Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!csd4.milw.wisc.edu!bionet!agate!ucbvax!hplabs!hp-pcd!hpvcfs1!neff
From: neff@hpvcfs1.HP.COM (Dave Neff)
Newsgroups: comp.sys.atari.st
Subject: Re: DeskJet Plus weirdness
Message-ID: <2150015@hpvcfs1.HP.COM>
Date: 11 Aug 89 19:28:15 GMT
References: <1546@ns.network.com>
Organization: Hewlett Packard, Vancouver, WA
Lines: 36

Re DeskJet+ weirdness:

A number of Atari ST users have run into the same situation.  DeskJet+
must be on when turning on the Atari.  Evidently the strobe line on the
Atari is stuck low so the DeskJet+ automatically asserts busy until the
strobe line is high.  Paul Close has posted a program that gets the
strobe line unstuck.  Some Atari users have the problem, some do not.
It may be that the current sink limitation on the Atari parallel port
is aggrivating the problem.  One DeskJet+ user said he is using the
7407 chips as line buffers and he has no problems powering on his
hardware in any order.  It isn't clear to me if he had the problem
before he buffered the line.  There is no good reason I know that
the Atari would be holding down the strobe line -- the strobe line
should be strobed :-).

I don't know why some Atari users run into this but others do not.
We have a couple Atari ST users here who have never run into this
problem (and they are not buffering their lines).  I would like
to know if the line buffers truly fix the problem or not.

Unfortunately (for you, actually fortunately for me) I am not in
product support and do not have the time to work very hard on this problem.
Paul and Dianne Close first contacted me about the power on order
dependancy and I worked with them on the problem for a while.

Would all DeskJet+ Atari ST users contact me telling me if they do
or do not experience this problem (must power on DeskJet+ before
the Atari)?  If someone had this problem but fixed it with the line
buffers would they please let me know.  I have no idea why DeskJet
users (only DeskJet+) run into this problem.  The DeskJet+ should
look identical to the parallel port from the standpoint of electrical
characteristics and handshaking -- especially when no data is being
sent :-). 

Dave Neff
neff@hpvcfs1.HP.COM