Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!uunet!abvax!iccgcc!herrickd From: herrickd@iccgcc.decnet.ab.com Newsgroups: comp.arch Subject: Re: CD-ROM documents (was Paperless Office) Message-ID: <2304.275e10db@iccgcc.decnet.ab.com> Date: 6 Dec 90 14:35:23 GMT References: <11191@pt.cs.cmu.edu> <00940487.15804140@KING.ENG.UMD.EDU> <28083@mimsy.umd.edu> <1990Nov29.162726.11411@mozart.amd.com> <11212@charm.UUCP> <2974@crdos1.crd.ge.COM> <1990Dec3.220850.18352@watdragon.waterloo.edu> Followup-To: .uiuc.edu> Lines: 22 In article , aglew@crhc.uiuc.edu (Andy Glew) writes: >>important architectural issue: CDROMs have a seek time of on the order of >>500 ms - that's right, half a second. This means that the selection of > > Why are CD-ROM seeks so slow? Is it related to the spiral tracking? The audio Compact Disk spec includes constant linear speed along the spiral track that runs the entire length of the disk. The average seek includes moving a significant distance along the radius of the disk, requiring a noticable change in the rotational speed of the disk. That average change in rotational speed is the critical path factor in the half second seek time. Seeks that skip only a few revolutions of the spiral are typically done by rotating a mirror and happen quite fast (they don't require much change in the rotation speed of the disk, either). A half second between selections in an audio program is not at all unpleasant, the slow seek time was a good design decision, darn it. dan herrick herrickd@astro.pc.ab.com