Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!ncar!husc6!bu-cs!bucsb!bear
From: bear@bucsb.UUCP (Blair M. Burtan)
Newsgroups: comp.sys.amiga
Subject: Re: Problems with A1000 expansion devices
Message-ID: <2895@bucsb.UUCP>
Date: 24 Jul 89 20:06:58 GMT
References: <20192@louie.udel.EDU>
Reply-To: bear@bucsb.bu.edu (Blair M. Burtan)
Followup-To: comp.sys.amiga
Organization: Boston Univ Comp. Sci.
Lines: 36

In article <20192@louie.udel.EDU> D88.R-WESTMAN@linus.ida.liu.se (Rickard Westman) writes:
>
>The guru meditation numbers are of type 00000003 and 00000008, indicating
>address errors and privilege violations.  Typically, they occur after about
>ten minutes use, but sometimes much earlier.  The hard drive does not even
>need to be turned on to cause these problems, so I guess there might be
>something wrong with the pass-thru/interface.
>
>I've read in an article by Perry Kivolowitz, that problems like these
>could be caused by poor grounding of some chips on the Kickstart RAM
>daughterboard, or possibly by PAL devices with weak drive capacity.
>
Well, I have the same problems with a lengthy history.  Here's the
scenario:   I've had my 1000 with a Starboard 2Meg and a Microbotics
20 meg harddrive for about a year running on DOS 1.2 with no problems.
Ever since I upgraded to 1.3, I've experienced frequent crashes to
the point of having to re-insert Kickstart.  When I say frequent,
I mean several times a day.  This was unacceptable.  During the crash,
the hard drive light would be lit.  The lead me to believe that there
was an incompatibility between the controller and 1.3.  I called
Microbotics and they told me to check/ground the PALS.  First, the
Texas Instruments PALs are supposed to be the good ones.  Next ground
the chips with some heavy wire (not wire-wrap wire).  There are several
articles on teh subject.  The end result.  Nothing.  No change.
Grumble, grumble, grumble.  So, I broke down and bought the 1.3
kickstart eliminator.  Unfortunately that's curing the symptom rather
than the disease.  Well, as it turned out.  I had a really old version
of Supraboot.  The new version helped a great deal.  But not entirely.
And now I hear that there are problems with SetPatch leaving the
writeable control store open to attack.  Oh, and by the way, GOMF 3.0
gets fried as well.  It can't catch this type of crash.
   So, does anyone have the same problem? And can we throw our 1.3
disks at CBM until they do something about it?  Let's hear some
feedback.

                                   - Bear