Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!usc!wuarchive!uwm.edu!ogicse!milton!serval!luke.eecs.wsu.edu!rnelson
From: rnelson@luke.eecs.wsu.edu (Roger Nelson - Grad Student)
Newsgroups: comp.sys.sgi
Subject: Re: Personal Iris 1 Meg Simms
Keywords: Simms Iris
Message-ID: <1991Apr17.173300.10850@serval.net.wsu.edu>
Date: 17 Apr 91 17:33:00 GMT
References: <1991Apr16.143152.13058@meteor.wisc.edu> <1991Apr17.163254.18574@colorado.edu>
Sender: news@serval.net.wsu.edu (USENET News System)
Distribution: usa
Organization: Washington State University
Lines: 23

In article <1991Apr17.163254.18574@colorado.edu> rainer@boulder.Colorado.EDU (Rainer Malzbender) writes:
>In article <1991Apr16.143152.13058@meteor.wisc.edu> bt@sleaze.meteor.wisc.edu (Bill Taylor) writes:
>>I put in Panasonic MN41C1000ASJ-08 which are supposed to be 80 ns.
>>It has toshiba 1 Meg simms in it. Anyone want to share constaints
>>in what can be installed?
>
>This may not be reliable since we are in the process of chasing down an
>intermittent memory error as well, but we have three kinds of SIMMs in
>our 4D25TG's (former SGI demo machines):
>
>	Fujitsu MB85235-10
>	Toyocom TH3C10009-T10
>	Mitsubishi MM1M09A0J-10
>
>I'm suspecting a motherboard problem, since the SIMMs seem to check out.
>It looks like putting *any* memory above 8 megs causes diagnostics to
>fail (with a wonderfully descriptive message like "error in Electronics
>Module :-). 
>

We had similar problems when we added memory to our machines.
The faster chips must be placed in the front.
Once we did this, the machines work just fine.