This message is from the T13 list server.
On Sat, 9 Feb 2002 21:53:32 +0200, Dimiter Popoff wrote: >This message is from the T13 list server. >The original CD and IO are SCSI (or rather, SASI...) signals, >solving more or less the same "somewhat serious" problem. >Indeed over ATA CD and IO are lost sometimes when really needed, >so moving them at the other end of the cable, where the >host is - as you propose - makes good sense. Finding a way to >have a coherent set of copies at each side of the cable (like we have >the CD and IO signals in SCSI) would do an even better job. The ATAPI IO and CD bits are nothing but "extra" and "unnecessary" information. On many ATAPI devices they are just "random" data that has no usefulness. In the PACKET command protocol they are completely unnecessary. The BSY and DRQ bits are the only information needed. Maybe SCSI needs the equivalent of IO and CD but having these bits in the ATAPI PACKET command protocol was a BIG mistake. They are nothing but confusion. Do you have any idea how many times I have told people to fix their host software so that it ignores these bits and use BSY and DRQ instead? Do you have any idea how many times this has fixed host software that did not work correctly? *** Hale Landis *** www.ata-atapi.com *** Subscribe/Unsubscribe instructions can be found at www.t13.org.
