From owner-freebsd-hackers  Sun Oct 26 17:46:20 1997
Return-Path: <owner-freebsd-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id RAA04991
          for hackers-outgoing; Sun, 26 Oct 1997 17:46:20 -0800 (PST)
          (envelope-from owner-freebsd-hackers)
Received: from misery.sdf.com (misery.sdf.com [204.244.210.193])
          by hub.freebsd.org (8.8.7/8.8.7) with SMTP id RAA04984
          for <freebsd-hackers@freebsd.org>; Sun, 26 Oct 1997 17:46:17 -0800 (PST)
          (envelope-from tom@sdf.com)
Received: from tom by misery.sdf.com with smtp (Exim 1.73 #1)
	id 0xPeEA-0006yr-00; Sun, 26 Oct 1997 17:44:22 -0800
Date: Sun, 26 Oct 1997 17:44:17 -0800 (PST)
From: Tom <tom@sdf.com>
To: Jerry Hicks <wghhicks@ix.netcom.com>
cc: Mike Smith <mike@smith.net.au>, freebsd-hackers@freebsd.org
Subject: Re: Parity Ram
In-Reply-To: <3452B4B9.8A4510A9@ix.netcom.com>
Message-ID: <Pine.BSF.3.95q.971026173258.26745H-100000@misery.sdf.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-freebsd-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk


On Sat, 25 Oct 1997, Jerry Hicks wrote:

> Mike Smith wrote:
> > 
> > > Yeah, but when you go to buy memory they have ECC *OR* parity types
> > > available. ECC generally costs more than parity...
> > 
> > Crap.  In the case of 72-pin parts, parity and ECC are both x36.  30-pin
> > parts are either x8 (no parity) or x9 (parity, or ECC in groups of 4).
> > 
> > mike
> 
> Hi Mike,
> 
> i understood that they're referred to as ECC vs PARITY because of the
> memory controller interface. True, the same number of bits are found in
> devices labeled both ways.
>
> Some ECC controllers must have special requirements of the interface to
> the memory (timing, interleaving ?)

  Depends.  PC motherboards implement ECC with standard parity RAM.  Most
motherboards all use the same Intel chipset, where is all handled.

  I've seen custom ECC memory before (x38), but it only worked on one
particular device.

> A quick AltaVista search on ECC found this link: 
> 
> http://www1.ibmlink.ibm.com/HTML/SPEC/6062ipme.html#pari
> 
> (IBM, ugh) In this example they are pretty explicit that parity memory
> is different from ECC-EDO devices.  i'll bet the prices are different
> too.

  Who cares... it is price sheet.

> Do you know anything of Richard Hamming's assertion that parity memory
> (the old fashioned even/odd type) is-a-bad -thing in large
> configurations?

  I think it bullshit.  I've never heard of this before.  Nor have you in
the two times you've mentioned it, actually stated what is supposed to be
so bad about it.

Tom