Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!cornell!uw-beaver!rice!sun-spots-request
From: smaug@eng.umd.edu (Kurt Lidl)
Newsgroups: comp.sys.sun
Subject: 12 Megs on a 3/60
Keywords: Hardware
Message-ID: <790@brazos.Rice.edu>
Date: 11 Aug 89 05:33:00 GMT
Sender: root@rice.edu
Organization: Sun-Spots
Lines: 42
Approved: Sun-Spots@rice.edu
X-Sun-Spots-Digest: Volume 8, Issue 95, message 4 of 11

>| Has anyone run any 3/60's with 12 meg of memory and noticed system
>| crashes, where there weren't any with 8 meg?  I haven't checked yet to
>| determine if it may be application specific, but using 3 different cpu's
>| and 2 different cpu cabinets, and 4 different SIMMs, we get crashes on the
>| 3/60 with 12 meg, and not with 8....I can't believe they all have bad SIMM
>| sockets. Most crashes seemed to be memory parity errors, however.  Sun
>| tech support is unaware of any problem with the 12 meg value. Could an
>| application do it?
>|
>|       Curiously yours,
>|       Dick Mead  <mead@hamal.usc.edu>

>We upgraded our 2 3/60's to 12MB of memory by adding Clearpoint SIMMs.
>We haven't had a single memory related reboot and it's been 8 months since
>the upgrade.

>Scott J. Kamin          {att!uswat, isis, nbires, dunike} !onecom!sjk
>TelWatch Inc. (formerly OneCom) (800) 669-1266 (outside Colorado)
>2905 Wilderness Place           (303) 440-4756 (switchboard)
>Boulder, CO 80301               (303) 938-6726 (direct)

I wish I could say the same with the memory expansion in some of our
3/60's here at the University.  We have taken a Sun supplied memory
expansion kit, put it in a 3/60 (expanding the memory from 8 to 12
megabytes) and have it fail to properly boot /vmunix.  Normally, the
system will not get to the fsck on the filesystems before it starts to
complain about a memory failure.

The worst part is that the problem is NOT with the memory SIMMS, but
rather the motherboards.  The same memory boards will work perfectly well
with another 3/60.  There seems to be no rhyme or reason to whether or not
the motherboards fail to run 12 megabytes of RAM or not.  Of the five
machines that failed, we were able to get all but two of them to work
properly with at least one combination of memory chips.  This suggests to
me that the memory is marginally too slow for the bus, or that some of the
chips are marginal...  A very irritating problem that was worked around
(in most cases) by swapping around SIMMS until everything worked...

==================================================================
==  Kurt J. Lidl  (smaug@bacchus.eng.umd.edu)	(301) 345-6243	==
==  UUCP: uunet!mimsy!bacchus!smaug		(301) 454-1526	==
==========="Amiga : The Computer for the Best of Us..."===========