From owner-freebsd-hackers Wed Oct 29 06:52:33 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id GAA19833 for hackers-outgoing; Wed, 29 Oct 1997 06:52:33 -0800 (PST) (envelope-from owner-freebsd-hackers) Received: from word.smith.net.au (ppp20.portal.net.au [202.12.71.120]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id GAA19824 for ; Wed, 29 Oct 1997 06:52:24 -0800 (PST) (envelope-from mike@word.smith.net.au) Received: from word.smith.net.au (localhost [127.0.0.1]) by word.smith.net.au (8.8.7/8.8.5) with ESMTP id BAA01029; Thu, 30 Oct 1997 01:18:15 +1030 (CST) Message-Id: <199710291448.BAA01029@word.smith.net.au> X-Mailer: exmh version 2.0zeta 7/24/97 To: Mikael Karpberg cc: dg@root.com, hackers@FreeBSD.ORG Subject: Re: Parity Ram In-reply-to: Your message of "Wed, 29 Oct 1997 03:29:12 BST." <199710290229.DAA07708@ocean.campus.luth.se> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 30 Oct 1997 01:18:14 +1030 From: Mike Smith Sender: owner-freebsd-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > Hmm... It's still not quite clear to me. That is, does this slow my > computer down, in case I use ECC? Slightly, yes. Unless you stand over it measuring things microscopically, you won't notice the difference. > It seems to me all this could be done on the DIMM/SIMM, or something, > possibly clocked at multiple of the bus clockspeed, and therefor > not effect the rate at which memory could be read/written over the bus > by the CPU. It's faster and cheaper to do it inside the memory pipeline. > If that's not the case, and the computer is actually slowed down by ECC, > how much performace do you loose? 0.1%? 5%? 30%? That depends; what's your system doing? The tests we did put the change down in the noise (<2%) for our workload. mike