[pups] Spares, et. al for 11/44 system

[pups] Spares, et. al for 11/44 system

[Fred N. van Kempen]

Greg,

> Thought about that too but the math doesn't work.  40+64 chips = 104
> chips * 65536 bits per chip = 6815744 total bits.
> 
> 6815744 doesn't divide cleanly by any of 9, 18, 19, 36, or 38
Right... 12 bits of ECC seems a bit much.  Dunno what that stuff is
about... weird! ;)

Fred


**********************************************************************
This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the system manager.

This footnote also confirms that this email message has been swept by
MIMEsweeper for the presence of computer viruses.

www.mimesweeper.com
**********************************************************************

Received: (from major at localhost)
	by minnie.cs.adfa.edu.au (8.9.3/8.9.3) id IAA27971
	for pups-liszt; Thu, 8 Mar 2001 08:20:35 +1100 (EST)
	(envelope-from owner-pups at minnie.cs.adfa.edu.au)
Received: from chiton.ucsd.edu (chiton.ucsd.edu [192.135.238.128])
	by minnie.cs.adfa.edu.au (8.9.3/8.9.3) with ESMTP id IAA27966
	for <pups at minnie.cs.adfa.edu.au>; Thu, 8 Mar 2001 08:20:30 +1100 (EST)
	(envelope-from cdl at mpl.ucsd.edu)
Received: (from cdl at localhost)
	by chiton.ucsd.edu (8.9.3/8.9.3) id NAA15151;
	Wed, 7 Mar 2001 13:15:48 -0800 (PST)
Date: Wed, 7 Mar 2001 13:15:48 -0800 (PST)
From: Carl Lowenstein <cdl@mpl.ucsd.edu>
Message-Id: <200103072115.NAA15151 at chiton.ucsd.edu>
To: greg at ciswired.com, pups at minnie.cs.adfa.edu.au
Subject: Re: [pups] Spares, et. al for 11/44 system
Sender: owner-pups at minnie.cs.adfa.edu.au
Precedence: bulk

> Date: Tue, 6 Mar 2001 20:07:00 -0500 (EST)
> From: "Gregory R. Travis" <greg at ciswired.com>
> To: pups at minnie.cs.adfa.edu.au
> Subject: [pups] Spares, et. al for 11/44 system
> 
> 
> p.s.  The Nat. Semi board is a bit strange.  It has an area of
> 16x4 TI 64K-bit chips (i.e  64x64k/8 = .5MB) and another
> area of 10x4 TI 64K-bit chips (i.e. .3MB).  No matter
> what I do, I can't do the math to get this board to
> fit into a 256/512/1MB size.  I ASSUME it's a .5MB
> board, but what about the extra chips?

Nat.Semi. used to make ECC memory boards for the PDP11.

I had one once, it looks like you have one now.  My arithmetic says 5
extra bits per 8-bit byte makes for single-error correcting, dual-error
detecting ECC on the byte level.  Vaxes do it with 39 bits per 32-bit
word, Alphas do it with 72 bits per 64-bit word.  Economy of scale.

As I remember, you could just ignore the ECC and it would work like
a standard parity memory board, except that it very rarely had any
parity errors.

    carl

        carl lowenstein         marine physical lab     u.c. san diego
        {decvax|ucbvax} !ucsd!mpl!cdl                 cdl at mpl.ucsd.edu
                                                  clowenstein at ucsd.edu

Received: (from major at localhost)
	by minnie.cs.adfa.edu.au (8.9.3/8.9.3) id SAA32608
	for pups-liszt; Thu, 8 Mar 2001 18:55:06 +1100 (EST)
	(envelope-from owner-pups at minnie.cs.adfa.edu.au)
Received: from mserv1c.vianw.co.uk (mserv1c.vianw.co.uk [195.102.240.33])
	by minnie.cs.adfa.edu.au (8.9.3/8.9.3) with ESMTP id SAA32604
	for <pups at minnie.cs.adfa.edu.au>; Thu, 8 Mar 2001 18:55:03 +1100 (EST)
	(envelope-from pete at dunnington.u-net.com)
Received: from p113.nas4.is5.u-net.net ([195.102.201.241] helo=indy.dunnington.u-net.com)
	by mserv1c.vianw.co.uk with smtp (Exim 3.22 #5)
	id 14avBh-0000Ji-00
	for pups at minnie.cs.adfa.edu.au; Thu, 08 Mar 2001 07:50:17 +0000
Received: by indy.dunnington.u-net.com (951211.SGI.8.6.12.PATCH1502/980207.PNT)
	for pups at minnie.cs.adfa.edu.au id WAA24282; Wed, 7 Mar 2001 22:10:25 GMT
Date: Wed, 7 Mar 2001 22:10:25 GMT
From: pete@dunnington.u-net.com (Pete Turnbull)
Message-Id: <10103072210.ZM24280 at indy.dunnington.u-net.com>
In-Reply-To: "Gregory R. Travis" <greg at ciswired.com>
        "RE: [pups] Spares, et. al for 11/44 system" (Mar  7,  7:59)
References: <Pine.LNX.4.10.10103070756400.21204-100000 at weasel.ciswired.com>
X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail)
To: pups at minnie.cs.adfa.edu.au
Subject: Re: [pups] Spares, et. al for 11/44 system
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 8bit
Sender: owner-pups at minnie.cs.adfa.edu.au
Precedence: bulk

On Mar 7,  7:59, Gregory R. Travis wrote:

> Thought about that too but the math doesn't work.  40+64 chips = 104
> chips * 65536 bits per chip = 6815744 total bits.
>
> 6815744 doesn't divide cleanly by any of 9, 18, 19, 36, or 38

It divides by 13, though (64K x 13 x 8, so the prime factors are 2 and 13).
 Unlikely as it sounds, could this be for a 12-bit word machine?

13 is not only 12+1, it's 8+5.  5 is a common number of bits for ECC
(though not, I admit, for 8-bit words!  Usually it's 8+3, 16+5, or 16+6).

It also strikes me as odd that a NatSemi board should have TI chips on it.
 NatSemi made their own RAM, and Texas made their own PDP-11 and Vax memory
boards.

-- 

Pete						Peter Turnbull
						Network Manager
						Dept. of Computer Science
						University of York