From owner-freebsd-chat Wed Apr 17 18:37:24 2002 Delivered-To: freebsd-chat@freebsd.org Received: from mired.org (dsl-64-192-6-133.telocity.com [64.192.6.133]) by hub.freebsd.org (Postfix) with SMTP id E18B237B400 for ; Wed, 17 Apr 2002 18:37:13 -0700 (PDT) Received: (qmail 618 invoked by uid 100); 18 Apr 2002 01:37:10 -0000 MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <15550.9030.396432.30948@guru.mired.org> Date: Wed, 17 Apr 2002 20:37:10 -0500 To: Giorgos Keramidas Cc: Bob Bomar , chat@FreeBSD.ORG Subject: RE: overclocking and freebsd In-Reply-To: <20020412144054.GB2610@hades.hell.gr> References: <20011110215343.C961@bsd.alexe.org> <20020411182041.H45395@darius.2y.net> <20020411200534.A25472@ns.museum.rain.com> <20020412042041.GA80748@peitho.fxp.org> <20020412144054.GB2610@hades.hell.gr> X-Mailer: VM 6.90 under 21.1 (patch 14) "Cuyahoga Valley" XEmacs Lucid X-face: "5Mnwy%?j>IIV\)A=):rjWL~NB2aH[}Yq8Z=u~vJ`"(,&SiLvbbz2W`;h9L,Yg`+vb1>RG% *h+%X^n0EZd>TM8_IB;a8F?(Fb"lw'IgCoyM.[Lg#r\ From: Mike Meyer X-Delivery-Agent: TMDA/0.51 (Python 2.2 on FreeBSD/i386) Sender: owner-freebsd-chat@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org In <20020412144054.GB2610@hades.hell.gr>, Giorgos Keramidas typed: > That's a highly subjective metric though. My FreeBSD machine feels a > lot more responsive than those Windows XP machines with faster CPUs a > and larger amounts of RAM I've seen friends work on. But how does one > define an objective metric of 'responsiveness'? Easy - to describe, anyway. First, get a large enough group of users for each system to be statistically significant, and watch them work for a week or so, noting all their activites and how many each one is done. Sort out the activities that are relevant to the question at hand, and see if the relative frequencies of them are reasonably close to each other. If they are, make up a sequence of events that hits those frequences. If not, make up three sequences, one for each group, and one for the two groups combined. Now measure how long it takes each member of each group to perform the sequence(s). After you've got the data, you can start running two-variable tests for the sequence(s), and see if there's a statistically significant difference. The tricky part will be when the timing is warped by the window manager. I.e., if I've got a WM set to follow the mouse and not raise the active window, then activating a window is one action for me (point), but two (point and click) for a windows users. On the other hand, raising a window is easy for a windows users, because they can point anywhere in the window, whereas I have to point at the frame. If I'm using a pointerless window manager, it starts getting really complicated. http://www.mired.org/home/mwm/ Independent WWW/Perforce/FreeBSD/Unix consultant, email for more information. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message