3 Sep
2003
3 Sep
'03
3:34 a.m.
Decserver 100's are terminal servers for going from terminals that use
3 wires (rcv,xmt, gnd) to ethernet. The connection is for an AUI, which
normally goes to thick-wire ethernet, but can adapt to normal ethernet.
The biggest problem, as I recall, is they only speak LAT, so if you want
to hook up terminals to something that doesn't run DECNET/LAT you have
a tough time. Of course, us VMS guys don't care.
That's what I remember. Anybody else, please correct me if the memory
is wrong...
Joe Heck
4 Sep
4 Sep
12:03 a.m.
TRASH3(a)splab.cas.neu.edu wrote:
Decserver 100's are terminal servers for going
from terminals that use
3 wires (rcv,xmt, gnd) to ethernet. The connection is for an AUI, which
normally goes to thick-wire ethernet, but can adapt to normal ethernet.
The biggest problem, as I recall, is they only speak LAT, so if you want
to hook up terminals to something that doesn't run DECNET/LAT you have
a tough time. Of course, us VMS guys don't care.
That's what I remember. Anybody else, please correct me if the memory
is wrong...
Joe Heck
No reason for a correction: it was indeed DECnet and LAT only. Somebody
please fill me in, but I think the DECserver 300 was the first to talk
Telnet, starting at firmware V2.0?
Except, perhaps, that thickwire was the "normal Ethernet" at those times.
Unlike some later / third-party models, the original DECservers always
had to download their operating system from a host on the network.
Commonly, the host was a VAX or PDP-11, but software for other machines
(including DOS PCs, provided they were running DECnet-DOS) existed. If
you turn power on without a load host on your network, you may,
depending on local settings on the DECserver, see messages on the
console port (any of the eight terminal ports) reporting "load failures".
DECnet (MOP, to be more precise) was used for software download and
upline dumps (the latter was rare, I remember having excellent stability
and very long uptimes with all DECservers except the MUXserver 100)
exclusively, while LAT was used for the actual sessions (connections
from terminal ports to hosts on the net). LAT is *not*
part of DECnet
(which is often claimed), it's a separate protocol. However, it
was
common to run LAT on a host that also would run DECnet.
The DECserver 100 hardware is, BTW, based on a Motorola 68000, and there
is even a low-level monitor program included in the load software, so
you can retrieve hex dumps from the DECserver's memory etc.
Hope that helps. In case you don't want your unit any more, throw it at
me! ;-)
Andreas
--
Andreas Freiherr
Vishay Semiconductor GmbH, Heilbronn, Germany
http://www.vishay.com
8:21 a.m.
No reason for a correction: it was indeed DECnet and
LAT
BTW: Someone else said that DS100 was the DEC first terminal
server. But the DECSA (PDP-11/24 based) preceded it. (I think
that was code named Pluto; if it wasn't, then there must have
been an even earlier one).
only. Somebody please fill me in, but I think the
DECserver 300 was
the
first to talk Telnet, starting at firmware V2.0?
The DS100 and DS200 didn't do telnet, at least none o fmine ever did
and I think I had the latest download code for both of them. The
DS300 did, and you may be right about V2.0 firmware being required.
Except, perhaps, that thickwire was the "normal
Ethernet" at
those times.
Thickwire was common, but Thinwire was certainly on the up and up.
Even thickwire tended to be mostly a big yellow snake in the ceiling
dropping down to racks with DELNIs and then AUI to everything in the
rack.
DECnet (MOP, to be more precise) was used for software
download and
from terminal ports to hosts on the net). LAT is *not*
part of DECnet
(which is often claimed), it's a separate protocol. However, it was
common to run LAT on a host that also would run DECnet.
MOP isn't part of DECnet either.
The DECserver 100 hardware is, BTW, based on a
Motorola
68000, and there
is even a low-level monitor program included in the load software, so
you can retrieve hex dumps from the DECserver's memory etc.
There used to be a tech manual for one of (or maybe both of) the
DS100 or DS200 with enough detail that you could write your own
firmware for it. I've never seen that manual, but it would be
nice if it magically appeared and someone scanned it for us!
68000 and variants were used by many products produced by various
parts of DEC's Networking group on both sides of the Atlantic.
(The DECservers, the HUB stuff, many of the WAN host interfaces).
Antonio
--
---------------
Antonio Carlini arcarlini(a)iee.org
12:19 p.m.
BTW: Someone else said that DS100 was the DEC first
terminal
server. But the DECSA (PDP-11/24 based) preceded it. (I think
that was code named Pluto; if it wasn't, then there must have
I thinkl you're right, I remember seeing a code name on at least one of
the PCBs in my DECSA, but I don't feel liking climbing over a dozen other
machines to check right now :-)
The DECSA does a heck of a lot more than operate as a terminal server, of
course...
-tony
5:20 p.m.
On Thu, 4 Sep 2003, Antonio Carlini wrote:
68000 and variants were used by many products produced
by various
parts of DEC's Networking group on both sides of the Atlantic.
(The DECservers, the HUB stuff, many of the WAN host interfaces).
It was also used in various XMI and BI boards I've seen over the years for
non-networking purposes. Those tended to have a lot of DEC and not DEC
chips on them of many different types.
16 Sep
16 Sep
2 a.m.
On Thu, 4 Sep 2003, Antonio Carlini wrote:
BTW: Someone else said that DS100 was the DEC first
terminal
server.
That was me.
But the DECSA (PDP-11/24 based) preceded it.
(I think
that was code named Pluto; if it wasn't, then there must have
been an even earlier one).
Yes, they prototyped the DS100 "idea" on a
bare 11/24, which is
why the DS100/200/300/500 code still has the DE (DEUNA/DELUA)
driver (== Unibus eth. controller.) - the DS series have chips
that are DQ (DEQNA/DELQA) compatible. Still, the DECSA was not
intened for mass production [use], so, I'd say the DS100 was
the first "for mass use" DEC terminal server..
