home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
kermit.columbia.edu
/
kermit.columbia.edu.tar
/
kermit.columbia.edu
/
newsgroups
/
misc.20020314-20021006
/
000041_ishikawa@yk.rim.or.jp_Wed Apr 24 14:13:11 EDT 2002.msg
< prev
next >
Wrap
Text File
|
2002-10-06
|
3KB
|
74 lines
Article: 13332 of comp.protocols.kermit.misc
Path: newsmaster.cc.columbia.edu!panix!howland.erols.net!news.maxwell.syr.edu!newsfeed.media.kyoto-u.ac.jp!newsfeed.rim.or.jp!news.rim.or.jp!not-for-mail
From: Ishikawa <ishikawa@yk.rim.or.jp>
Newsgroups: comp.protocols.kermit.misc
Subject: Re: a bug on GNU/linux: speed reset to unintended value occasionally.
Date: Thu, 25 Apr 2002 02:22:31 +0900
Organization: Ye 'Ol Disorganized NNTPCache groupie
Lines: 50
Message-ID: <3CC6E9D7.F4F2C624@yk.rim.or.jp>
References: <3CAFF81C.8039CBF8@yk.rim.or.jp> <a8q34m$l3l$1@watsol.cc.columbia.edu> <3CB1D4F3.10D79B1F@yk.rim.or.jp> <a8svsu$3vl$1@watsol.cc.columbia.edu> <3CBAB0BC.1D3ABF7B@yk.rim.or.jp>
NNTP-Posting-Host: pl1447.nas911.n-yokohama.nttpc.ne.jp
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-2022-jp
Content-Transfer-Encoding: 7bit
X-Trace: news.rim.or.jp 1019668953 68363 210.139.44.167 (24 Apr 2002 17:22:33 GMT)
X-Complaints-To: root@rim.or.jp
NNTP-Posting-Date: Wed, 24 Apr 2002 17:22:33 +0000 (UTC)
X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.4.18 i686)
X-Accept-Language: ja, en
Cache-Post-Path: duron!unknown@localhost
X-Cache: nntpcache 2.3.3 (see http://www.nntpcache.org/)
Xref: newsmaster.cc.columbia.edu comp.protocols.kermit.misc:13332
>Has anyone noticed this sub-optimal speed
>using Kermit on two closely
>placed PCs at 38400 bps?
It turned out that the apparent slowdown
was an artifact of
sending binary file, namely wermit.
I got a much better throughput when
I tried to send largish ascii file such as
/etc/termcap.
This was again the two ports connected
directly by cross/null cable.
It puzzled me for a while, but
finally I figured it was a slowdown
caused by the prefixing or quoting of
binary characters:
If I disable the prefixing via
set control-character unprefixed all
then I got an improved throughput.
(With prefixing I got sub-3KB/sec throughput.
Without prefixing I got over-3KB/sec throughput.)
One would not usually remove prefixing
unless we know the transmission occurs via
clean direct connection.
Kermit has a reason to be conservative. It is one of
the best file transfer program
under a noisy/lossy/disruptive
environment.
The reason I got interested in the throughput was
that I was curious to learn if
the throughput improves when I use full 8 bits
connection instead of 7 bits connection.
Obviously at this speed (38400), we are already close
to the limit of serial port, and
whether we use 7 bits or 8 bits isn't that important.
Other overhead such as prefixing masks
the improvement, if any, of using full 8 bit character
transmission path.
Thank you again for the great software package.
Happy Hacking!