Received: from bryggen.bgnett.no (datadok@bryggen.bgnett.no [194.19.41.2]) by minnie.cs.adfa.oz.au (8.6.13/8.3) with ESMTP id GAA14024; Fri, 17 Jan 1997 06:37:24 +1100
Received: (from hamartun@localhost) by bryggen.bgnett.no (8.8.4/8.7.3/brage 1.0) with UUCP id TAA21253 for oldunix@minnie.cs.adfa.oz.au; Thu, 16 Jan 1997 19:21:08 GMT
Received: from barsoom.Hamartun.Priv.NO (tih@barsoom.Hamartun.Priv.NO [192.168.1.33])
          by barsoom.Hamartun.Priv.NO (8.8.4/8.8.4) with ESMTP
	  id UAA12286 for <oldunix@minnie.cs.adfa.oz.au>; Thu, 16 Jan 1997 20:20:12 +0100 (CET)
Date: Thu, 9 Jan 1997 20:18:56 +0100 (CET)
From: Tom I Helbekkmo <tih@Hamartun.Priv.NO>
To: info-pdp11@transarc.com
Subject: Weird TS05 gotcha
Message-ID: <970109195614.25256A@barsoom.Hamartun.Priv.NO>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
ReSent-Date: Thu, 16 Jan 1997 20:20:08 +0100 (CET)
ReSent-From: Tom I Helbekkmo <tih@Hamartun.Priv.NO>
ReSent-To: PDP Unix Preservation Society <oldunix@minnie.cs.adfa.oz.au>
ReSent-Message-ID: <970116202008.10317D@barsoom.Hamartun.Priv.NO>

I just wanted to share an interesting experience that might save
someone else some time if the occasion should arise...

I've got a Q4 cabinet with two BA23s in the middle, an RA81 below them
and a TS05 tape station on top.  It's had a KA630 (a MicroVAX II, that
is) CPU running VAX/VMS, Ultrix and 4.3BSD-Reno before, and the TS05
has always behaved perfectly.  Just recently, I reconfigured this box
with a MicroPDP-11/73 (a great little system: two RD54s, twin RX50s,
TK50 and TS05, DELQA, DHQ11 and the RA81), and was dismayed to find
that my trusty old 9 track tape station no longer worked!

After several, unsuccessful attempts to get some (too) old diagnostic
software to work, I put the KA630 back in, and ran MDM, the MicroVAX
Diagnostic Monitor.  The tape station checked out perfectly.  Back in
went the 11/73 -- sure enough, it didn't work.  Experimenting showed
that I could fsf, bsf rewind and stat the tape station with mt all I
wanted, but I couldn't read nor write.  The controller always gave an
illegal address error, which the manual says is what happens when you
use it in a 22 bit qbus while it's configured for 18 bit operation.

Of course, that couldn't be for real, right?  I mean, the KA630 is a
22 bit system, and it worked on that with several operating systems!
Still, it doesn't hurt to check, so I pulled the controller.  Yup, it
was set to 18 bit mode.  Flipped it to 22, turned on buffering at the
same time -- and I now have a fast, dependable 1600bpi 9 track again!

If anyone can explain how this thing worked in the first place, I'd
appreciate it!  (Oh, and if anyone has some hints for this youngster
about the proper care and feeding of my TS05 as the years go by, that
would come in very handy as well!)  (Heck, while I'm asking all this,
an RK05 with qbus controller and a few packs would be great, too, and
would go real well with this old /23 I've got sitting here!)

-tih
-- 
Popularity is the hallmark of mediocrity.  --Niles Crane, "Frasier"



