On Thu, Jun 18, 2020 at 2:42 PM Peter Coghlan via cctalk
<cctalk at classiccmp.org> wrote:
Thanks for your reply Paul. My eventual goal is to be
able to use the
synchronous serial interface on a MicroVAX to connect to IBM machines that
only support bisync lines.
I'm curious which software package you are using. In the 80s and 90s,
I worked for Software Results Corp, and we sold hundreds (small
thousands?) of Unibus, Qbus, and VAXBI interfaces and software suites
under the name COMBOARD for Bisync and SNA comms from DEC-to-IBM
(RSTS, RSX-11, VMS, and UNIX). In my experience, people paid us
$10K-$25K per line because they had problems with whatever other
solution they were looking at that cost less. I don't know any
specific details on where the gaps were, but just that people did
experience bugs or missing features that made them come to us. I'd
like to hear about what you are encountering once you get to the point
of passing bytes.
However, I don't have access to any such IBM kit
at the moment so I have to make do with trying to get the MicroVAX to talk
to another instance of itself for now.
For bisync (3780/HASP) that should be just fine. There's a simple
difference in the protocol so that each end can tell who is the
master, and AFAIK, any implementation you encounter should be able to
be set in the appropriate mode, but you _do_ have to tell each end
what their role is or they will chatter endlessly if they both think
they are the master.
Apologies if you already know this. It's been a long time and I'm not
sure who remembers what details about obscure comms from 30-40 years
ago. I myself haven't even set up a bisync connection in 25 years. I
used to do this stuff every day, then by the mid-90s, it entirely
evaporated.
-ethan