Why does Altos UNIX V "lprint" print so S L O O O W?

[Bill Vermillion]

In article <1991Mar21.033308.21226 at NCoast.ORG> allbery at ncoast.ORG (Brandon S. Allbery KB8JRR) writes:
>As quoted from <1143 at twg.bc.ca> by bill at twg.bc.ca (Bill Irwin):
>+---------------
>| Altos has added an extension/enhancement to the lp spooling
>| system called "lprint", which is supposed to use the TERM setting
>| to look up the entry in /etc/termcap and find the start/stop
>| transparent printing codes to send to the terminal in order to
>| print the file.  This is all working (after turning on onlcr in
>| stty to get the file to print on more than one line 8^( ) except
>| that the printer is running very slowly.  I mean, print a line -
>| count one, two - print another line - count one, two.
>+---------------
>

I had a similar problem under SCO Xenix.   The terminal was a Wyse
60, and the program OSAS was written in Business Basic and used
their way of handling printing (which I hate!).

The Wyse 60 used a parallel printer port and the printer was an
Okidata - in the 320 or 390 series (they had several of each).

You could watch the status line togges the xprt on and off,  and
printing had a puase of 1 to 2 seconds between each line.

Changed the terminal to a Wyse 60 with a serial port and put a
ser/par convertor on, and it ran at proper speeds.  Putting a
par/par buffer on the Wyse 60 with the parallel printer port did
NOT change anything.

Never took the time to actually find out what handshaking was being
done to cause the problem as moving a serial printer port Wyse into
place was the quickest/easiest solution.   I have seen slow
parallel printing before and it appeared to be dependant on the
printer handshake.  One printer would be painfully slow, and swap
another brand in and speed was back.

-- 
Bill Vermillion - UUCP: uunet!tarpit!bilver!bill
                      : bill at bilver.UUCP