5 Jul
2010
5 Jul
'10
9:19 p.m.
Fred Cisin wrote:
On Sun, 4 Jul 2010, Charlie Carothers wrote:
Yep. I
realized after I clicked send I should have probably said "bit
of logic, if needed" as I thought possibly only a cable might be needed.
If it were me though, I think I would tend to
just grab an old PC with a
parallel port and arrange a bit of logic to make that talk to the
Centronics output of the classic computer.
That "logic" consists of a bidirectional printer port (See Bruce Eckel's
Interfacing book), and a cable that you WILL have to make yourself.
If it IS an old PC (less than about a dozen MHz), then you will need to
add some trivial circuitry to extend the pulse, otherwise your polling
loops will be too slow. If the alien machine is like MOST, then it cheats
on the spec, and the pulse is WAY too short to use interrupts.
Yeah, I was thinking later that
the dot-matrix impact printers that I
mostly visualize originally being used with these machines could do ISTR
Epson emulation as one of their interface options (either that or they
actually *were* Epsons). So maybe you only need to recognize the Epson
escape sequences as long as you can convince the classic it is talking
to an Epson or emulated equivalent. Of course I'm sure there are
several exceptions where that idea won't fly. Seems like there are
always exceptions...
Later,
Charlie C.
The software (under DOS yet)
should be pretty much a piece of cake with that arrangement. Just
mostly accept input up to CR, CRLF or whatever and write the text lines
to a file. I suppose handling escape sequences might not be quite so
much fun. It sort of depends on the capabilities of the target physical
printer and just how close to the classic's printout look one was
willing to support.
Even the escape sequences are not hard to figure out. Many printers are
well documented. With a little trivial study, you can do post-processing.
It does NOT have to be done during the capture, IFF you are writing to a
file. If you choose to not write to a file, and use the PC as a print
buffer, then you will need to do ad-hoc processing.
If you do a good job of it, then you won't need to start from scratch NEXT
time.
--
Grumpy Ol' Fred cisin at xenosoft.com