From owner-freebsd-chat Mon Sep 8 00:29:08 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id AAA15253 for chat-outgoing; Mon, 8 Sep 1997 00:29:08 -0700 (PDT) Received: from wakky.dyn.ml.org (lee@1Cust72.tnt1.manassas.va.da.uu.net [153.37.113.72]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id AAA15247 for ; Mon, 8 Sep 1997 00:29:03 -0700 (PDT) Received: (from lee@localhost) by wakky.dyn.ml.org (8.8.5/8.8.3) id DAA06240; Mon, 8 Sep 1997 03:28:33 -0400 (EDT) Message-ID: <19970908032832.44490@wakky.dyn.ml.org> Date: Mon, 8 Sep 1997 03:28:32 -0400 From: Lee Cremeans To: Mike Smith Cc: chat@freebsd.org Subject: Re: lousy disk perf. under cpu load (was IDE vs SCSI) Reply-To: hcremean@vt.edu References: <19970908024325.42427@wakky.dyn.ml.org> <199709080628.QAA01795@word.smith.net.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.79e In-Reply-To: <199709080628.QAA01795@word.smith.net.au>; from Mike Smith on Mon, Sep 08, 1997 at 04:28:40PM +1000 X-OS: FreeBSD 2.2-RELEASE X-Evil: microsoft.com Sender: owner-freebsd-chat@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > > Makes sense...I had an old (1985) Seagate ST-4026 that used linear seek, and > > the magnets in it were HUGE. There's also a Hitachi DK511-8 here, of about > > the same vintage, that uses rotary voice-coil. > > I was talking pre-1980 stuff; you would have to find a real fossil to > tell you what they were using before that. 8) Heh...fwiw, the oldest HD I've ever seen was from 1982, a stepper-based IMI. > > > > Yep...most stepper-motor drives I've seen used rack-and-pinion and > > sector-and-pinion linkages, joined together with bands or gears--about the > > same as a floppy drive. How they got those things to track accurately is > > beyond me, though. > > That's what guardbands are for; you use a bias current to pull the > stepper slightly off centre until you find one band, then push it the > other way to find the other. Center between the two is the middle of > the track. 8) Hmmm...I didn't know you could do that with a stepper motor, but then I'm no stepper motor expert :) > Lots of disks use the spindle to power the drive in spindown; listen to > any of the 5.25 CDC/Imprimis/Seagate disks winding down; as the power > finally drops off the spindle brake comes on. I presume they did this > to minimise the wear on the brake pad. Yup...we have three of those old-style CDCs here, and they all do that. They don't have brake pads though, from what I've seen; it looks like they just short the phases together, making an eddy-current brake. The click you hear when the brake engages is a relay on the motor-drive board, and I would imagine the braking system has something to do with the drive transistors. They may well have delayed it to keep the current from frying something, especially in the full-height drives. > Head retraction is pretty common too; naturally on removables it's > mandatory 8) Indeed; the first drive I saw that did that was a 44MB SyQuest (which is sitting in my closet, awaiting a decent SCSI card...). -- Lee C. -- Manassas, VA, USA (WakkyMouse on DALnet #watertower) A! JW223 YWD++^i WK+++r P&B++ SL++^i GDF B&M KK--i MD+++i P++ I++++ Did $++ E5/10/70/3c/73ac Ee34/1/36 H2 PonPippi Ay77 M | hcremean (at) vt.edu FreeBSD/Linux/Unix hacker...Win95 and M$ evil! (go see www.freebsd.org) My home page: http://wakky.dyn.ml.org/~lee | finger me for geek code