[2045] in linux-scsi channel archive
Re: scsi-problem (phase change ?)
daemon@ATHENA.MIT.EDU (Hauke Johannknecht)
Sun Jun 22 14:35:30 1997
Date: Sun, 22 Jun 1997 20:31:39 +0200 (MET DST)
From: Hauke Johannknecht <ash@ash.ccc.de>
To: "Leonard N. Zubkoff" <lnz@dandelion.com>
cc: groudier@club-internet.fr, linux-kernel@vger.rutgers.edu,
linux-scsi@vger.rutgers.edu, ncr53c810@colorado.edu
In-Reply-To: <199706221734.KAA07468@dandelion.com>
On Sun, 22 Jun 1997, Leonard N. Zubkoff wrote:
> (1) If you run even normal Fast SCSI-2 (10.0 mega-transfers/second) on your
> SCSI bus, you should have *active* termination at both ends. Some
> configurations *might* get away with passive termination, but at the
> first sign of any problems you should immediately move to active
> termination.
smile. i remember it as a one end active for scsi-2,
both for long cables. ...
> (2) Ultra SCSI (20.0 mega-transfers/second) is even more stringent. If I
> recall correctly, you are allowed 3 meters cable length only if you have
> at most 4 devices, and 1.5 meters if you have more. The cables must also
> be built to much more stringent specifications. This doesn't apply to your
> configuration since the 53c810 isn't Ultra-capable, but it's here for
> reference.
ACK.
> With this configuration, I'd suggest it's time for a second SCSI bus.
hmm, ok, you convinced me. (not exactly you. i got this hint more
than one time ... :)
i will add my old FD TMC-950 for the Seagate ...
Gruss,
Hauke
--
Hauke Johannknecht 5johannk@informatik.uni-hamburg.de
-> Hamburg / Germany <- - tschechow@geocities.com
http://www.geocities.com/TimesSquare/Arcade/9242/ ash@ash.ccc.de
--> pgp-key via public-key-server and on request <--> use pgp ! <--
