Postscript printer on centronics interfaces [summary]

From: Karl-Jose Filler (pla_jfi@pkinbg.UUCP)
Date: Fri Jan 24 1992 - 04:43:07 CST

My original Question was :

I am trying to use the sun transcript software ( links sun's to
postscript printers via V.24 terminal lines ) with a parallel
S-Bus card.
The S-Bus Card + Printer is working, if you do an
        cp file /dev/pp0
But I cannot get it working with spooling.
Both with the PS Communication filter and without the printer
gets no PS job.
The current printcap entry :
[Printcap entry deleted]
The Sbus Card is :
        Antares Parallel Port
The Printer:
        HP LaserJet III Si + Postscript

        Thank you in advance
                                Karl-Jose Filler
There have been lots of answers, which fall in three categories :
1. Postscript needs a bidirectional port and most of the
   printers / interfaces don't support this ( at least not the III Si )
2. How to establish a V.24 communication to the printer.
3. How to install a printer, and to restart the spooler

All answers, which fall in 1. are summarized below :

                Karl-Jose Filler

Subject: Re: Postscript printer on centronics ports

We've been running generic Adobe transcript modified to work
with a parallel printer, but I assume the problem's the same...

The Centronics interface supports only one-way data communication, but
pscomm gets job completion info by querying the printer and reading
the response from the port. That's how it gets accounting information
and syncs itself with the printer's state.

Most (?) PS printers that have parallel ports use a serial port as a
back channel. pscomm must be modified to open the serial port and
do all reads from it. Sun may (should) support this already.
From: (Mike Heisler)
Subject: Parellel Postscript
Date: Wed, 22 Jan 92 9:16:01 EST

Unless you have changed it the Transcript software expects a response
from the PS printer, PostScript is inherently bidirectional, the parrellel
port is not. In theory you should see something to that affect in the
error log but sometimes Transcript doesn't deal with the errors like it
should. I would look at the code and set debugging on.

Mike Heisler		607-255-7344
Materials Science Ctr.	302 Thurston Hall	Cornell Univ, Ithaca, NY 14853
From: rhoward%matd%gatech@unido.uucp
Subject: Re: Using Sun Transscript Software with parallel Interfaces

You can't make Transcript work with a parallel port. Period. The software assumes a serial port and does ioctl()'s based on that. I had the bright idea to try to modify the source until I also realized that there are no PostScript printers (that I know of) that have a bidirectional parallel port (the serial one is). Since Transcript needs to establish a bidirectional link with the printer, it's a no go.

Robert -- | Robert L. Howard | Georgia Tech Research Institute | | | MATD Laboratory | | (404) 528-7165 | Atlanta, Georgia 30332 | ------------------------------------------------------------------ From: John DiMarco <> To: Subject: Re: Using Sun Transscript Software with parallel Interfaces

Bad news: the HP IIIsi doesn't support bidirectional communications over the serial port when in postscript mode. Transcript requires this, since it queries the printer before and after each job for a pagecount.

Either you have to use the serial port, or you have to replace transcript with some other package. If you want to use the parallel port, you almost definitely will not be able to do printer accounting, however.

John -- John DiMarco Computing Disciplines Facility Systems Manager University of Toronto EA201B,(416)978-1928 ------------------------------------------------------------------ From: (Bob Netherton) To: Subject: Re: Using Sun Transscript Software with parallel Interfaces

The problem is that the Adobe/Sun Transcript software is really designed for a serial connection to a printer. The output filter fires up a program called pscomm that is responsible for talking to the printer. The first thing it does is to start up a child that does a read on the port. I believe this is where you are having problems. You are going to have to hack up pscomm to deal with a readonly device and replace the child process with something that does ioctl's to get the device status. It is not too bad, but you will need source to do it.

You can get it from Adobe for about $3000-5000 (US dollars).

Hope this helps

Bob -- = Bob Netherton = ObQuote: See the signs they're always there = Ericsson Network Systems = But you know you never care = PO Box 833875 M/S B-07 = You're always looking for another reason = Richardson, Tx 75083-3875 = - Hawkwind, Paradox = (214) 907-5992 = ObCDplayer: Deep Purple, Live in London - 1974 ------------------------------------------------------------------ From: (Klamer Schutte) Subject: Re: Using Sun Transscript Software with parallel Interfaces

At a time we had a similar setup, with another parallel cart && a HP III D + HP Postscript. We made an ouput filter ourselves (the $comm program in the script /usr/tran/sparc/lib/psif).

And it worked for 99.99 %. A pity that a great deal of all out PS files are bigger than 10000 bytes, and PostScript is lousy in error recovery...

We traced back the error to the parallel card, got it refunded, and are yet using an rs232 connection. Grmbl.

So if you get it all, errorfree, running, please let me know. If you want the output program we wrote, let me know as well.


-- Klamer Schutte Tel: +31-53-892786 Fax: +31-53-340045 Faculty of electrical engineering -- University of Twente, The Netherlands preferred: SMTP: ------------------------------------------------------------------ From: Wilson N G <> To: Message-Id: <> Subject: Re: Postscript printer on centronics ports Status: R

I think your problem is likely to be that the output filter (of=) is trying to get a response from the printer, and the printer cannot send characters UP a parallel port. It isn't quite clear from your message whether you mean that you've tried it without an of= entry, but if so (and it still fails), it is possible that lpd or some default filter is also attempting to do so.

------------------------------------------------------------------ From: Mike Raffety <> To: Subject: Re: Postscript printer on centronics ports Status: R

PostScript, being a true, interpreted, language, needs a two-way communications line back to whoever is running the program (e.g., TranScript). Parallel ports, in general, are uni-directional (one-way).

I don't know of a reliable way to run a PostScript printer on a parallel interface.

Please be sure to summarize back to the list; thanks.

------------------------------------------------------------------ ------------------------------------------------------------------ ------------------------------------------------------------------ ------------------------------------------------------------------ ------------------------------------------------------------------ ------------------------------------------------------------------ ------------------------------------------------------------------ ------------------------------------------------------------------ ------------------------------------------------------------------ ------------------------------------------------------------------ -- Karl-Jose Filler Voice: +49-911-526-2870 Philips Kommunikations E-Mail: pla_jfi@pkinbg.uucp Industrie Ag D-8500 Nuernberg Germany

This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:34 CDT