Huw Davies wrote:
Ah, the good old 2741. We had one here at La Trobe
which had an 8080 based
interface so that it could talk (current loop?) to our DECsystem-10 at
134.7 baud. I understand the 134.7 baud rate is the fastest speed that you
can drive the Selectric mechanism before it flies apart :-)
Did my honors thesis on this back in 1977 using a typesetting program known
as Cicero running on the -10 written at ARL (Aeronautical Research Labs, in
Melbourne Australia). Used it again for version 1 of my M.Sc - it could do
golf ball changes to allow the pages of mathematical formulae to be
printed. From memory, a typical page took 10 minutes to print due to the
time wasted in golf ball changes. I keep trying to remember this when I
complain that the Xerox printer I now use prints about 10 double sided
pages a minute...
Be glad you weren't doing APL, which I have seen type balls for. As I
recall, it took _two_ typeballs to do APL properly, so between changing
balls and overstriking, the language fell out of popular favor. These
days, with bit-mapped displays and printers, the language should be
_blossoming_ except for all of the crap about "object-oriented" or
otherwise feeble languages.
--
Ward Griffiths
They say that politics makes strange bedfellows.
Of course, the main reason they cuddle up is to screw somebody else.
Michael Flynn, _Rogue Star_