Date: Mon, 6 Feb 95 10:08:06 MST From: terry@cs.weber.edu (Terry Lambert) To: mnewell@lupine.nsi.nasa.gov (Michael C. Newell) Cc: jerryk@indy.net, questions@FreeBSD.org Subject: Re: Kermit @ 14.4? Message-ID: <9502061708.AA05717@cs.weber.edu> In-Reply-To: <Pine.SUN.3.91.950206094351.14429D-100000@lupine.nsi.nasa.gov> from "Michael C. Newell" at Feb 6, 95 09:46:39 am
next in thread | previous in thread | raw e-mail | index | archive | help
> > Whenever I attempt to dial my service provider and the modems handshake at > > 14.4, Kermit barfs saying _it_ can't support 14.4. What gives? If I use > > tip, I can connect at 14.4 although the entry in /etc/remote for the port is > > set at 19.2. I have set the speed (with the Kermit 'speed' command) at both > > 57600 and 38400. > > 14.4 is not a "standard" serial communication speed (your serial port > cannot support 14.4Kbs) ; it is a standard carrier speed (V.32bis) though. > What I do is set my modem and serial card for hardware flow control, then > set the DTE (i.e. Kermit's) speed to the highest available. The modems > will handle bufferring between them. If you have V.42bis compression > you'll actually get better throughput than the raw 14.4Kbs carrier will > allow. The clock crystal divider register in the UART doesn't support 14.4; it goes from 19.2 to 38.4 to ... The good thing about flow control is that the higher you set your modem to computer baud above the modem to modem baud, the larger the latency you introduce between typing, for instance, ^C, and anything actually happening as a result as the modem buffers drain down. Er. Uh... actually, that's *not* a good thing. 8-) 8-). It is unlikely you are really pushing 1440 cps over a 14.4 anyway, since all baud rates above 4800 (assuming line turnaround) imply the use of compression, and compression algorithms aren't that deterministic. Unless you buy a Telebit and do multicarrier modulation, or go digital. For a 14.4 modem, you probably want to set the interface speed to 19.2 to handle bursting, and live with the small latency thus introduced. Opinion: flow control sucks. It is only useful when your hardware is lying to you. Damn dishonest hardware <grumble>. Terry Lambert terry@cs.weber.edu --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?9502061708.AA05717>