Amcnc.1174
net.dcom
utzoo!decvax!harpo!duke!mcnc!taylor
Tue Apr 27 09:31:31 1982
3270 converter evaluations
We are completing tests of two 3270 protocol converters, and the following
is a preliminary analysis of these units. I will attempt to post other
reviews as the work continues, and would welcome comments from those
who have experience with these devices.
Alanthus (macom):
This unit is one of the best on paper, but, from what I have seen is
not there yet. When I talked with the factory reps, I was assured that
each problem i mentioned was being corrected "in the next software
release", but we all know the story on that one. The unit is advertised as
a 3270 converter, port selector, terminal--terminal device, and local
processor for cpm and unix all in one. The primative methods for this
seem to be present (it is based on a multiple z-80 design), but the
path for accomplishing this is not quite as rosy. On the whole, i found that
their marketing was much better than their product, tho I must admit that
my view was somewhat prejudiced in that i am much more interested in a
good 3270 converter than a multifunction box.
Some specific objections:
1) Although the box was advertised as 'dual host', the method for
host selection is so cumbersome that it is impractical for specific ports
to be able to address more than one host.
2) For speed detection, a specific character is used for each
type of terminal, and both speed and term type are detected. However,
the character must be entered once for each speed. At 1200 baud, this
means entering the character 9 times!
3) No status line available at all.
4) For several terminal types, highlighting was not available or
not desirable with their implementation.
5) We could not get the system to work well at all with our
WYLBUR system, as it seemed to retransmit fields which have already been
transmitted once or more.
RENEX:
On the whole, we were much more impressed with this system, and have no
specific serious objections. To date, all the problems we have found
have been with our system rather than the box. Although it is only a
single host system, it has some quite desirable features, including
single-character (CR) autobaud to 19200. There is also a modified
beehive terminal which has been modified to look like a 3278, and even
has special keycaps for exact emulation.
Some Comments (as opposed to specific objections):
1) As they were among the first terminals supported, there is
no highlighting on the 3101 or vt-100.
2) I was not wild about their choice of substitute key for the
'new line' function, but then i haven't found an implementation that I
really liked.
3) For the most part, the function keys on the terminal are not
defined as the pfk's. Rather, a STANDARD escape sequence is used for
most or all terminals (esc 1 = pfk 1, etc.). I don't really see this
as a fault as much as a design decision on their part.
4) For terminals that check parity, they use mark parity rather
than even parity for some reason.
As compared with what we have seen so far, I give this unit a very high
overall rating.
Your comments, questions, views, etc. welcomed.
Steve Taylor
decvax!duke!mcnc!taylor
-----------------------------------------------------------------
gopher://quux.org/ conversion by John Goerzen <
[email protected]>
of
http://communication.ucsd.edu/A-News/
This Usenet Oldnews Archive
article may be copied and distributed freely, provided:
1. There is no money collected for the text(s) of the articles.
2. The following notice remains appended to each copy:
The Usenet Oldnews Archive: Compilation Copyright (C) 1981, 1996
Bruce Jones, Henry Spencer, David Wiseman.
