organized by category: Handheld computers: HP-71B Business calculators: HP-10B, HP-14B, HP-17B, HP-17BII, HP-18C, HP-19B, HP-19BII Scientific calculators: HP-20S, HP-21S, HP-22S, HP-27S, HP-28C, HP-28S, HP-32S, HP-42S, HP-48SX Eric -- Eric L. Smith Opinions expressed herein do not necessarily reflect those esmith@apple.com of my employer, friends, family, computer, or even me! :-)
emv@math.lsa.umich.edu (Edward Vielmetti) (10/10/90)
In article <509@ssp9.idca.tds.philips.nl> dolf@idca.tds.PHILIPS.nl (Dolf Grunbauer) writes:
started. So I am only expecting trouble with VAX when it is going to
implement user programmable interfaces in their vacuum cleaners or when
DEC builds vacuum cleaners into their machines.
When I used to care for a powerful Vax 11/750 the periodic maintenance
schedule involved vacuuming out the dust bunnies from the system. I
don't think the field engineer used a VAX tho.
Followups to alt.folklore.computers.
--Ed
Edward Vielmetti, U of Michigan math dept <emv@math.lsa.umich.edu>
dhoyt@vx.acs.umn.edu (10/10/90)
>In article <14900017@hpdmd48.boi.hp.com> sritacco@hpdmd48.boi.hp.com (Steve Ritacco) writes:
Didn't the DEC Systems 10 & 20 allow the program to change the size of
characters? Five bits gave you radix 50 (used by the file system). Nine was
great as it gave and even number of characters per word with no extra bits as
well as giving you meta characters. Eight and seven were good for Lisp
programming as it gave you extra tag bits in a word. I know we did this all in
software, but I can't remember if it was an 'offical' system option or not.
david
kym@bingvaxu.cc.binghamton.edu (R. Kym Horsell) (10/10/90)
In article <2383@ux.acs.umn.edu> dhoyt@vx.acs.umn.edu writes: >>In article <14900017@hpdmd48.boi.hp.com> sritacco@hpdmd48.boi.hp.com (Steve Ritacco) writes: > Didn't the DEC Systems 10 & 20 allow the program to change the size of >characters? Five bits gave you radix 50 (used by the file system). Nine was >great as it gave and even number of characters per word with no extra bits as >well as giving you meta characters. Eight and seven were good for Lisp >programming as it gave you extra tag bits in a word. I know we did this all in >software, but I can't remember if it was an 'offical' system option or not. The DECSystem-10 & 20 had general byte manipulation instructions; bytes weren't actually defined to be any particular size by the h/w. Unlike the VAX line, the 10/20's had the concept of a one-word "byte pointer" that specified a field of up to 36 bits somewhere in memory (perhaps even in the same bytepointer)! Byte pointers contained a byte size (6 bits) , offset (6 bits), index register (sic), displacement and indirect bit -- you effectively got the word from which the field was extracted by performing the whole general addressing stuff (e.g. if the indirect bit in the bytepointer was set you could end up performing an indefinite number of mem reads since each of _these_ could indirect). There were various insructions for manipulating bytepointers; load byte/deposit byte from mem to/from reg, increment byte pointer (by one position), increment&load, increment&deposit, and on the KL10 & 20's various more general byte pointer adjustments and restrictions removed (e.g. bytes across word bdys). (I only played low-level on a KI10 so I'm a bit unsure as to the details of these latter). Another use of 7-bit bytes in a 36-bit word in _text_ files involved using the left-over bit (which occured in the lsb) as a line/page marker word -- if an editor, compiler or the lineprinter software came across a word that was otherwise supposed to be text & found the lsb set, it interpreted the word as a page/line number pair -- typically expanding same into the appropriate decimal string. -Kym Horsell
zombie@uucp (Mike York) (10/10/90)
In article <61237@fuquad.ccur.com> simon@fuquad.UUCP (Simon Rosenthal) writes: >Not to mention its descendants, the PDP-9 and PDP-15. I remember there being >a whole roomful of these in use at Massachusetts General Hospital (running >MUMPS), as late as 1983/4. We retired our 14 PDP-15's in December 1988. They served us well for 10 years. We used them for an interactive graphics application to prepare illustrations for airplane maintenance, repair and operation manuals. We started looking for a replacement platform in 1983, but didn't find anything that could pick as fast us the PDP-15's (which had vector refresh monitors) until SGI came out with the 2400T. We now have 120 SGI 4D/25's. As much as I like 'em, there are still times that I miss programming those 15's in assembler ;^). -- Mike York | "Lord help me, I'm just not Boeing Computer Services | that bright." (206) 865-6577 | zombie@voodoo.com | -Homer Simpson
sxr@cs.purdue.EDU (Saul Rosen) (10/10/90)
In article <4155@bingvaxu.cc.binghamton.edu> kym@bingvaxu.cc.binghamton.edu.cc.binghamton.edu (R. Kym Horsell) writes: >>It is my understanding that DEC actually made a few 18 bit word machines. >>It was the PDP-? (I can't remember)! > >PDP-9 & PDP-15. I used them for a while. There have been a number of postings here and in alt.folklore.computers in which the posters ask questions and make comments about various generations of DEC computers. The following is a comment that I recently posted to alt.folklore.computers: If you are interested in the history of all of the DEC machines, from the beginning of DEC through most of the 1970's, you will find it very interesting to read the book "Computer Engineering, A DEC View of Hardware Systems Design" edited by C. Gordon Bell, J. Craig Mudge, and John E. McNamara. It was published by Digital Press in 1978. The title makes the book sound very technical, and the book does contain a great deal of technical detail. It also contains a great deal of general historical information about the various DEC series of computers. It can be read with interest by computer people who do not consider themselves to be engineers. It answers all of the questions relative to DEC computers that I have seen posted to this newsgroup. Saul Rosen
