Exactly. Luckily the designer of the other end of the link was aware of the
stringent timing and used a decent XTAL OSC that has hardly aged a day in
almost 30 years :-)

The 30-50ppm error I can hope for from a crystal works out fine (somewhere
about .01 bit time over the 240 bits (?)).

I'm not familiar with the DOS 6.22 fast-running-clock scenario. Any idea
what caused it?

The official (original PC) clock into the timer chip is:  1,193,181.6667 Hz
(from IBM docs believe it or not).

On division by 65536 we get :

PC tick rate is 18.206507... (more or less :-) )

Bob

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]: PIC only [EE]: engineering [OT]: off topic [AD]: advertisements