Image Map Image Map
Page 3 of 3 FirstFirst 123
Results 21 to 29 of 29

Thread: Which machines did not have LOWER CASE?

  1. #21
    Join Date
    Mar 2013
    Location
    SW Quadrant of Michigan
    Posts
    289

    Default

    From what I can tell, the International Telegraph and Telephone Consultative Committee (CCITT) [which was the precursor organization to the International Telegraph Union Telecommunication Standardization Sector {ITU-T}], International Organization for Standardization (ISO), & the United States of American Standards Institute (USASI) [which was the precursor organization to the American National Standards Institute {ANSI}, as well as the successor organization to the American Standards Association {ASA}] all voted and agreed to adopt the addition of the lowercase alphabet into various domestic and international standards of character encoding and communication in 1963. This resulted in an update of ASCII as standard ASA X3.4-1963, which only contained the uppercase alphabet, to ASCII standard USAS X3.4-1967, which was the very first ASCII standard approved and published that included character encoding for the lowercase alphabet.

    ASCII standard USAS X3.4-1967 was the starting point for all the future character encodings which we love and are familiar with: CCSID/Code Page 437, ISO/IEC 8859-1, Windows/CP-1252, & Unicode.

    Interesting, after the the various international and domestic organizations mentioned above decided to adopt character encodings for the lowercase alphabet in 1963, IBM started work on the other well known and used character encoding standard Extended Binary Coded Decimal Interchange Code (EBCDIC), based on IBM's implementation of Binary Coded Decimal Interchange Code (BCDIC).

    As such, I guess one can confidently say before 1963, everything was likely uppercase only, and after 1963, it depends on what character encodings were implemented.

  2. #22
    Join Date
    May 2009
    Location
    Connecticut
    Posts
    4,840
    Blog Entries
    1

    Default

    The manual for the IBM 1401 indicates that the 1407 console can enter lower case characters and the 1403 printer can print lower case characters. That was from 1959 so the first computer with lower case I have been able to find.

    See page 100 of the 1401 Operator's Guide.

    There was also widespread enough support for lower case before the standards came about for RUNOFF to be viable in 1964. The standard committees just seemed to be grabbing onto what the manufacturers were doing.

  3. #23
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    35,369
    Blog Entries
    18

    Default

    Nope. The 1950s Friden Flexowriter was used on quite a number of machines (e.g. LGP-30) and had upper- and lower-case models. Techref.

    What systems took advantage of that, I'm not clear about.

    The IBM 7030 production dates from 1961 and as I noted, had lowercase, depending on the print train used. It wasn't EBCDIC or ASCII.

    UNIVAC systems using Fieldata encoding used control codes to switch between UC and LC. I'm not sure which early UNIVAC systems actually used the encodings, but it's there.

    Until about the S/360 architecture, the character set for a computer was mostly a side issue. That is, it wasn't hard-coded into the instruction set. S/360 introduced instructions that used EBCDIC or ASCII (e.g. ED and EDMK) (depending on the setting of bit 12 in the PSW.

    The issue for older machines using 6-bit characters was how to represent more than 64 characters. So, there was no one-for-one representation in 6-bit ASCII or BCDIC. Thus, one could shift (a la 5-level teletype code and the DEC PDP8-based WIPS) or use more than 6 bits (CDC 6000/Cyber extensions to display code using, e.g. octal 01 = "A"; octal 7601 = "a".
    Last edited by Chuck(G); October 23rd, 2020 at 08:56 PM.

  4. #24

    Default

    Quote Originally Posted by vwestlife View Post
    It's in the September 1985 issue of Stereo Review magazine, pages 80 and 81 of the PDF:

    https://worldradiohistory.com/Archiv...ew-1985-09.pdf

    And don't miss the ad for the rare SVI-328 computer on 88-89, and the Olivetti "spark-jet" printer on 112-113.
    I wonder how much the Stereo Review readers enjoyed having almost 80 pages of their issue consumed by a DAK catalog.

    Boy, those were the days.

