MicroVax TCP/IP ?
Douglas Robinson
dbr at cybvax0.UUCP
Wed Nov 7 04:40:49 AEST 1984
Ok, for the full scoop (as much as I can tell):
1. Unisource is a company, NOT a piece of software.
It has NOTHING to do with the MicroVAX-I (uVAX-I).
2. ULTRIX is DEC's version of 4.2BSD (NOTE: They track
USENET and fix errors in code! No more maintenance
problems.)
3. ULTRIX for the uVAX-I has just been released by
DEC and is known as: 'ULTRIX-32m rel. 1.0'.
4. This is a non-changeable (unless you get source) linked
kernel with all drivers that DEC supports for the
uVAX-I configuration. A relinkable object file distribution
is in the works (a la ULTRIX-32 for the larger VAXen), but
won't be available for rel. 1.0.
5. Adding drivers is a non-trivial problem! The memory
map registers that you find on all larger VAX's are
NOT on uVAX-I so all devices are required to be MSCP
(so the driver can do scatter/gather for the machine).
6. ULTRIX-32m WORKS WELL! I have personnally MOVED
(NOT RECOMPILED, BUT COPIED) executables from an 11/750
running 4.2 BSD to this O/S and had them run without
problem! (5Mb in all!)
7. uVAX-I is about the same speed as the 11/73. It excells over
the 11/73 when overlays would be necessary and where a lot
of floating point is used (I have comparison benchmarks that
could be posted if enough desire becomes apparent). It is
a little slower where 16 bit integers are involved.
8. I am in the process of obtaining enough hardware to Ethernet
the uVAX-I and our 11/750 together. This will be via 4.2BSD/
ULTRIX standard utilities using the DEQNA (Ethernet for
Q-Bus(22bit)) and the DEUNA (Ethernet for UNIBUS). If desired,
I can post the results as well.
I hope this clears up some of the confusion.
--
Doug Robinson Jobs don't kill programmers... programmers kill jobs!
Cybermation, Inc. 617/492-8810
377 Putnam Avenue
Cambridge, MA 02139 (USA)
...!{mit-eddie, harvard, mirror}!cybvax0!dbr
More information about the Comp.os.eunice
mailing list