>> Unless you are using an unusually fast and non-standard baud rate,
>> the printer port can easily overrun the RS-232 data rate by an
>> order of magnitude or two.
>
>Yes, but (most?  All?) parallel ports tend to implement a form of
>flow control on a per-byte basis, so that they can wait for printers
>to do very slow things like print head or carriage movement.  The
>algorithm usually goes something like:

My experience with centronics interface printers of recent origin is that
many machines that send data to them do not worry about the handshake from
the printer. They just put the data out and pulse the strobe line, and then
change the data and give another pulse on the strobe line. On machines like
older PC's the whole thing is done in software, and the time taken for the
software loop is sufficient for the printer to get the data from its input
buffer into an internal fifo. The computer keeps squirting data until the
end of line (at which point it knows the printer needs to do something) and
then checks the busy flag until the printer is ready for more data.

For your purpose you may be able to do something similar, but should allow
for the possibility that the handshake in hardware is not properly dealt
with by the host.

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist