Newsgroups: comp.sys.ibm.pc.hardware
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!noc.near.net!howland.reston.ans.net!usc!elroy.jpl.nasa.gov!swrinde!network.ucsd.edu!pacbell.com!att-out!cbfsb!cbnewsg.cb.att.com!rnichols
From: rnichols@cbnewsg.cb.att.com (robert.k.nichols)
Subject: Re: how to search for bad memory chips.
Message-ID: <1993Apr17.235135.5397@cbfsb.cb.att.com>
Sender: news@cbfsb.cb.att.com
Organization: AT&T
References: <N5s42B8w165w@c-cat.UUCP>
Distribution: na
Date: Sat, 17 Apr 1993 23:51:35 GMT
Lines: 29

In article <N5s42B8w165w@c-cat.UUCP> david@c-cat.UUCP (Dave) writes:
>i came upon this idea i would like to share with everyone.
>
>to check for bad memory chips
>
>1. create a boot disk with emm386 himem.sys and ramdrive.sys in the
>   config/autoexec.bat.
>
>2. boot the PC to create a RAM drive as large as possible.
>
>3. use a disk repair utility ( I use NDD ). Run it on the RAM
>        drive, yes it will run, its only a device driver
>
>4. run 1000 or so passes, they go very quick
>
>5. if your machine fails, there is a definate bad memory chip
>
>6. if your machine passes, there is a conflict with programs you
>        are loading in memory.
...

It's an interesting idea, but the worst-case data patterns developed to
test magnetic media are totally different than the patterns used to detect
common faults in memory chips.

--
Bob Nichols
AT&T Bell Laboratories
rnichols@ihlpm.ih.att.com
