Path: utzoo!utgpu!water!watmath!clyde!burl!codas!killer!academ!soma!masscomp-request
From: dalesys@lamont.columbia.edu (dale chayes)
Newsgroups: comp.sys.masscomp
Subject: Re: PC-Masscomp Ethernet
Summary: pc's on ethernet
Message-ID: <3337@soma.bcm.tmc.edu>
Date: 18 Feb 88 08:47:41 GMT
References: <3301@soma.bcm.tmc.edu>
Sender: masscomp@soma.bcm.tmc.edu
Lines: 34
Approved: masscomp@soma.bcm.tmc.edu

In article <3301@soma.bcm.tmc.edu>, (Stan Barber, Moderator) writes:
Bryan Costales, of   EEG Systems Laboratory asks about conecting PC-ATs
and Masscomps via ethernet.


We (at Lamont-Doherty) have been using Compaqs (portables and 286's) with
3Com 501 boards in our lab network and in our shipboard network.  We have
been running Sun's PC-NFS package as well as MIT's TCP (an old version)
sucessfully.  For file transfer, its great.

However:

I find TELNET uncomfortable as a terminal emulator because of
the block buffering enforced by the network protocol, and (since the 
distance is short) have both a serial line and a network connection 
for the compaq that is normal on my desk.
 
The older version of the MIT sofware (jan 85) has a timeout built into
their TELNET which disconnects and drives me wild.

I have just recently gotten a copy of the NCSA Telnet which I have not
had a chance to try out yet.

The big advantage of Sun's PC-NFS is that it WOULD allow you to use your
masscomp's disk (or parts of it) as if they were a part of your PC's file
system. IE: from DOS you could mount node:/usr/dale as drive F:. Now 
that is really a neat and in may cases useful feature. Unfortunately,
it looks like Masscomp is not going to support the pcnfs daemon in the
forthcomming release of RTU and NFS. Perhaps someday that will happen.

IN SUMMARY:
Almost any network connection will greatly improve the speed of file
transfers, but you may find the terminal emulation agravating if you
are used to a terminal with moderate response.