Talk:Seek time

Terminology
Wouldn't this term also apply to the time it takes your CD player to arrive at the next track when you skip tracks? Or no? --KQ

Yes, it probably would, but I've never heard the term used for music CD players. Makes sense though. I've just realised that it isn't hard disk specific anyway. Certainly people use it when talking about computer CD players and floppy disks. Oops. --drj
 * CD's dont have fixed tracks, but have a format very much like an old style vinyl record. That means that the transition needs to be along the path of the track, and so the seek time is mostly the fastest speed that the disk can be fast forwarded --mike dill

Ok then, thanks. Should we mention something about the seek time on Zip Disks, or is that playing too rough? ;-) --KQ

Er, (music) CDs do only have one track true, but that's not how they seek a random music track. They do it more or less like a conventional disk drive, they drive the head in or out, crosssing the spiral track as it goes. It'll try and guess where to stop the head (it's a little bit difficult because different mastering processes give the spiral a slightly different curve) and then it'll read where it has stopped and jump forward or backward in smaller steps until it has found where it wants to be. This effect was easily noticable by listening to the worm gear on my friend's portable CD player. One long burst followed by a couple of very short ones. If it were to fast forward the CD until it got to where it wanted to go it would need an increase in rotation speed of about 1000 (to skip from the beginning of a CD to the last track in only a few seconds, say). This certainly doesn't happen (your CD player would need to consume more power than your kettle to do that, to say nothing of the noise). --drj

Solid state media?
Article currently reads:


 * Solid state media can have seek times of under 0.1 ms because there are no moving parts to align.

Surely solid state media have zero seek time? JulesH 10:37, 15 September 2006 (UTC)


 * Yes. There may be a negligible delay for implementation reasons, but I doubt it's measurable compared to the throughput. -- intgr [talk] 17:02, 9 September 2007 (UTC)


 * It's around 0.1 ms, but I'd guess it would be access time rather than seek time. -- Ddxc (talk) 11:57, 14 January 2008 (UTC)

Seek time vs. access time
I think the distinction between seek time and access time isn't made clear enough in this article. (The access time article could use some improvement as well, by the way.) See also (somewhat unreliable, but perhaps still useful),  and. (I'm not enough of an expert on this to improve it myself, unfortunately.) -- Ddxc (talk) 11:52, 14 January 2008 (UTC)

no defenition
71.156.42.169 (talk) 00:53, 6 February 2008 (UTC)what is wrong cant you find a defenition for seeking

Merge
I see no reason to merge the Seek time article into the Rotational delay article or vice versa. One is not part of the other; rather they are two of the three components of Access time. Jim.henderson (talk) 12:05, 25 May 2008 (UTC)

Please add a graph !
I'm under the impression that seek times have not decreased significantly during the last two decades and that they have increasingly become a bottle neck in modern computers. A graph will illustrate that nicely. -- Nic Roets (talk) 17:33, 2 December 2010 (UTC)