home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!uunet.ca!xenitec!zswamp!geoff
- From: geoff@zswamp.UUCP (Geoffrey Welsh)
- Newsgroups: comp.dcom.modems
- Subject: Re: V.42bis effectiveness (was Re: V.42bis - on 2400 baud)
- Message-ID: <9m5uqB5w165w@zswamp.UUCP>
- Date: Wed, 09 Sep 92 21:57:55 EDT
- References: <1992Sep2.231452.3907@cco.caltech.edu>
- Organization: Izot's Swamp
- Lines: 28
-
- andy@cs.caltech.edu (Andy Fyfe) writes:
-
- > I did a bit of experimenting sending a compressed file over a v42bis
- > link. I sent both the file as it stood, and an ASCII-fied version,
- > make with btoa. I found that the V42bis compression on the ascii
- > version almost completely compensated for its greater size, compared
- > to the binary file.
-
- This is interesting, but I refuse to even consider it for my arsenal of
- benchmarking trivia until I know a *lot* more than you've posted there! For
- one thing, I don't trust throughput measurements on any file that takes less
- than ten minutes, and even then I use the *receiver's* CPS rate to eliminate
- the error introduced by transmit buffering.
-
- Try this: make a file one million bytes long, containing only spaces.
- Transmit it via V.42bis compression, and then transmit a version compressed
- with compress, zip, or even zoo. I think you'll find that traditional
- compression doesn't do any worse than V.42bis, but it can do considerably
- better; unless there's a reason to send uncompressed data, you're almost
- always better off to send it compressed. Hence when I discuss modem
- throughput I prefer to confine myself to the throughput attained without data
- compression and hope that it's indicative of throughput on precompressed
- files.
-
- Geoffrey Welsh, 7 Strath Humber Court, Islington, Ontario, M9A 4C8 Canada
- geoff@zswamp.uucp, [xenitec.on.ca|m2xenix.psg.com]!zswamp!geoff (416)258-8467
- Coincidentally, most people who fight for "fairness" would be
- significantly better off under the system they call "fair"...
-