kahn@xanth.UUCP (10/30/87)
Does anyone know why floppy drive accesses are so slow? If the disk spins at 300 rpms, and a track has 8 sectors, it should only take 0.0222 seconds to read a sector. This translates to a maximum access rate of as high as something like 45K bytes/sec, plus any time required to switch tracks. The data bus, at million(s) of bytes per second, shouldn't be the bottleneck. Why is it necessary to go to a hard disk to get closer to the speeds which, in theory, the floppy can reach?
john@hpcvlo.HP.COM (John Eaton) (11/02/87)
<<< < Does anyone know why floppy drive accesses are so slow? If the disk spins < at 300 rpms, and a track has 8 sectors, it should only take 0.0222 seconds < to read a sector. ---------- To read a sector from a floppy: Turn on motor and wait 250-500 ms for it to come up to speed Step to correct track at 6ms per step and then wait 15 ms for head to settle Issue read command and wait an average of 100 ms for sector to come around Read sector in next 16.384 ms John Eaton !hplabs!hp-pcd!john
justin@inmet.UUCP (11/04/87)
/* Written 4:33 pm Nov 1, 1987 by john@hpcvlo.UUCP in inmet:comp.sys.ibm.pc */ <<< < Does anyone know why floppy drive accesses are so slow? If the disk spins < at 300 rpms, and a track has 8 sectors, it should only take 0.0222 seconds < to read a sector. ---------- To read a sector from a floppy: Turn on motor and wait 250-500 ms for it to come up to speed Step to correct track at 6ms per step and then wait 15 ms for head to settle Issue read command and wait an average of 100 ms for sector to come around Read sector in next 16.384 ms John Eaton !hplabs!hp-pcd!john /* End of text from inmet:comp.sys.ibm.pc */ Also, bear in mind that the sectors of a file are rarely contiguous, unless the disk is brand-new. Thus, you have to go through the entire above procedure for a large fraction of the sectors of the file. -- Justin du Coeur