On Tue, 15 Mar 2005 07:21:42 +0100, woodelf <bfranchuk at jetnet.ab.ca> wrote:
William Donzelli wrote:
Well the
1620 was a variable length machine ... A sign/flag bit made
more sence at the time since
you only had as many BCD digits as you needed.
It is still very inefficient, with lots of wasted bits. It would not
matter with a small machine like a 1620, but it does when the system
gets
larger. Even a small S/360 dwarfs a 1620. All those wasted bits add up.
The 1620 is BCD serial ... slow but then lots less $$$ than a 360.
Since I only got to know small computers like a PDP 8 and a IBM1130
I never had to deal with the bigger stuff.
The 1620 was one of the first computers I used, but only at high-level
(FORTRAN II).
Compiling was interesting - put the source cards in the reader, punch out
an intermediate deck, then pass both through for the second pass. I used
it for my physics lab tests, plotted nice graphs on the Calcomp drum
plotter. We also had a line printer making 50 lines per minute. I never
saw one like it, characters were on quarter-circle segments which rotated
back and forth.
The Univac 1107 was a more impressive beast, with both a drum storage and
a drum printer.
One of my major goofs was when I first used the drum, not knowing that it
did not write any end of file marker.
Computer time was valuable in the sixties, and my program was reading from
the drum for 20 minutes before the operator terminated it.
My favourite old big iron is the NCR 315. I worked for NCR Norway, who
hade one in their data processing center. The 315 was, of course, a BCD
system. It had 12-bit bytes (which were calles slabs, since the byte term
was not established when it was new), and 24-bit words. I just found a
couple of old pictures of the operator looking helplessly at a mess of
magnetic tape. The really cool thing was that most of the input came from
OCR - cash register journals written with National Optical Font. Scanned
by a rotating drum with a tiny diamond-shaped opening and a
photomultiplier inside.
--
Bj?rn