> And scsi just works, at least in my experience. (As long
> as cabling is within spec, at least.) IDE users seems to
> have some "don't use this disk with that chipset at
> those dma speeds" issues that I never hear about with scsi.
> And such issues comes up over and over everytime there is a
> new IDE standard. New scsi standards seems to be less trouble,
> you automatically get the speed of the slower device without
> those spectacular hangs.
>
> Helge Hafting
I've noticed the absolute opposite. IDE is the one that usually just works
or doesn't work. And even when it doesn't work that can usually be traced
down to a bad cable, a very unusual event itself. And in this respect SCSI
itself seams to be much more sesitive to cable problems. I have encountered
a lot of bad SCSI cable so far and only one bad IDE one (obvious at that as
it was torn) despite all the computers that have passed by me, which also
were overwelmingly IDE. Because of this you are practically force to go out
and buy supossedly super high quality SCSI cables that can go for around $50
in order to avoid problems. Then you also have the extra complexity of SCSI
which adds to the areas in which possible errors can arise termination which
depends on how your SCSI bus is setup, 15 ID number selections, crappy
connectors (especially SCSI 2), incompatable bus types (i.e. LVD vs
UltraWide), etc. IDE because of its much simplier design usually limits
most of these configuration problems down to one jumper either master or
slave (and even that should not be relevent on some machines that use cable
select, though I have yet to see one of these myself).
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Jun 07 2000 - 21:00:25 EST