Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!wuarchive!uunet!overload!dillon From: dillon@overload.Berkeley.CA.US (Matthew Dillon) Newsgroups: comp.sys.amiga.datacomm Subject: Re: 19200bps Message-ID: Date: 8 May 91 20:50:35 GMT References: <1991May5.185645.12902@cunixf.cc.columbia.edu> <1991May5.201708.452@cec1.wustl.edu> <1991May7.125031.2331@cec1.wustl.edu> Organization: Not an Organization Lines: 28 In article <1991May7.125031.2331@cec1.wustl.edu> amc4919@cec2.wustl.edu (Adam M. Costello) writes: >In article dillon@overload.Berkeley.CA.US (Matthew Dillon) writes: >> half of it, which is what 9600bps (V.32) modems do now (9600bps full >> duplex = 19.2KB bandwidth). The best you will ever see on a phone line > >You say that current 9600bps modems can sustain that rate in both directions? >Then it should be no problem to have 19200bps one direction (and, say, 1200bps >the other). I think this would be far more useful anyway. In my experience, >the traffic is always much greater in one direction. >AMC Well, close... if you run half duplex like the HST and Telebit, you still generally need a low-speed return channel to handle ACKs and other incidental data. Why? Because it takes time to 'turn around' the high-speed data direction so if it is possible to return some data without switching you get much better throughput. The split is usually something like 14400/4800 or 18800/400, but you have to be very careful, the equation isn't as simple as choosing a point to split the bandwidth. -Matt -- Matthew Dillon dillon@Overload.Berkeley.CA.US 891 Regal Rd. uunet.uu.net!overload!dillon Berkeley, Ca. 94708 USA