From owner-freebsd-hackers Sat Oct 10 17:02:49 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id RAA17691 for freebsd-hackers-outgoing; Sat, 10 Oct 1998 17:02:49 -0700 (PDT) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from ridge.spiritone.com (ridge.spiritone.com [205.139.108.2]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id RAA17668 for ; Sat, 10 Oct 1998 17:02:37 -0700 (PDT) (envelope-from dionn@spiritone.com) Received: from spiritone.com (us4b-121.spiritone.com [206.98.120.121]) by ridge.spiritone.com (8.8.8/8.8.8) with ESMTP id RAA32130; Sat, 10 Oct 1998 17:02:09 -0700 (PDT) Message-ID: <361FEDC9.613F95FC@spiritone.com> Date: Sat, 10 Oct 1998 16:29:13 -0700 From: Haze Organization: Organized? Me? Ha! X-Mailer: Mozilla 4.07 [en] (Win95; U) MIME-Version: 1.0 To: Alex CC: Luigi Rizzo , hackers@FreeBSD.ORG Subject: Re: how fast are "fast" CDROM drives ? References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On 10-Oct-98 @ 21:46:29, Alex wrote: > > On Sat, 10 Oct 1998, Luigi Rizzo wrote: > [...] > > I got a 32x ATAPI unit one month ago and there is no way (under > > -stable) i can get more than 2.2MB/s by dd'ing from the disk. > > Considering i get 5-6MB/s from a raw IDE disk (and from iozone with > > large files) on the same machine, i don't think the limiting factor is > > CPU or the OS. > > > > I wonder if anybody is able to exploit the speed of their 24/32/40x > > IDE or SCSI drives > > Well, some of the earlier high speed CDs 12x+ cheat a bit by rotating the > disk faster only for certian parts of the disc (the outter edges IIRC). > The most inner part of the disc would then be read at somewhere around 6x. Drives over 16x are CAV (most are anyway). This means that the listed speed of 32x can only be achieved at the outermost tracks. If you filled a CD completely, then read from the very last files you would get 32x throughput. Consequently, you almost never see it IRW situations. CAV drives will xfer the outermost tracks at about 2.2x the rate of the innermost tracks. So your 32x will get about 14x on the inside tracks. A throughput of 2.2MB/s (~15x) is quite typical. The best I've ever seen on my 32x drives is ~3.7MB/s. -- Knowledge is the perception of truth distorted by reality PGP Key: www.spiritone.com/~dionn/pgpkey Fingerprint: 312E F076 35D0 8EEB 4B2A DC8A 04FC 1543 3B2C 0502 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message