Path: utzoo!mnetor!tmsoft!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!samsung!olivea!apple!portal!cup.portal.com!cec From: cec@cup.portal.com (Cerafin E Castillo) Newsgroups: comp.dcom.modems Subject: Re: connect problems with T2500 Message-ID: <37140@cup.portal.com> Date: 21 Dec 90 23:38:06 GMT References: <1990Dec19.044634.26743@sopwith> Distribution: na Organization: The Portal System (TM) Lines: 56 Here are some answers, which I have verified with Telebit, about some of the aforementioned S-Register settings in the NEW GF7.00/BC7.00 code and some of the BA4.00/BC4.00 code that is still around. S120=16 works only in GF7.00/BC7.00 firmware releases. The S120 register is normally associated with changes to PEP packetization when PEP is used over adverse communications links (ie long distance, international, satellite, etc.). S120=12 is a common setting in such situations. In earlier firmware such as 5.00/4.00/3.01, S120=16 is accessible but does not make any changes to PEP. In GF7.00/BC7.00 S120=16 cancels the 3 second answer tone issued before the PEP tones (ie Telebit moose call ;-). Combined with S93=4, which reduces the AC V.32 answer tone, this allows for a faster train-up in PEP. S120=16 also alleviates some problems caused by the V.25 tone used in GF7.00 /BC7.00. That is, if a BA5.01/GF7.00/BC7.00 firmware modem, with its S50=0 set, hears the V.25 tones issued for a PEP connection (S50=255 OR S92=0) by the answering GF7.00/BC7.00 modem; it will try and connect in V.32. As it becomes aware of the PEP connection being requested by the answering modem, it will adjust to issuing the PEP tones to connect. Meanwhile, you have gone through two train-up sessions which are eating up phone line cost and, possibly, causing your automated file transfer (ie UUCP) to time out. This is where S120=16 in the answering modem and S50=255 in the calling modem will be of great help. Similarly, when a GF7.00/BC7.00 answering modem is set to S50=255 or S92=0 and a calling BA4.00/BC4.00 modem is set S50=0, the call will fail. This may be due to the PEP tones following the V.25 tones. Remember, the V.25 tones issued in GF7.00/BC7.00 PEP connections are the same as those issued for the V.32, V.22bis, etc., answer sequences. Revision 4.00 modems may react adversely to V.25 tones being followed by PEP tones. The 7.00 modem is set to answer in V.25 tones followed by PEP tones (S92=0). As the 4.00 modem starts to connect and react positively to the V.25 tones of the GF7.00/BC7.00 answering modem, the PEP tones kick in from the answering modem and, in the confusion that follows, the connection is aborted. This is as close to the scenerio that was given to me by Telebit as I could come. My conclusion on the aborted connection is based on my experience with the modem answering sequences and train-up procedures. I welcome any corrections on the V.25 aspect of the answering sequence, since I'm not too familiar with the info exchanged in this tone. This V.25 tone is used to prevent problems caused by echo suppressors and companders in certain long distance carriers such as MCI and SPRINT. Again, S120=16 and S93=4 might help, but an upgrade (BA5.01 for TrailBlazer Plus/T2000) is recommended to alleviate this situation. I hope this info helps. This is meant to be an informational posting. ============================================================================== Cerafin E. Castillo || //\\ ||\\ || Network Consultant || //__\\ || \\ || Los Altos Los Altos Networks || // ---\\|| \\|| Networks 340 Second St. #6 ||___// \ | \ | Los Altos, CA 94022 (415) 941-8031 UUCP: {apple,sun,uunet}!portal!cup.portal.com!cec INTERNET: cec@cup.poral.com "...No hay mal que por bien no venga..." ===============================================================================