7 Feb
2008
7 Feb
'08
5:37 a.m.
On Feb 7, 2008 6:16 AM, <cctalk-request at classiccmp.org> wrote:
Message: 6
Date: Wed, 6 Feb 2008 19:39:17 +0000
From: Ethan Dicks <ethan.dicks at usap.gov>
Subject: Re: VAX-11730 (was Any word on OS/2 for PDP-11?)
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Message-ID: <20080206193917.GC17779 at usap.gov>
Content-Type: text/plain; charset=us-ascii
On Wed, Feb 06, 2008 at 08:12:11AM -0800, Bob Armstrong wrote:
780
To do it on the 11/780 you needed to putchase the optional second WCS
(Writable Control Store) board. The first WCS board was used for patches
for microcode bugs from DEC.
Ethan
Dicks wrote:
The 11/730 (and 11/725) has an on-board FEP - an
8085. You talk to
it much in the same way as you do to the LSI-11 in an 11/780, except
its console medium is TU58 tape, not RX01 disk. Syntactically, though,
I believe it's similar (I know the KA730 well, but not the KA780).
Indeed, the 730 CFE (Console Front End or FEP) was very much like the - the commands were similar and the 730 even used
indirect command files
for
functions like booting and power up loading of
microcode. All the
microstore on the 730 was RAM; the only ROM was a little EPROM that
booted
the 8085 operating system from the console TU58.
Everything else - all
the
CPU microcode, the microcode for the IDC
(integrated disk controller)
the
FPA microcode, and VMB - were all loaded from the
TU58. Although the
8085
operating system was some custom thing DEC wrote
themselves, the TU58s
used
an RT11 file system (again, just like the 780)
and you could easily
manipulate them with EXCHANGE or FILEX.
That all sounds familiar.
The bad news was that TU58s are really, really,
slow. With all the
microcode and indirect files that had to be read from the TU58, your 730
could easily take ten minutes from power on to the point where VMB was
even
ready to start thinking about loading VMS. When
working on one, you
want to
do everything you can to avoid turning off the
CPU box power:-)
Using the stock console tapes from DEC, that was absolutely true. One of
the scripts I wrote a while back, built a load-order optimized console
tape.
I think the 8085 must cache the directory, since the tape doesn't seek
back
to the directory blocks between each file. With the files in the right
order,
the file transfer time doesn't change, but the file-to-file time is just
about
nil (I think there may be one seek from end-to-end because of how many
files
there are to read and the block interleave on the tape).
Because everything about the KA730 microcode
was "soft", it's fairly
easy
to change the CPU microcode, update the console
TU58 and reboot to
change
the CPU behavior. I believe DEC even sold a set
of microprogramming
tools
for the 730, but I've never seen them and I
don't know what's become of
them
today.
Back in the day, I remember occasionally seeing docs on PDP-11
microcoding,
but I don't recall seeing anything for any model of VAX.
-ethan
--
Ethan Dicks, A-333-S Current South Pole Weather at 6-Feb-2008 at
19:20 Z
South Pole Station
PSC 468 Box 400 Temp -48.6 F (-44.8 C) Windchill -75.5 F (-59.7C)
APO AP 96598 Wind 8.2 kts Grid 50 Barometer 675.4 mb (10802
ft)
Ethan.Dicks at usap.gov
http://penguincentral.com/penguincentral.html
--
d|i|g|i|t|a|l had it THEN. Don't you wish you could still buy it now!
pechter-at-gmail.com