On this side of the Atlantic it was always '10, 11, 0...' not '12, 11, 0...' I've no idea why.
John -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Clark Kidd Sent: 28 May 2009 20:35 To: [email protected] Subject: Punched Card Combinations (WAS Book on Poughkeepsie) It was not all that difficult to look at the holes on a punched card and figure out what each column represented. If you have an older "System/360-370 Reference Summary Card" (the old green or yellow multi-folded cards), there was a section called CODE TRANSLATION TABLE, and a column titled "EBCDIC Card Code" that contains the card columns that would be punched out to generate the specific values from X'00" to X'FF'. My handy yellow card is dated March, 1974 and it contains this column. For example, the letter "A" or X'C1' could be represented with a 12-1 punch. As I remember, there were three "control" rows at the top of the card (12, 11, 0) and then 9 "data" rows (1-9) under those. So each possible column would contain up to 12 rows that could be punched: Col1 Col2 ... Col80 12 12 12 11 11 11 0 0 0 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 6 6 6 7 7 7 8 8 8 9 9 9 For example, a column representing an "A" would have the first of the control rows punched out (12), and the first of the data rows (1) punched out. Some of the more exotic hex combinations required quite a number of punches. For example, X'FF' was represented as 12-11-0-7-8-9. This is why a keypunch machine would be much louder when you were duplicating an object deck. If a particular column didn't get punched correctly, you could always "patch" the card by putting it back in the keypunch machine and re-punching that one column. But how about if there were too MANY holes? I remember they made some small silver adhesive squares that could be placed over a hole so that it wouldn't be read. I have seen some very large programs (object decks of 1000+ cards) that were patched this way rather than recompiling the program and wasting CPU time and punched cards. Of course, it was usually a simple change such as changing a constant or correcting a bad branch. There were REAL system programmers even in the days of punched cards... Clark ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
