home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.dcom.modems
- Path: FreeNet.Carleton.CA!an171
- From: an171@FreeNet.Carleton.CA (Anthony Hill)
- Subject: Re: To V42Bis or not to V42Bis??
- Message-ID: <DM9p11.Gn6@freenet.carleton.ca>
- Sender: an171@freenet2.carleton.ca (Anthony Hill)
- Reply-To: an171@FreeNet.Carleton.CA (Anthony Hill)
- Organization: The National Capital FreeNet
- References: <4ej6rh$oa2@canopus.cc.umanitoba.ca> <4eruq2$e2@hg.oro.net> <4etjse$a0g@azure.acsu.buffalo.edu> <4f0jnq$dr7@hg.oro.net>
- Date: Sun, 4 Feb 1996 20:03:01 GMT
-
-
- Ed Starry (estarry@oro.net) writes:
- > millevil@acsu.buffalo.edu (Andrew Milleville) wrote about {Re: To V42Bis or
- > not to V42Bis??} in 'comp.dcom.modems'...
- >
- > ~: ~ I find that at 28.8 connections I get faster CPS rates with it off.
- > ~: ~ Does this make any sense?
- > ================================
- >
- > ~: Yes. With compression 'Off' the data stream is slow enough to cross the
- > ~: serial port without errors. With compression 'On' the extra amount of data
- > ~: that's decompressed within your modem is too much for the serial port to
- > ~: handle, there are errors and retries are necessary. Retries slow down CPS
- > ~: rates.
- > ================================
- >
- > ~I think what you're trying to say is, with compression on, and while
- > ~you're transferring pre-compressed files, it will do you no
- > ~good. Compressed files are notrious for not compressing very well. In
- > ~fact, the extra overhead involved in compressing files can slow down
- > ~your throughput.
- >
- > ~It has nothing to do with serial port speed.
- > ===================================
- >
- > I said exactly what I meant. With a 28.8 modem and V.42bis 'Off' the maximum
- > throughput will never exceed 2,880 CPS! Data can cross the 'serial port'
- > error free at this rate with almost any 28.8 modem, even under the most
- > adverse of conditions.
-
- Actualy, 28.8 modems have a max. theoretical throughput of about
- 3350-3450cps without data compression, depending on the size of the error
- control packets used. Error control strips the start and stop bits from
- your data, so while the you're transfering 10bits/byte over the DTE
- connection, you're only sending 8 bits/byte (minus a bit of overhead) over
- the DCE connection.
-
- > With V.42bis 'On' the 'Expanded Data throughput rates' can easily exceed
- > 7,000 CPS! Many 'serial ports' do poorly (die) at these rates, therefore
-
- If the user has their system set up right, they should never see
- overruns, even when sending extremely compressible data. Of course, it's
- a bit much to expect that some new users to understand these stupid
- machines and get them workign right..
-
- > forcing retries. The more retries the slower the CPS rate becomes. Obviously
- > this person is experiencing so many retries the true CPS data throughput
- > crossing the 'serial port' is dropping below the rate of V.42bis 'Off'.
- > ====
- >
- > ~the extra overhead involved in compressing files can slow down
- > ~your throughput.
- >
- > If you really believe this you had better get your calculator fixed. When I
- > see 7,000 CPS throughput rates I don't care how much 'extra overhead' there
- > is, 7,000 is 7,000! I always thought 7,000 CPS was faster than 2,880 CPS?
-
- If you transfer pre-compressed files (eg .ZIP files), you can not
- compress them any more using v.42bis (or MNP5 for that matter). MNP5
- actualy tries to compress them again, and ends up actualy increasing the
- size of the data packets, lowering throughput. v.42bis is smart enough to
- test which of the "compressed" or "uncompressed" packets are smaller, and
- sends that one instead of just blindly sending out compressed packets, so
- the only reduction in speed with v.42bis would be a very small one for
- overhead (about 2 or 3 cps on a 28.8 connection).
-
- Anthony
-
- --
- Anthony Hill | an171@FreeNet.Carleton.CA
-