The DS100 and DS200 didn't do telnet, at least
none o fmine ever did
and I think I had the latest download code for both of them. The
DS300 did, and you may be right about V2.0 firmware being required.
Correct. DS100
and DS200 are LAT-only, DS300 V1.x was LAT-only, and
DS300 V2.x (and up) and others had IP support as well.
Thickwire was common, but Thinwire was certainly on
the up and up.
17 Sep
17 Sep
12:40 p.m.
Yea. I have seen the DS500 source code, and it looks
like that has
actually evolved from the earlier DS100 and DS200 systems.
This I find surprising since, like PLUTO, the DS500 was PDP-11 based
ans the DS100 and DS200 (and the 300 & too AFAIK) were 68000-based
(maybe moving to some 68K variant in the later lines). The DECserver
series came out of LKG (Littleton) - the folks that bounced
through the Cabletron acquisition and came out the other side
as DNPG - and they tended to use 68K variants wherever possible.
The UK side did use 68K variants for some of the synch adapters
but tended to use VAX processors for the network boxes (DECnis,
DEMSA).
Antonio
--
---------------
Antonio Carlini arcarlini(a)iee.org
3:29 p.m.
On Wed, 17 Sep 2003, Antonio Carlini wrote:
The DS200
sure is PDP-11 based. Dunno 'bout the 100, since I dont
have that one.
Fred
--
Fred N. van Kempen, DEC (Digital Equipment Corporation) Collector/Archivist
Visit the VAXlab Project at http://www.pdp11.nl/VAXlab/
Visit the Archives at http://www.pdp11.nl/
Email: waltje(a)pdp11.nl BUSSUM, THE NETHERLANDS / Sunnyvale, CA, USA
Yea. I have
seen the DS500 source code, and it looks like that has
actually evolved from the earlier DS100 and DS200 systems.
This I find surprising since, like PLUTO, the DS500 was PDP-11 based
ans the DS100 and DS200 (and the 300 & too AFAIK) were 68000-based 
3:56 p.m.
Antonio Carlini wrote:
This I find surprising since, like PLUTO, the DS500
was PDP-11 based
ans the DS100 and DS200 (and the 300 & too AFAIK) were 68000-based
"Fred N. van Kempen" <waltje(a)pdp11.nl> wrote:
The DS200 sure is PDP-11 based. Dunno 'bout the
100, since I dont
have that one.
Just had a DS200 open a few months ago, and there was a 68000 in it,
with no PDP-11 of any sort (not even a T11). Also, if you look at
the contents of the ROMs or the MOP download code, they are full of
68K code. For instance, there are lots of hex 4E75 words, which is a
68000 RTS instruction. Shows up as "Nu" in ASCII, or "uN" if your
byte
order is different.
4:10 p.m.
On Wed, 17 Sep 2003, Eric Smith wrote:
Just had a DS200 open a few months ago, and there was
a 68000 in it,
with no PDP-11 of any sort (not even a T11). Also, if you look at
the contents of the ROMs or the MOP download code, they are full of
68K code. For instance, there are lots of hex 4E75 words, which is a
68000 RTS instruction. Shows up as "Nu" in ASCII, or "uN" if your
byte
order is different.
Hrrm. Thats weird. I could've sworn they were F11 based.
Also,
they used RSX taskbuilder images. Guess they were using an 68K
cross-assembler setup then....
Cheers,
Fred
--
Fred N. van Kempen, DEC (Digital Equipment Corporation) Collector/Archivist
Visit the VAXlab Project at http://www.pdp11.nl/VAXlab/
Visit the Archives at http://www.pdp11.nl/
Email: waltje(a)pdp11.nl BUSSUM, THE NETHERLANDS / Sunnyvale, CA, USA
18 Sep
18 Sep
9:24 a.m.
Also, they used RSX taskbuilder images. Guess they
were
using an 68K cross-assembler setup then....
IIRC the last stage of producing a downloadable image for
the various DEMSA images involved sticking a header on
the front which the code referred to as RSX-style or some such.
Antonio
--
---------------
Antonio Carlini arcarlini(a)iee.org
19 Sep
19 Sep
4:52 a.m.
On Thursday 18 Sep 2003 12:47 am, Eric Smith wrote:
Antonio Carlini wrote:
I'll confirm this, both the ds200 and the ds300 have 68k's in 'em...
alex/melt
--
Dance like no-one's watching, sing like no-one can hear you, love like
you've never been hurt, wash like everyone has a nose, eat like
cholesterol is a myth.
This I find surprising since, like PLUTO, the
DS500 was PDP-11 based
ans the DS100 and DS200 (and the 300 & too AFAIK) were 68000-based
"Fred N. van Kempen" <waltje(a)pdp11.nl> wrote:
The DS200 sure is PDP-11 based. Dunno 'bout
the 100, since I dont
have that one.
Just had a DS200 open a few months ago, and there was a 68000 in it,
with no PDP-11 of any sort (not even a T11). Also, if you look at
the contents of the ROMs or the MOP download code, they are full of
68K code. For instance, there are lots of hex 4E75 words, which is a
68000 RTS instruction. Shows up as "Nu" in ASCII, or "uN" if your
byte
order is different.
7646
days inactive
7662
days old
11 comments
8 participants
participants (8)
-
Andreas.Freiherr＠Vishay.com -
arcarlini＠iee.org -
ard＠p850ug1.demon.co.uk -
eric＠brouhaha.com -
meltie＠meltie.com -
thompson＠new.rr.com -
TRASH3＠splab.cas.neu.edu -
waltje＠pdp11.nl