>In fact, IDE controllers have become so dirt-cheap that anyone with
>two IDE drives may want to look into putting the second drive on its
>own controller.  The speed increase should more than make up for the
>loss of one free IRQ.  YMMV, though.

Yes, I had thought of this also.  If you're trying to breath a few
more years life into a couple old IDE drives, this may be a very cheap
performance enhancement.

But, IDE still uses programmed-I/O (busy loops), tying up the
processor for all I/O (unless there have been enhancements for
EIDE/ATAPI that I don't know about).  As opposed to SCSI, which allows
the controller to do all the I/O independently of the CPU (at least on
most of the high-end boards supported by NetBSD -- obviously this
doesn't include dumb SCSI boards like the AHA-1522).

So, for new equipment you're still better off getting quality SCSI
components than IDE on multiple controllers.  But, for existing IDE on
a small budget, two IDE controllers would probably buy you a bit of
enhanced performance for very little money.

-----------------------------------------------------------------------------
  Michael L. VanLoon                                 michaelv@HeadCandy.com
       --<  Free your mind and your machine -- NetBSD free un*x  >--
     NetBSD working ports: 386+PC, Mac 68k, Amiga, HP300, Sun3, Sun4,
                           DEC PMAX (MIPS), DEC Alpha, PC532
     NetBSD ports in progress: VAX, Atari 68k, others...
-----------------------------------------------------------------------------