Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!rutgers!ucsd!ames!lll-winken!uunet!ncrlnk!ncrwic!encad!entec!jlohmeye
From: jlohmeye@entec.Wichita.NCR.COM (John Lohmeyer)
Newsgroups: comp.periphs
Subject: Re: SCSI & Q-bus
Message-ID: <325@entec.Wichita.NCR.COM>
Date: 25 Mar 89 03:53:43 GMT
References: <834@mplvax.EDU>
Reply-To: jlohmeye@entec.UUCP (John Lohmeyer)
Distribution: usa
Organization: NCR Corporation, Wichita, Kansas
Lines: 27

In article <834@mplvax.EDU> cdl@mplvax.ucsd.edu.UUCP (Carl Lowenstein) writes:
>  [stuff deleted]
>1) Can the DTC 11-2, which is fundamentally a SASI host adapter, be
>used to control a SCSI device (disk, in particular). 
>
>2) Can any SASI host adapter which can assert a variable device ID
>control more than one SCSI device, if one is careful to avoid
>overlapping operations?

I would answer a qualified yes to both questions.  You won't get great
performance because you will have to be careful not to overlap I/O, but
it _should_ work.  I am not familiar with the DTC 11-2, but if it is
really old, it probably has the following characteristics:

 No arbitration
 Doesn't drive the ATN signal
 Doesn't generate or check parity
 Only provides the target ID during selection (does not include its own)

All of these things were permitted in SCSI-1, but are now disallowed in SCSI-2.
Unless you have a very new disk that is religious about these things, it
probably will work with the DTC host adapter.  One caveat: many vendors do
not do a lot of testing for compatibility with SASI. So, you may find bugs
in how they handle the lack of these features.

Good luck,
John Lohmeyer  j.lohmeyer@wichita.ncr.com