Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!ut-sally!husc6!rutgers!ames!sdcsvax!ucbvax!decvax!tektronix!reed!omen!caf
From: caf@omen.UUCP (Chuck Forsberg WA7KGX)
Newsgroups: rec.ham-radio.packet,comp.dcom.modems
Subject: Re: Gilmore responds...
Message-ID: <581@omen.UUCP>
Date: Sun, 6-Sep-87 18:06:53 EDT
Article-I.D.: omen.581
Posted: Sun Sep  6 18:06:53 1987
Date-Received: Mon, 7-Sep-87 03:47:59 EDT
References: <115@splut.UUCP> <7176@steinmetz.steinmetz.UUCP>
Reply-To: caf@omen.UUCP (Chuck Forsberg WA7KGX)
Organization: Omen Technology Inc, Portland Oregon
Lines: 50
Keywords: public access to ham spectrum
Xref: mnetor rec.ham-radio.packet:508 comp.dcom.modems:930

Having "finished" a debate with megalink's author on Compuserve, I can
add some comments about these protocols.  Normally I don't criticize
other file transfer protocols, but I did not care to let false
criticism of YMODEM and ZMODEM by megalink's auhors go unchallenged.

For the record, the fact that my name appears about 9 times in the
latest megalink protocol description does *not* mean I endorse
megalink in any way.

Megalink uses fixed length blocks both for the filename (can't handle
full pathnames) and data blocks.  Each 512 byte megalink data block
has a minimum of seven overhead characters.  As a result, megalink has
considerable overhead, especially on short files.  BTW, the 98.5 %
transfer efficiency "gttutor" reports for an 880k .ARC megalink file
transfer is mathematically impossible.

Megalink does not protect all of the XOX/XOFF and network control
characters that ZMODEM does, so megalink has a theoretical efficiency
advantage over ZMODEM on .ARC files longer than about 20kb.  In
practice, ZMODEM's automatic receiver startup and crash recovery
increase the operational efficiency in ways competitive performance
tabluations tend to ignore.

SEAlink uses 128 byte XMODEM style packets with no protection
for flow control or network command characters.  A SEAlink receiver
generates an ACK every 128 bytes, impacting throughput on buffered
modems etc.

The only implementations of megalink are those supplied by Meiners
himself.  Tests to date show this software to be overly sensitive to
buffer overruns and transmission impairments.  Under some conditions,
it is possible to speed up a megalink transfer (as reported by the
transmitter!) by halting the receiving computer at the right instant.

Actually, ZMODEM isn't the fastest file transfer protocol in the
world.  YMODEM-g is faster, but its error recovery is worse than
megalink.

ZMODEM is designed to provide user friendliness, high efficiency, and
high reliability in environments that support (or nearly support)
XMODEM.  ZMODEM's weakness is a coding complexity greater than XMODEM,
a failing that is more important to Turbo Pascal mavens than to C
hackers and comm software users.

Chuck Forsberg WA7KGX Author of Pro-YAM communications Tools for PCDOS and Unix
...!tektronix!reed!omen!caf  Omen Technology Inc "The High Reliability Software"
17505-V Northwest Sauvie Island Road Portland OR 97231  VOICE:503-621-3406:VOICE
    TeleGodzilla BBS: 621-3746 19200/2400/1200  CIS:70007,2304  Genie:CAF
  omen Any ACU 2400 1-503-621-3746 se:--se: link ord: Giznoid in:--in: uucp
  omen!/usr/spool/uucppublic/FILES lists all uucp-able files, updated hourly