Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!rutgers!ames!orion!medin
From: medin@orion.arpa (Milo S. Medin)
Newsgroups: comp.dcom.lans
Subject: Re: Looking for DDN/X.25 to TCP/IP router
Message-ID: <168@ames.UUCP>
Date: Fri, 16-Jan-87 02:33:34 EST
Article-I.D.: ames.168
Posted: Fri Jan 16 02:33:34 1987
Date-Received: Fri, 16-Jan-87 21:49:38 EST
References: <2885@gitpyr.gatech.EDU> <1504@cit-vax.Caltech.Edu>
Sender: usenet@ames.UUCP
Reply-To: medin@orion.UUCP (Milo S. Medin)
Organization: NASA Ames Research Center, Moffett Field, Calif.
Lines: 43
Keywords: DDN Arpa X.25 TCP/IP router

In article <1504@cit-vax.Caltech.Edu> mangler@cit-vax.Caltech.Edu (System Mangler) writes:

>Has anyone gotten *anything* to work with an X.25 port on
>a PSN (aka IMP)?  Some local folks have been trying to get
>their X.25 ARPAnet connection going for some months now,
>so I keep getting the impression that the X.25 stuff is
>not exactly debugged yet.
>
>Don Speck   speck@vlsi.caltech.edu  {seismo,rutgers,ames}!cit-vax!speck

Don, you're exactly right.  If you are running 4.3 and the ACC 
DDN X.25 board, you can get a relatively fixed driver from
Mike Petry (petry@trantor.umd.edu).  It's really new as he made
a breakthrough recently.  Note the ACC driver on the 4.3 tape has
some serious problems with it.  As does the TWG code for VMS no
doubt.  

My feelings are that for local connections to the IMP (PSN), 
1822DH is far superior if you can get an interface from
ACC for your system.  A straight 1822DH connection will run
twice as fast as a comparable X.25 connection simply because 
the 1822 can signal at about 100-200 Kb/s and you can't connect X.25
to an IMP faster than 64 Kb/s.  That may not be 
important for a host with a few connects open to other DDN
hosts, but for a gateway which may be talking to a very large 
set of other hosts (or gateways) on the network it can make
a big difference.  I see no technical reason to inflict X.25 
and HDLC on a 20 ft cable going between a gateway and the IMP.

Also, the DDN X.25 spec is very complicated, as are most of the
X.25 interface boards that talk to the IMP.  It takes great care
and skill to write a good driver for these beasts.  Sooner or later
we'll no doubt be forced to convert, but that won't be for a 
couple years at least, and by then we may have more stable drivers
and faster signalling to overcome the current problems of the 
system.  Currently there are a lot of hosts on MILNET that use
X.25 interfaces, but about 94% don't pass IP with them --  they
use DDN BASIC X.25 service (not Standard).

All the usual disclaimers apply...

			Milo "I love 1822" Medin
			NASA Ames/ Bendix Aerospace