From pat@ucc.gu.uwa.edu.au Mon Jun 28 10:09:15 1993
Return-Path: <pat@ucc.gu.uwa.edu.au>
Received: from gossip.pyramid.com by dutecai.et.tudelft.nl (4.1/1.34JP)
          id AA24756; Mon, 28 Jun 93 10:09:10 +0200
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA18874; Mon, 28 Jun 93 01:09:04 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA20651; Mon, 28 Jun 93 01:08:45 -0700
Received: from mackerel.gu.uwa.edu.au 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA18691; Mon, 28 Jun 93 01:08:33 -0700
Received: by ucc.gu.uwa.edu.au id AA03883
  (5.65c/IDA-1.4.4 for riscy@pyramid.com); Mon, 28 Jun 1993 16:00:59 +0800
From: Pat Mackinlay <pat@ucc.gu.uwa.edu.au>
Message-Id: <199306280800.AA03883@ucc.gu.uwa.edu.au>
Subject: SIMM sockets...
To: riscy@pyramid.com
Date: Mon, 28 Jun 1993 16:00:55 +0800 (WST)
X-Mailer: ELM [version 2.4 PL11]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 739       
Status: RO
X-Status: 


Neil, are you still planning on using 32 SIMM sockets? I've been looking,
and I just can't see how you plan to fit that many on the board if you
plan to do an ISA bus as well. 32 sounds a little excessive in any case.
I think we might do better to think about 8, possibly 16. More than 4
sockets is going to mean external memory drivers anyhow, so there's no
real difference between 8 and 16 apart from board space. Perhaps we can
allow 16M SIMMS to be used? With 8 x 16M SIMMS, we get the same 128M
maximum, perfectly reasonable from my point of view <grin>.

Pat -- "There's only one thing left to do Mama, I got to ding a ding dang
	my dang a long ling long" (Jesus Built My Hotrod -- Ministry)
GCS d* -p+ c++ l++ m--- s+/- !g w- t- r

 
From Steven.D.Ligett@Dartmouth.EDU Mon Jun 28 15:04:58 1993
Return-Path: <Steven.D.Ligett@Dartmouth.EDU>
Received: from gossip.pyramid.com by dutecai.et.tudelft.nl (4.1/1.34JP)
          id AA25607; Mon, 28 Jun 93 15:04:48 +0200
Received: from sword.eng.pyramid.com 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA04563; Mon, 28 Jun 93 06:04:43 -0700
Received: from goss.pyramid.com
	by sword.eng.pyramid.com (5.61/Pyramid_Internal_Configuration)
	id AA24887; Mon, 28 Jun 93 06:04:27 -0700
Received: from dartvax.dartmouth.edu 
	by gossip.pyramid.com (5.61/OSx5.1a Pyramid-Internet-Gateway)
	id AA04444; Mon, 28 Jun 93 06:04:25 -0700
Received: from prancer.dartmouth.edu by dartvax.dartmouth.edu (5.65+D5/4.5HUB)
	id AA21049; Mon, 28 Jun 93 09:03:06 -0400
Message-Id: <5283630@prancer.Dartmouth.EDU>
Date: 28 Jun 93 09:03:03 EDT
From: Steven.D.Ligett@Dartmouth.EDU (Steven D. Ligett)
Subject: Re: SIMM sockets...
To: riscy@pyramid.com
Status: RO
X-Status: 

--- Pat Mackinlay wrote:
Neil, are you still planning on using 32 SIMM sockets? I've been looking, and
I just can't see how you plan to fit that many on the board if you plan to do
an ISA bus as well.... Perhaps we can allow 16M SIMMS to be used? With 8 x
16M SIMMS, we get the same 128M maximum, perfectly reasonable from my point
of view <grin>.
--- end of quoted material ---
I think that the problem is the plan to use the glue chip that only supports
1MB and 4MB chips.  (If I understand what's been said about the 7370 (?).)  
A typically shortsighted chip design, I'd say.

Let's take a quick look at the costs of the board and sockets for these 32
SIMMs.  Ignore the space and cost of the memory drivers that will be needed.

For a 6-layer board, $1.50 per square inch for the PCB.  Each socket is about
0.3" by 4".

board cost = $1.50 * 32 * 0.3" * 4" = $57.60  
just for the board space.

socket cost = $1.95 * 32 = $62.40
for the sockets.

$120 to put 32 empty SIMM sockets on the board.  Assume that I'm
misremembering the socket cost, and that we go with a 4-layer board, and that
goes down to $100.  Still too much.

 
