dvv@hq.demos.su (Dmitry V. Volodin) (10/03/90)
OK, I've already got several responses on my RTS yell. Some of them explaining what is RTS, others asking what I've ment. I've posted my message in order to attract netpeople's attention to VERY common misinterpretation of RTS. Very typical is to think that RTS controls the flow from modem to computer (DCE to DTE officially). Here is a quote from one of letters: >The serial port standard is asymmetrically named, but is actually >symmetric. One of the above is output, and the other is input. >The signal, output or input, means that "this direction is ready >to receive another byte". Serial chips can be programmed so that >they will respect this input signal and not send until it's ON. >They also can be programmed so that they receive a byte and drop >the signal output until you pull that byte out, then they raise >the output signal and the other end sends another byte. And here is my reply: >The above doesn't comply with RS-232 and V.24. RTS is raised by computer >to indicate the modem (or whatever) that there's some data to send. >CTS is an indication from the modem that RTS is felt and all the >procedures to send away data are up and running. This scheme was >designed primarily for half-duplex modems and has nothing to do with >stopping the modem (or whatever, forgot that official name) to deliver >data to computer. The computer is supposed to have fast enough procedures >to capture data from modems at full speed. > >This RTS/CTS misunderstanding is VERY common and, I suspect, appeared >when smart guys used to wire one computer's RTS to other's CTS to >make flow control on direct links. What is really painful - other >smart guys write smart drivers INCOMPATIBLE with canonical ones. >SCO's interpretation is quite V.24-compliant, and FAS's - just >clever incompatible hack. Regards. Dima P.S. Don't think there's no more half-duplex modems in the Unix world - think about faxes. -- Dmitry V. Volodin <dvv@hq.demos.su> | fax: +7 095 233 5016 | Call me Dima (D-'ee-...) phone: +7 095 231 2129 |
raymond@ele.tue.nl (Raymond Nijssen) (10/05/90)
In article <1990Oct3.151720.22811@hq.demos.su> dvv@hq.demos.su (Dmitry V. Volodin) writes: >Don't think there's no more half-duplex modems in the Unix world - >think about faxes. .... or HST modems. ______________________________________________________________________________ Raymond X.T. Nijssen / Don't speak if you / Oh VMS, please forgive me all raymond@ele.tue.nl / speak for yourself / unfriendly things I said about you
src@scuzzy.in-berlin.de (Heiko Blume) (10/07/90)
dvv@hq.demos.su (Dmitry V. Volodin) writes: >>SCO's interpretation is quite V.24-compliant, and FAS's - just >>clever incompatible hack. it suffices to make modern modems work. my HST uses CTS to tell the computer to stop sending data, and the computers use RTS to tell the modem to stop sending data. (i know it for sure, someone just downloads x11r4 from my machine....) who cares about recommendations from a standard if it doesn't work with a given hardware ? ;-) -- Heiko Blume c/o Diakite blume@scuzzy.in-berlin.de FAX (+49 30) 882 50 65 Kottbusser Damm 28 blume@scuzzy.mbx.sub.org VOICE (+49 30) 691 88 93 D-1000 Berlin 61 blume@netmbx.de TELEX 184174 intro d scuzzy Any ACU,f 38400 6919520 ogin:--ogin: nuucp ssword: nuucp [HST,V.42bis]
gemini@geminix.in-berlin.de (Uwe Doering) (10/07/90)
dvv@hq.demos.su (Dmitry V. Volodin) writes: >>This RTS/CTS misunderstanding is VERY common and, I suspect, appeared >>when smart guys used to wire one computer's RTS to other's CTS to >>make flow control on direct links. What is really painful - other >>smart guys write smart drivers INCOMPATIBLE with canonical ones. >>SCO's interpretation is quite V.24-compliant, and FAS's - just >>clever incompatible hack. I'm aware of this problem. But if you look at high speed modems you'll find that the manufacturers have used this incompatible method for years. Therefor, it's only logical to write a driver like FAS which fully supports these modems (and other DCE). BTW, just a few months ago this full duplex method was adopted by the CCITT and its US counterpart. Now each pin (RTS and CTS) has two names. I don't remember the new ones at the moment but they use the old names (RTS and CTS) for the half duplex handshake and the new names for the full duplex method. The problem is if I use these new names in the FAS documentation nobody will know what I am talking about because the new names are almost unknown to the UNIX community, at least at this time. Therefor, I'll stick with RTS/CTS for the time being. To make clear what method is meant one could call it "full duplex RTS/CTS" or otherwise "half duplex RTS/CTS". I think this complies with the terminology modem (and other DCE) manufacturers use. Uwe -- Uwe Doering | Domain : gemini@geminix.in-berlin.de Berlin |---------------------------------------------------------------- Germany | Bangpath : ...!unido!fub!geminix!gemini