> -----Original Message-----
> From: Don Maslin [SMTP:donm_at_cts.com]
> Sent: Thursday, May 27, 1999 6:47 PM
> To: Discussion re-collecting of classic computers
> Subject: RE: What's a "computer console" selectric called?
>
> On Thu, 27 May 1999, Arlen Michaels wrote:
>
<snip>
> > Actually, the biggest challenge in interfacing this thing to a computer
> was
> > to sort out how to read one particular status signal from one of the
> > microswitch contacts in the print mechanism, so your computer could
> start
> > sending the next character at just the right moment before the
> mechanical
> > cycle completely finished. Else your software had to pause a few
>
> Didn't you also have to feed it EBCDIC instead of ASCII in order for it to
> 'understand' what you wanted printed?
>
> - don
>
<snip>
I don't think the Model 735 Selectric could even handle EBCDIC directly. I
seem to recall the electrical interface was defined as tilt-and-rotate
signal names. My Selectric terminal certainly didn't do any translation by
itself from character-codes to solenoid signals, at least not from ASCII. I
had to do translation myself before sending to the printer. One way would
have been with hardware between the computer and the Selectric, eg- using an
eprom to translate each ASCII code into the correct combination of Selectric
tilt-and-rotate signals. My lazier way was to simply put a look-up table in
my driver code, to intercept each ASCII character enroute to the printer and
translate it into the appropriate pattern of solenoid signals first.
Imagine if you had to drive a dot-matrix print head with raw pin-driver
signals instead of the printer hardware figuring it out for you : same kind
of problem.
Some vendors did indeed supply an interface that took ASCII from the
computer and sent the necessary tilt-and-rotate signals out to the
Selectric.
Arlen
--
Arlen Michaels amichael_at_nortelnetworks.com
Nortel Networks, Ottawa, Canada
Received on Fri May 28 1999 - 10:37:02 BST