  5. #25
    Join Date
    Sep 2006
    Location
    Silicon Valley
    Posts
    2,338

    Default

    Quote Originally Posted by Paralel View Post
    Interesting, after the the various international and domestic organizations mentioned above decided to adopt character encodings for the lowercase alphabet in 1963, IBM started work on the other well known and used character encoding standard Extended Binary Coded Decimal Interchange Code (EBCDIC), based on IBM's implementation of Binary Coded Decimal Interchange Code (BCDIC).
    "interesting" isn't how I would put it

    Bob Bemer fought IBM for ASCII and Big Blue gave him the finger over collation sequencing.

    the ASCII bit in S360 is vestigial.

    and if you don't know who Bob Bemer was
    http://www.cnn.com/TECH/computing/99...idg/index.html

    bobbemer.com has bitrotted, you have to go to the wayback machine now to see his site.
    CHM has his papers https://www.computerhistory.org/coll...alog/102785380
    The interesting bit starts at page 300
    Last edited by Al Kossow; October 24th, 2020 at 03:48 AM.

  6. #26
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    35,369
    Blog Entries
    18

    Default

    The thing (way back when) that tripped me up again and again, was the non-consecutive sequences in EBCDIC (and BCDIC) alphabetics. You have to remember that "J" is not "I"+1, nor is "S" "R"+1 -- there are other characters stuck in those gaps. On other internal codes, such as CDC Display code, alphabetics are consecutive.

    Of course, BCDIC and EBCDIC are based on punched card codes, which are essentially decimal. so, "I" is 12-9 and "J" is 11-1. (The 0 row is essentially a zone punch in alphabetics). You can see this reflected in the PDP-8 instruction 6632, "Read card alphabetic" which takes a card column and returns it as a 6-bit "zoned" representation.

    ASCII is not as easily translated to punched card codes (or to 7 track BCD tape codes). But even this isn't straightforward. In BCDIC, binary 000000 is a space on some systems or an 8-2 punch on others--but not all: 8-2 is a record mark on the 1401 and 1620. Since you can't have all tracks zero on an even-parity 7 track tape, that's also encoded as 001010 instead of 000000. The history of IBM BCDIC encodings is complex and, as far as I can determine, no fewer than 4 distinct versions were in use at the same time. So, in a sense, EBCDIC straightened things out a bit, while retaining some of the warts of the old code and not going whole-hog to ASCII (USASCII as it was termed at the time).

    The notion that EBCDIC eliminated BCDIC for IBM after 1965 is a bit misleading. The IBM System/3 (1969) used BCD for its 96-column cards.

  7. #27
    Join Date
    Mar 2013
    Location
    SW Quadrant of Michigan
    Posts
    289

    Default

    Quote Originally Posted by Al Kossow View Post
    "interesting" isn't how I would put it

    Bob Bemer fought IBM for ASCII and Big Blue gave him the finger over collation sequencing.

    the ASCII bit in S360 is vestigial.

    and if you don't know who Bob Bemer was
    http://www.cnn.com/TECH/computing/99...idg/index.html

    bobbemer.com has bitrotted, you have to go to the wayback machine now to see his site.
    CHM has his papers https://www.computerhistory.org/coll...alog/102785380
    The interesting bit starts at page 300
    That seems about right for IBM.

    Not terribly different from their decision to use CCSID 437 for the IBM PC MDA ROM font, making it an "Extended" ASCII, but not exactly ASCII compatible.

    (Although, I will admit, Code Page 437 is still useful even in this day and age to quickly pop in a non-ASCII keyboard standard glyph on a windows machine. I use it rather frequently. ☺)
    Last edited by Paralel; October 24th, 2020 at 10:23 AM.

  8. #28
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    35,369
    Blog Entries
    18

    Default

    Quote Originally Posted by Al Kossow View Post
    the ASCII bit in S360 is vestigial.
    Are you saying that if the ASCII bit is set in the PSW, EDIT doesn't unpack packed BCD to its ASCII (not EBCDIC) version?

    I've never gotten to run on a S/360 box with the ASCII bit set, so I can't say. The CDC STAR-100 certainly observed the distinction--tested it myself out of curiosity ( http://bitsavers.org/pdf/cdc/cyber/c...00hw_Dec75.pdf PDF page 284 et. seq.).

  9. #29
    Join Date
    Mar 2013
    Location
    SW Quadrant of Michigan
    Posts
    289

    Default

    That is a rather interesting question. Now I wonder if that function does indeed work correctly.

    Anyone have an S/360 up and running to test it out?

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •