Aucbvax.2917
fa.unix-wizards
utzoo!decvax!ucbvax!unix-wizards
Wed Sep  2 21:57:50 1981
Unibus problems on Vaxes
>From Csk@UCLA-SECURITY Wed Sep  2 19:04:19 1981
One of the messages to this group was about problems with a controller
on the 11/750 unibus.  It turns out that there are all sorts of
strange timing and other weirdnesses that I have heard rumored
about on the 11/750 (and some on the 780) which all the
plug compatible unibus companies are struggling with.
They are quickly discovering that things that worked on 11's, don't
work on 11/44s and don't work even worse on 11/780s and 750s.

Does anyone out there know about these weirdnesses and can describe
how an 11/750 unibus differs from a normal "DEC standard" unibus
as described in dec published documents?  At UCLA we will be
connecting all kinds of weird stuff to our vaxen (some
home brew) and would like to avoid problems if we can.

Here are some of the problems that I have heard about:

1.  Unlike DEC's published spec. which claims that a slave device
can place data and slave sync on the bus at the same time (the
master is supposed to include all delays necessary), apparently on the
VAX 11/750 a device had better delay slave sync approx 40 nanoseconds.
Dec accomplishes this on all new unibus peripherals
by putting an RC delay before the 8881 chip which
drives slave sync, (and by using an 8881 rather than an 8640, and the
8881 is a slower chip!).

2.  I have heard rumors that many old dec devices don't work for this
and other reasons, but have not been able to get the reasons clarified

3.  I have heard rumors that a master device had better remove master
sync "quickly" after seeing slave sync, which is not required by the
dec spec.

Anybody out there know about this stuff?

--charley
-------

-----------------------------------------------------------------
gopher://quux.org/ conversion by John Goerzen <[email protected]>
of http://communication.ucsd.edu/A-News/


This Usenet Oldnews Archive
article may be copied and distributed freely, provided:

1. There is no money collected for the text(s) of the articles.

2. The following notice remains appended to each copy:

The Usenet Oldnews Archive: Compilation Copyright (C) 1981, 1996
Bruce Jones, Henry Spencer, David Wiseman.