home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!news.acns.nwu.edu!nucsrl!ddsw1!infopls!andyross
- From: andyross@infopls.chi.il.us (Andrew Rossmann)
- Newsgroups: comp.bbs.waffle
- Subject: Re: Old XT + Waffle + Fossil + V.32bis = ?
- Message-ID: <LyToXB2w165w@infopls.chi.il.us>
- Date: Wed, 20 Jan 93 12:16:44 CST
- References: <JaFLXB1w165w@amikoj.rn.com>
- Organization: INFOPLUS support, Wheeling, IL
- Lines: 28
-
- jeff@amikoj.rn.com (Jeff Kunce) writes:
-
- > FURTHER IMPROVEMENT:
- > From what I hear, I am convinced that, if you use a uucp protocol with
- > larger packets (like 7x512 "G"), V.42bis compression *does* increase
- > throughput. I have not had the time to verify this (patch the uucico,
- > arrange things with my feed, etc.)
-
- Actually, there is still no actual compression taking place. You'd get the
- same 1600+ cps under MNP4 or v42. The fact of turning on error correction
- reduces the number of bits that needs to be sent.
-
- A normal connection requires 10 bits/byte (1 start, 8 data, 1 stop). Under
- MNP/v42, the start and stop bits are not used, and several bytes of data are
- 'combined' into packets. Every 8 bits in the data packet is called an octet.
- This alone would give a 20% speedup. The actual speedup is slightly less due
- to the headers needed to define and error-check the packets.
-
- This is for the data actually sent between the modems. The modems will
- 'convert' the data to the standard 10 bits when it talks to the computer.
-
-
-
- --
- Andrew Rossmann | Sysop of Infoplus BBS, +1 708 537 0247
- andyross@infopls.chi.il.us | Infoplus Support, latest version available
- uunet!infopls!andyross | by logging in as infoplus. [v32bis/v42bis]
- NO new taxes...NO new gambling
-