adrian@ru-cs44.UUCP (Adrian Pell) (07/22/83)
Thanks to all the people who replied to my question on usage of Versatec
printer/plotters. Credits at the end. Here's a summary:
1) What software do you use?
The general standard seems to be vtroff (from Berkeley, see
below), which is a postprocessor for troff. Other software
used it [dt]itroff, the device-independent typesetter package
of Kernighan, vplot, pic, ideal, DIGS (Device Independent
Graphics System).
2) Who supplies it?
Vtroff comes from Berkeley in the 4.xBSD distributions, although
not in the 2.xBSD distributions (why?). It's reputed to be
available from Tom Ferrin at UC San Francisco (ucbvax!ucsfcgl!tef),
for the 11's. (Tom, are you out there?) Vplot is in Berkely
or SIII; pic, ideal and [dt]itroff are available (maybe) from AT&T;
DIGS comes from Heriot-Watt University, Edinburgh, Scotland
- no details of availability.
3) Will it fit on a PDP11/44?
It seems that it will - well most of it. Vtroff and vplot
are known to be working on I?D 11's. Some bits of the rest
are too.
4) How much of a drain on the system is it?
If you have a lightly loaded 780, not a lot. A bit more on
the smaller CPU's! Vtroff generates a lot of output (figures
of .5-1 MB per foot have been mentioned), and if you can't
generate it fast enough to drive the printer direct you'll
need some (large) temporary file space. (The Versatec needs
to be kept moving, otherwise black lines result). Suggestions
that vtroff is about double the load of troff which is about
double the load of nroff seem to be sustained.
5) Reliability?
Opinions on reliability seem very subjective. It probably
depends on how good your engineer is, or whether the animal
was made on a Friday afternoon, ... There are bad experiences:
lack of intensity, slipping paper, power supplies burning
out; but there are also good experiences. You take your choice.
Thanks to:
vax135!ukc!root44!pdl
philabs!mcvax!jaap
philabs!mcvax!teus
ihnp4!kpno!downey
decwrl!qubix!msc
duke!dbl
decwrl!turtlevax!ken
utcsstat!laura
vax135!edcaad!edmrc!jimp
utcsstat!geoff
harpo!presby!burdvax!puder
ucbvax!UCBERNIE:carl
for taking the trouble to reply.
Adrian Pell ({vax135,mcvax}!ukc!ru-cs44!adrian)
Dept of Computer Science
University of Reading
Englandellis@flairvax.UUCP (Michael Ellis) (07/23/83)
We have a Versatec V80 matrix printer/plotter running on our Vax
780/Unix 4.0bsd system (is there a law somewhere that I gotta say
`trademark of Bell Labs'?) and it's the most bitchiest hardware we
have, I mean constantly hanging, so we have to power cycle the thing
and kick /usr/lib/vpd -- frequently having to resubmit the `job'.
With persistence, you can still get anything to run, however.
Now I don't mean to bad mouth Versatec. I think I've seen the thing
work well on some systems. Does anyone know if the driver `vp.c' has
severe brain damage (and have fixes for it)? Does anyone else have
problems with their Versatec V80?
I already installed William Sebok's vmpage fix for raw DMA devices,
and it's likely that fix corrected many failures. But the thing is
still bloody cantankerous as hell. Especially on warm summer days.
Michael Ellis - Fairchild AI Lab - Palo Alto CA - (415) 493-7250msc@qubix.UUCP (07/24/83)
I have been involved in the installation of 2 Versatec v80's on Vax
11/750's running 4.1bsd. Both installations were painful. The first
installation was the first v80 to be put on a 750 running Unix. Because
of this the finger pointing when it didn't work was greater than usual.
In the end it turned out we had the long lines option in the printer but
not in the controller so they couldn't talk to each other. (Versatec's fault.)
On the second installation, I took the vp.c documentation literally
and asked the Versatec FE to wire both interrupts to the same vector.
I then built the system to expect just one interrupt. That didn't
work. The system would hang when autoconf was probing for the controller..
I tried again the more normal 2 interrupt way only to find that the
controller was not correctly configured.
After the installation woes were over, both v80's have performed
without any problems at all.
--
Mark
...{decvax,ucbvax,ihnp4}!decwrl!
...{ittvax,amd70}!qubix!msc
decwrl!qubix!msc@Berkeley.ARPAthomas@utah-gr.UUCP (Spencer W. Thomas) (07/24/83)
The only problems we've had with our V80 (at least since it's been attached to our Vax) have been strictly user-caused mechanical failures (like opening the front without pushing the little "push to open" lever.) And we're even running it through a multiplexor! (Maybe that's the secret?) Its print quality is a little ragged when running vtroff (alternate pixels are off by as much as 1/400th of an inch (i.e., half a dot)), but that's more or less acceptable. Considering the amount of output that gets pushed through the beast, it's remarkably reliable (I don't remember the last time it was down). Maybe you got a lemon? =Spencer
ellis@flairvax.UUCP (Michael Ellis) (07/25/83)
My apologies to this overloaded newsgroup, my original article seeking advice for Versatec trouble should have read `4.1(a)bsd', (soon to be 4.2). -michael