From owner-freebsd-hardware Tue Feb 4 23:37:34 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA10779 for hardware-outgoing; Tue, 4 Feb 1997 23:37:34 -0800 (PST) Received: from dfw-ix10.ix.netcom.com (dfw-ix10.ix.netcom.com [206.214.98.10]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA10774 for ; Tue, 4 Feb 1997 23:37:32 -0800 (PST) Received: (from smap@localhost) by dfw-ix10.ix.netcom.com (8.8.4/8.8.4) id BAA28191; Wed, 5 Feb 1997 01:36:56 -0600 (CST) Received: from wck-ca14-60.ix.netcom.com(207.92.174.124) by dfw-ix10.ix.netcom.com via smap (V1.3) id sma028180; Wed Feb 5 01:36:48 1997 Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.5/8.6.9) id XAA25845; Tue, 4 Feb 1997 23:35:27 -0800 (PST) Date: Tue, 4 Feb 1997 23:35:27 -0800 (PST) Message-Id: <199702050735.XAA25845@silvia.HIP.Berkeley.EDU> To: skynyrd@opus.cts.cwu.edu CC: freebsd-hardware@FreeBSD.ORG In-reply-to: (message from Chris Timmons on Tue, 4 Feb 1997 19:41:31 -0800 (PST)) Subject: Re: (not seeing) ECC penalty on Natoma From: asami@vader.cs.berkeley.edu (Satoshi Asami) Sender: owner-hardware@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk * With 128MB (2x16x36) and an async /usr, i see no difference in the mw * build time between checking nothing and doing ECC. I would think that It can be either because (1) Natoma doesn't do anything when ECC is turned on (2) Natoma's ECC is as fast as non-ECC (or non-ECC is as slow as ECC) (3) Your benchmark does not exercise memory bandwidth very much I don't know how you can tell the difference between 1 and 2, but at least to see if it's 3 or not, pick up ftp://stampede.cs.berkeley.edu/pub/bcopy/bcopy-960524.tar.gz (yeah I know, haven't updated for ages) and look at the plot. (You'll need gnuplot to get a gif output.) You can see some other (old) results in http://now.cs.berkeley.edu/Td/bcopy.html Satoshi