home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Columbia Kermit
/
kermit.zip
/
newsgroups
/
misc.20010921-20020314
/
000005_fdc@watsun.cc.columbia.edu_Sat Sep 22 14:06:46 2001.msg
< prev
next >
Wrap
Text File
|
2020-01-01
|
5KB
|
107 lines
Flags: 000000000001
Article: 12812 of comp.protocols.kermit.misc
Path: newsmaster.cc.columbia.edu!watsun.cc.columbia.edu!fdc
From: fdc@watsun.cc.columbia.edu (Frank da Cruz)
Newsgroups: comp.unix.solaris,comp.protocols.kermit.misc
Subject: Re: A Unix Tip problem
Date: 22 Sep 2001 16:06:51 GMT
Organization: Columbia University
Lines: 89
Message-ID: <9oicur$5j4$1@newsmaster.cc.columbia.edu>
References: <113ed076.0109070921.45c34569@posting.google.com> <1000398035.902707@blake.timetraveller.org> <9nqp4i$hau$1@newsmaster.cc.columbia.edu> <1001172990.577070@blake.timetraveller.org>
NNTP-Posting-Host: watsun.cc.columbia.edu
X-Trace: newsmaster.cc.columbia.edu 1001174811 5732 128.59.39.2 (22 Sep 2001 16:06:51 GMT)
X-Complaints-To: postmaster@columbia.edu
NNTP-Posting-Date: 22 Sep 2001 16:06:51 GMT
Xref: newsmaster.cc.columbia.edu comp.unix.solaris:351053 comp.protocols.kermit.misc:12812
In article <1001172990.577070@blake.timetraveller.org>,
Robert Brockway <robert@timetraveller.org> wrote:
: In comp.unix.solaris Frank da Cruz <fdc@watsun.cc.columbia.edu> wrote:
: : : Kermit is both a serial utility & a serial transfer protocol. For
: : : transfering data, kermit is significantly (2-3 times from memory)
: : : slower than Z-modem (sz/rz under unix)
: : :
: : No, see:
: :
: : http://www.columbia.edu/kermit/perf.html
:
: My statements on kermit were based on performance tests I carried out
: personally. All protocols were tested out of the box. I stand by the 2-3
: time performance estimate as it is what I saw with my own eyes.
:
: I note the article you mentioned shows kermit to have superior performance
: but the fact is I did not see this when I tested it. Which was a pity as
: it was otherwise very convenient to use. The article goes on to suggest
: that various implementations of Kermit were poor, which i do not doubt,
: but I was using real Kermit on a unix platform on both ends.
:
Please state what software you were using in your tests (including version
numbers), the performance related parameter settings, the details of the
connection, and so forth.
The article I cited (and wrote) is also 100% accurate. I ran the tests
myself and I stand behind the results. The article gives all the
particulars of each computer, connection, software product name and version,
and settings so anybody who cares to can check the results themselves.
None of this is to say that there can not be a situation in which Zmodem is
faster than Kermit (or that Kermit is faster than Zmodem). But the
once-common perception that Kermit is intrinsically ORDERS OF MAGNITUDE
slower than Zmodem, Ymodem, and even Xmodem is based on the situation that
existed more than 15 years ago, and even then was mostly based on
low-quality 3rd-party half-baked Kermit implementations.
: The article itself mentions that Kermit takes care of stability first &
: speed second. This statement I agree with 100%. I found Kermit to be
: very reliable, just not fast :)
:
Until recently, Kermit software was delivered with conservative performance
settings to ensure maximum robustness. Perhaps this accounts for your
results. The current version of C-Kermit, 7.0 is delivered with fast
settings by default because the world has changed sufficiently to allow it:
http://www.columbia.edu/kermit/ckermit.html
(most connections now are over the Internet, over flow-controlled
error-correcting modems, or both). Kermit 95 also uses fast settings by
default. But note that the fast settings have been there all this time,
just not enabled by default, even though we knew this would provoke a
great deal of unfair criticism:
http://www.columbia.edu/kermit/kermit.html#notslow
: : : Before ppp/slip was around I used to do alot of serial file transfers
: : : and used to always look for Z-modem, then X-modem and finally kermit.
: : :
: : This is the kind of advice that was common in the 1980s. Please bring
:
: I did prefix my statement with "Before ppp/slip" :)
:
: : yourself up to date before giving advice in public:
:
: Why? Have any of the mentioned protocols (X-Modem, Z-Modem, or Kermit)
: been redrafted since then? They're still the same protocols they were.
: You canna change the laws of physics :)
:
I can't speak for the others, but Kermit is constantly being developed
and improved. In fact, it has improved by leaps and bounds even since 1993
when the performance article was written. See, for example, sections
4.18 through 4.20 of:
http://www.columbia.edu/kermit/ckermit2.html
The Kermit protocol was designed from the very beginning to be extensible.
Thus it is quite possible to improve it without breaking its
interoperability with older versions, and that is indeed what we do here.
Anyway, I don't think the world really needs yet another Kermit versus
XYZmodem debate at this late date. It's all been done at great length
before, and you can read all of it in the Kermit newsgroup archives:
ftp://kermit.columbia.edu/kermit/newsgroups/
The software is out in the open for all who wish to check for themselves.
- Frank