From owner-freebsd-questions@FreeBSD.ORG Fri Mar 20 23:53:20 2009 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 434F3106566C for ; Fri, 20 Mar 2009 23:53:20 +0000 (UTC) (envelope-from ghirai@ghirai.com) Received: from ghirai.com (ghirai.com [195.74.52.87]) by mx1.freebsd.org (Postfix) with ESMTP id 0F8068FC15 for ; Fri, 20 Mar 2009 23:53:20 +0000 (UTC) (envelope-from ghirai@ghirai.com) Received: from localhost (localhost [127.0.0.1]) by ghirai.com (Postfix) with ESMTPSA id 5E8C816FC3 for ; Fri, 20 Mar 2009 23:53:13 +0000 (GMT) Date: Sat, 21 Mar 2009 01:53:16 +0200 From: Ghirai To: freebsd-questions@freebsd.org Message-Id: <20090321015316.87f9531b.ghirai@ghirai.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Subject: speed in extracting rar files - unrar vs. 7z X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 20 Mar 2009 23:53:20 -0000 Hey, Does anyone know a faster way of extracting big rar files, or why is it so slow? I did a quick test with time (extracting the same ~800MiB file, consisting of split archives): unrar: real 4m29.637s user 0m4.969s sys 0m3.131s 7z: real 3m50.020s user 0m4.784s sys 0m1.821s While not a very good test, as i was having other apps idling around at the time, i did notice that neither of them were fully utilizing the CPU (Core2Duo E6550, clocked at 2.9GHz). Usage was ~8% at most. As you can see, 7z is marginally faster. The shareware WinRAR on windows seems to be better implemented (?), as it uses both cores to the fullest, and as such the time needed to extract stuff is a lot shorter. If this an implementation problem in unrar/7z, or is it the scheduler's 'fault' so to speak? Or does it have something to do with disk IO? If so, how do i check, and how can it be improved? Looking at top again, while extracting, showed both apps' state to be getblk most of the time, if it matters. I'm running 7.1-RELEASE, i386, generic kernel. Any ideas/thoughts? -- Regards, Ghirai.