Path: utzoo!attcan!uunet!husc6!uwvax!dogie!uwmcsd1!ig!agate!ucbvax!TRINITY.RICE.EDU!farrell
From: farrell@TRINITY.RICE.EDU (Farrell Gerbode)
Newsgroups: comp.protocols.ibm
Subject: Re: SUNLINK Users ??
Message-ID: <8805181647.AA02093@jade.berkeley.edu>
Date: 18 May 88 14:51:35 GMT
Sender: daemon@ucbvax.BERKELEY.EDU
Reply-To: Farrell Gerbode <farrell@trinity.rice.edu>
Organization: The Internet
Lines: 32


    We were using an AS9000, which is alleged to be IBM-compatible.
    Apparently there are slight differences in the channels, and Sun's
    channel attachment uses features in a slightly unusual way.  So if
    you are considering this product, and you aren't using a box actually
    made by IBM, make sure that your arrangement with Sun allows for
    the possibility that there might be problems.

Perhaps you should also bring this problem to NAS' attention as an
 incompatibility
of their product with IBM's?

Prior to our acquisition of an AS/9000 in 1982, we had another NAS product,
 namely,
an AS/6.  When we were unable to successfully install an IBM Series/1 channel
attach on the AS/6's channels, we pointed out the incompatibility of that
 machine's
channels with IBM's due to the Series/1 channel attach's violation of IBM's OEMI
Channel Specification during the inital selection sequence.  (Hmmm..perhaps
the IBM Series/1 was not really IBM compatible either! :-) NAS and Hitachi,
the manufacturer of both the AS/6 and AS/9000, devised and tested a hardware
modification to SLOW DOWN the initial selection sequence to that of an IBM 3033
and Hitachi flew three hardware people in from Tokyo to install it on our
 machine.

We found that NAS took very seriously our claim of their incompatibility with
IBM even in what was, at the time, a rather unique situation.  (Of course, the
Series/1 became MUCH more popular within a couple of years.)  The AS/9000 worked
with the Series/1 without modification.

Farrell Gerbode
Rice University