Path: utzoo!utgpu!watserv1!watdragon!rose!ccplumb From: ccplumb@rose.waterloo.edu (Colin Plumb) Newsgroups: comp.sys.amiga.tech Subject: Re: AmigaDos directory knowledge Message-ID: <18978@watdragon.waterloo.edu> Date: 4 Dec 89 07:22:06 GMT References: <88.filbo@gorn.santa-cruz.ca.us> <8742@cbmvax.UUCP> <3459@jhunix.HCF.JHU.EDU> <5615@cps3xx.UUCP> <5141@nigel.udel.EDU> Sender: daemon@watdragon.waterloo.edu Reply-To: ccplumb@rose.waterloo.edu (Colin Plumb) Organization: U. of Waterloo, Ontario Lines: 19 In article <5141@nigel.udel.EDU> new@udel.edu (Darren New) writes: > Something else I don't understand: > The disk should be able to read about 5 tracks/second if you look only > at the revolutions. Yet when doing something like DiskCopy, the disk > takes more than one second/track to read or write. What's the slowdown? > Is the blitter and system overhead really four times the read speed? First of all, you have to read about 1.1 revolutions to be sure to get a complete copy of each sector. Then remember that you need to read both top and bottom surfaces, and write the worst-case (slowest rotation speed) track, which is a bit under 1.1 nominal tracks. That adds up to 4.4 rotations to copy a cylinder, of which a 90mm floppy has 80. 70.4 seconds. The fastest copy program I know takes 69 seconds, so it's pretty close. It, I believe, reads one track while processing the track before so it's always reading or writing. This hides essentially all the blitter overhead. But Diskcopy isn't quite as smart. -- -Colin