[8698] in linux-scsi channel archive
Re: NCR53C8XX
daemon@ATHENA.MIT.EDU (Mark Hahn)
Sat Apr 22 14:40:34 2000
Date: Sat, 22 Apr 2000 14:37:59 -0400 (EDT)
From: Mark Hahn <hahn@coffee.psychology.mcmaster.ca>
To: Michael Gerdts <gerdts@cae.wisc.edu>
Cc: linux-scsi@vger.rutgers.edu
In-Reply-To: <20000421171405.B17599@cae.wisc.edu>
Message-ID: <Pine.LNX.4.10.10004221428150.13447-100000@coffee.psychology.mcmaster.ca>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
> > > > lowest device (ie: if you put a 10MB/sec CDROM on an 80MB/sec
when people say "80 MB/s", they usually mean Ultra2,
which is LVD signalling, at 40 MHz and 16b-wide.
Ultra2 has a special provision for supporting old (non-LVD actually)
devices, which _does_ remove most of the benefits of Ultra2,
including cable length, bandwidth and probably even CRC's.
in other words, don't put sub-Ultra2 devices on your Ultra2 chain.
> speeds. I have an NCR53C810 controller, a ~1 gig SCSI II hard drive, and
> an ancient Sony DDS II (aka DAT) drive.
this is pretty much a different issue: a normal single-ended chain
that has, say, a disk that wants to do ultra (20 MHz) transfers,
and a tape that wants to do slow (5 MHz) transfers will cooperate
just fine, since the speed of a transfer is negotiated between
host and target (independent of other targets).
in other words, SCSI compatibility within the SE domain (slow/fast/ultra,
narrow/wide) is a lot more transparent than the nasty sort of crippling
back-compatibility that Ultra2 offers.
regards, mark hahn.
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
