I'm really old. During 1973-5, I worked at Hewlett Packard's Advanced Product Division in Cupertino. I programmed the HP-21 and HP-25 Pocket Scientific Calcula

I'm really old. During 1973-5, I worked at Hewlett Packard's Advanced Product Division in Cupertino. I programmed the HP-21 and HP-25 Pocket Scientific Calculators. The chip architecture was bit serial with 14 decimal digits (56 bits). The HP-21 had 1024 by 10 bits of ROM, the HP-25 had 2046 by 10 bits. The HP-25 was user programmable, storing keystrokes, some strokes merged. There were 49 locations. The original assembly microcode was on punched cards, and debugged with an hardware emulator attached to an HP-2100 mini computer. I still have the paper listings. Some years ago, I gave copies to Eric Smith, who scanned, OCR'ed, and very carefully proof read and debugged the result. The listing for the HP-25 is at http://www.jacques-laporte.org/Woodstock/images/hp25.txt http://www.jacques-laporte.org/Woodstock/images/hp25.txt Here is a scan of a handwritten program for the MIT Whirlwind, from 1953 (from Don Knuth's collection): http://archive.computerhistory.org/resources/text/Knuth_Don_X4100/PDF_index/k-3-pdf/k-3-u2774-Whirlwind-program.pdf http://archive.computerhistory.org/resources/text/Knuth_Don_...