esmith@goofy.apple.com (Eric Smith) (10/11/90)
In article <2383@ux.acs.umn.edu> dhoyt@vx.acs.umn.edu writes (talking about the DECsystem-10): > characters? Five bits gave you radix 50 (used by the file system). Nine was Sorry, five gave you radix 32. The file system used six bits per character, or radix 64. -- Eric L. Smith Opinions expressed herein do not necessarily reflect those esmith@apple.com of my employer, friends, family, computer, or even me! :-)
jkenton@pinocchio.encore.com (Jeff Kenton) (10/11/90)
From article <ESMITH.90Oct10173750@goofy.apple.com>, by esmith@goofy.apple.com (Eric Smith): > >> Five bits gave you radix 50 (used by the file system). > > Sorry, five gave you radix 32. The file system used six bits per character, > or radix 64. > -- Radix 50 (50 is octal == 40 decimal) gave 3 characters per 16 bit word. There were 39 encodable characters: A-Z, 0-9, and "$", "%", ".". Each character had a value from 1 to 39 (in the order shown). A 16 bit value was created by calculating 1600*<first char> + 40*<second char> + <third char>, giving slighly fewer than 64000 possibilities. The six bit per character encoding scheme was known as SIXBIT (surprise!). ----- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ----- ----- jeff kenton: consulting at jkenton@pinocchio.encore.com ----- ----- always at (617) 894-4508 ----- ----- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -----
esmith@goofy.apple.com (Eric Smith) (10/12/90)
In article <12916@encore.Encore.COM> jkenton@pinocchio.encore.com (Jeff Kenton) writes: > From article <ESMITH.90Oct10173750@goofy.apple.com>, by esmith@goofy.apple.com (Eric Smith): >> >>> Five bits gave you radix 50 (used by the file system). >> >> Sorry, five gave you radix 32. The file system used six bits per character, >> or radix 64. >> -- >Radix 50 (50 is octal == 40 decimal) gave 3 characters per 16 bit word. There >were 39 encodable characters: A-Z, 0-9, and "$", "%", ".". Each character >had a value from 1 to 39 (in the order shown). A 16 bit value was created >by calculating 1600*<first char> + 40*<second char> + <third char>, giving >slighly fewer than 64000 possibilities. Yes, but the original posting said 5 bits/character = radix 50. That is still incorrect. Radix 50 is actually a little less than 5+1/3 bits per character. I stand by my assertion that 5 bits per character is radix 32 (decimal). Eric -- Eric L. Smith Opinions expressed herein do not necessarily reflect those esmith@apple.com of my employer, friends, family, computer, or even me! :-)
als@bohra.cpg.oz (Anthony Shipman) (10/12/90)
In article <ESMITH.90Oct10173750@goofy.apple.com>, esmith@goofy.apple.com (Eric Smith) writes: > In article <2383@ux.acs.umn.edu> dhoyt@vx.acs.umn.edu writes (talking about > the DECsystem-10): > > > characters? Five bits gave you radix 50 (used by the file system). Nine was > > Sorry, five gave you radix 32. The file system used six bits per character, > or radix 64. > -- > Eric L. Smith Opinions expressed herein do not necessarily reflect those > esmith@apple.com of my employer, friends, family, computer, or even me! :-) The "50" is in octal. The actual radix was 40 decimal. The code was used mainly in symbol tables in the assembler (and linker). The legal symbol characters (0-9, A-Z, '.', "something", "something" and NUL) were mapped to the values 0-39. The 6 chars of the symbol were then combined in the usual way sum (Di x 40^i), i=0..5 The resulting value could fit in 32 bits leaving 4 for symbol flags like local/global etc. -- Anthony Shipman ACSnet: als@bohra.cpg.oz.au Computer Power Group 9th Flr, 616 St. Kilda Rd., St. Kilda, Melbourne, Australia D
tj@Alliant.COM (Tom Jaskiewicz) (10/12/90)
From article <2721@crdos1.crd.ge.COM>, by davidsen@crdos1.crd.ge.COM (Wm E Davidsen Jr): > . . . > many minis were made with 16 or 18 bit words, so did anyone ever build a > 9-bit byte system? Yes. I recall such a system on a Honeywell mainframe that I had the misfortune to use back in 1974. I don't recall the model number of the machine (I've been trying to forget it for years). It had 36 bit words and packed four 9-bit characters to a word. In Fortran, you could read into an integer using A4 format. -- ########################################################################## # The doctrine of nonresistance against arbitrary power, and oppression is # absurd, slavish, and destructive of the good and happiness of mankind. # -- Article 10, Part First, Constitution of New Hampshire
richard@aiai.ed.ac.uk (Richard Tobin) (10/12/90)
In article <ESMITH.90Oct10173750@goofy.apple.com> esmith@goofy.apple.com (Eric Smith) writes: >> Five bits gave you radix 50 (used by the file system). >Sorry, five gave you radix 32. The file system used six bits per character, >or radix 64. Radix 50 was a way of packing six characters into a 32-bit number. There were 40 characters (50 octal = 40), thus using 64000^2 of the 65536^2 possible 32-bit values. The characters were nul 0-9 A-Z . $ % in that order, and were encoded simply by multiply by powers of 40. -- Richard -- Richard Tobin, JANET: R.Tobin@uk.ac.ed AI Applications Institute, ARPA: R.Tobin%uk.ac.ed@nsfnet-relay.ac.uk Edinburgh University. UUCP: ...!ukc!ed.ac.uk!R.Tobin