Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!usc!wuarchive!uunet!math.fu-berlin.de!fub!einoed!bitcave!cosheff From: cosheff@bitcave.in-berlin.de (Charles O. Shefflette) Newsgroups: comp.dcom.modems Subject: Re: CompuComm CSP modems 9600??? Message-ID: <1991May8.234800.22128@bitcave.in-berlin.de> Date: 8 May 91 23:48:00 GMT References: <19.28263B5D@zswamp.uucp> Organization: BitCave BBS, Berlin, FRG Lines: 61 root@zswamp.uucp (Geoffrey Welsh) writes: > In a letter to All, Larry Snyder (larry@nstar.rn.com ) wrote: > >what protocol do these modems use? Is it full duplex? > It is definitely *not* full duplex.[...] > I've been told - but can't confirm, nor can I find in the propoganda > anywhere - that the SpeedModem is asymmetrical, like the HST. From quite a bit of experience with the CompuComs here in Germany, and the work it took to get them to work relatively reliably (along with Joern Bush and a couple of other people here) we've made a few observations: 1. The protocol is definately asymmetrical. It appears to use 300bps mode (or something close to that) while in single character mode - at that point it is full duplex. When a packet of characters is to be transmitted the modem goes to a 9600bps forward channel with a low speed (presumably around 300bps) back channel. This tends to cause some problems on extremely noisy lines, but in general is not too inconvenient. We've seen transfer rates approaching 930cps using Zmodem here. 2. After listening to it for a while, it appears that CompuCom is using the V.29 modulator to do the 9600bps data transfer. As you may or may not be aware, V.29 is the signalling protocol used by high-speed CCITT Group III FAXes. This makes quite a bit of sense, actually, since the modulator and demodulator are already existing and work quite well. This does mean, though that this is not a full duplex modem - it has a high speed forward channel (for the one which is sending the most data) and a low speed reverse channel (as mentioned above). 3. We've had quite a bit of problems getting the modems to work correctly here in Germany. You may or may not have the same types of problems in other countries. The modems appear to work quite well in the US, though. In any case, most of the problems we've had are in relation to the 9600bps modes - the 2400bps MNP5 modes are quite good in any case. After experimenting quite a bit and spending many hours working on the modems and modifying them (with a few suggestions from CC) we managed to find a combination which seems to be fairly close and only requires a bit of "playing" to make it work in most circumstances. As a result of this work we did, CC is now shippingtheir Rev 1.4 and 1.6 modems with the modifications we worked outinstalled (at least the ones headed this way.) This modem is really quite a nice modem, it does CCITT V.22, V.22bis, MNP4 and MNP5, plus the CCITT 9600bps protocl, PLUS CCITT Group III FAX at up to 9600bps! That's quite a steal for the price, even if the 9600bps is a bit of a problem. I'm really waiting for the external version of the modem to come out though. I'd really rather not have an internal one if I could avoid it.... Hope this helps. Chuck Shefflette =========================================================================== cosheff@bitcave.in-berlin.de | Reality? We don't need no steenking | reality here!