Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watmath!clyde!rutgers!mit-eddie!uw-beaver!uw-june!uw-entropy!dataio!bright
From: bright@dataio.UUCP
Newsgroups: comp.sys.ibm.pc
Subject: Re: More memory chip stuff
Message-ID: <1306@dataio.Data-IO.COM>
Date: Tue, 26-May-87 16:17:55 EDT
Article-I.D.: dataio.1306
Posted: Tue May 26 16:17:55 1987
Date-Received: Wed, 27-May-87 05:31:06 EDT
References: <19522@sun.uucp>
Reply-To: bright@dataio.UUCP (Walter Bright)
Distribution: na
Organization: Data I/O - FutureNet Corp., Redmond, WA
Lines: 27

In article <19522@sun.uucp- dbercel@sun.UUCP (Danielle Bercel, MIS Systems Programming) writes:
-To those of you who sent me chip locations, thank you.
-Unfortunately, none of you agree with each other. So, let
-me make my plea one more time.
-
-My friend has a memory parity error and the chip is in
-location: 0480 (hex). From what I've been able to determine,
-this chip is in bank zero on the motherboard. That would mean it
-was one of the soldered chips. However, I've been told that
-the chip is bank five, bank one, and bank zero (expansion).
-
-Is there anyone out there who really knows the chip location?
-According to the hardware reference manual the "04" indicates
-the 2nd 16K of memory which makes it bank zero. But,
-according to the IBM diagnostics the "04" are the eight
-high order bits from a 20-bit memory address. This has
-now got my tiny little brain completely confused. 

There are two ways to determine this:

	1. Read the BIOS listing where the memory test is done and
	the error message is printed out. I've done this before, and
	replaced the right chip the first time.

	2. Pull out a memory chip and see how the diagnostic location
	changes. A few experiments will make it obvious what the
	encoding scheme is.