Path: utzoo!attcan!darkover!telly!lethe!torsqnt!news-server.csri.toronto.edu!helios.physics.utoronto.ca!ists!yunexus!xrtll!silver From: silver@xrtll.uucp (Hi Ho Silver) Newsgroups: comp.sys.ibm.pc Subject: Re: Disk interleave. Keywords: interleave disk spinrite Message-ID: <1990Jun5.024638.24889@xrtll.uucp> Date: 5 Jun 90 02:46:38 GMT References: <4627@ursa.UUCP> Organization: Not around here, pal! Lines: 52 In article <4627@ursa.UUCP> jawitz@ursa.UUCP (Eric Jawitz) writes: $I'm curious to know what a 1:1 interleave disk controller card does for $performance. [...] $The talk about a 1:1 interleave disk controller puzzles me because if $the interleave can be changed with software, what's the need for a $special controller card? How is a 1:1 controller any different from $my (apparently) 3:1 controller? Should I consider getting one? I guess a brief, nontechnical overview of interleave is in order. In the past, the typical disk drive could provide information to the controller faster than the controller could ship it off to the PC, so by the time the controller was ready to receive the next sector from the disk drive, the read/write head had already gone past the next sector and you'd have to wait until the disk completed a revolution, which obviously is not good for performance. To remedy this, interleave factors are used so that the position the read/write head is in once the controller is ready for the next sector is about where the next sector in the file will be on the diskette. I hope that explains it well enough ... this is something that's much easier to explain with a diagram. So how is the 1:1 controller different? Well, it can handle the data quickly enough that interleave factors are not needed, and it allows you to use the disk drive to the limit of its performance. $On a related note, my system has a selectable bus speed (8/12 megahertz I $believe). I realize that some cards may not work at the higher speed. However $assuming that everything checks out OK, are there any drawbacks to increasing $the speed? Would the system run hotter? Would I be shortening the life $of my boards? Am I likely to see an increase in disk throughput? Is this $in turn likely to affect the optimum interleaving for my system? Just because everything checks out OK when you try it out for a minute or two doesn't mean it will always work OK ... if you find it works at the higher rate, don't be surprised if you start thinking there are gremlins in your computer from time to time, as some cards may be at the very limit of their performance and may work some of the time but not all of the time at the higher rate. If all your cards can handle it, though, you should have no problems. The life of your system may be shortened a little bit, but most systems will normally last beyond the time when you want to replace them with more recent technology anyway. Also, as long as you make sure everything doesn't get clogged up with dust, you should have enough airflow in most machines to keep things adequately cool. As for disk throughput, you will quite likely have to alter the interleave for optimal performance if you change the bus speed, since that alters the rate at which the controller can transfer data between itself and the PC. -- /Nikebo \ Nikebo says "Nikebo knows how to post. Just do it."\silver@xrtll/ /---------\_____________________________________________________\----------/ /yunexus!xrtll!silver (L, not 1)\ Hi Ho Silver \ just silver for short / /Silver: Ever Searching for SNTF \ Life sucks. \ someone buy me a BEER! /