newsham@wiliki.eng.hawaii.edu (Timothy Newsham) (05/26/91)
Someone please tell me the details of Telebit Trailblazer's PEP communications protocol. Does PEP utilize V.32 (9600bps) or V.32bis (14,400bps)? If not, and PEP uses it's own protocal incompatible with CCITT's V.32 or V.32bis (proposed) standard, what is the real data rate of PEP? (Not it's "effective throughput"). And what kind of error correction and data compression does PEP use? Does it use MNP1-4 or V.42 data correction? And does PEP use MNP5 or V.42bis data compression? I am aware that this modem is compatible with slower and lower standards of modem communications, but I'm only interested in the standard and protocals that are used during the Trailblazer's optimum throughput...PEP... Help me! :)
tech@mich-ns.Michigan.COM (Mich. Network Sys. TECH SUPPORT) (05/26/91)
In article <13207@uhccux.uhcc.Hawaii.Edu> newsham@wiliki.UUCP (Timothy Newsham) writes: >Someone please tell me the details of Telebit Trailblazer's PEP "communications protocol. " "Does PEP utilize V.32 (9600bps) or V.32bis (14,400bps)? "If not, and PEP uses it's own protocal incompatible with CCITT's "V.32 or V.32bis (proposed) standard, what is the real data rate "of PEP? (Not it's "effective throughput"). " PEP is entirely different from V.32/V.32bis. PEP uses 511 different frequencies as opposed to just two, like most other protocols. The number of bits per baud depends on the quality of the phone line. PEP modems have microprocessors which monitor the quality of the phone line and can upshift (ie: use more bits per baud) if the phone line quality gets better, or downshift if it gets worse. This gives you two things: reliability and optimal use of the phone line. All of this over PSTN lines. Reliability comes into play because PEP modems can operate over very noisy phone lines with conditions that would cause other modems to drop the line. Optimal use of the phone line is acheived by the modems because they constantly monitor the quality of the line at all 511 frequencies and can change (up or down) the number of bits per baud at each frequency, based on the quality of the line. This means that you get as much throughput as the line can handle. "And what kind of error correction and data compression does PEP "use? Does it use MNP1-4 or V.42 data correction? And does PEP "use MNP5 or V.42bis data compression? " PEP uses its own 16 bit CRC error correction. The T2500 and Trailblazer Plus use a special form of Lempel-Ziv compression. "I am aware that this modem is compatible with slower and lower "standards of modem communications, but I'm only interested in "the standard and protocals that are used during the Trailblazer's "optimum throughput...PEP... " "Help me! :) Hope I've been of some help. -- Michigan Network Systems Technical Support Division 1-800-736-5984 BBS: +1 313 343 0800 E-MAIL: tech@mich-ns.Michigan.COM TELEBIT DIGIBOARD WESTERN DIGITAL 3COM SCO INTERACTIVE UNIX MICROPOLIS ADAPTEC
newsham@wiliki.eng.hawaii.edu (Timothy Newsham) (05/28/91)
>Hope I've been of some help. You're article was very helpful in answering all but one of my questions regarding the PEP protocal. (Thanks so much!) >The number of bits per baud depands on the quality of the phone line. Ok, so can you give me some numbers here so that I can compare it to other high speed modems? Like, under optimum phone line conditions, what is the real data rate (before compression) of PEP? And are optimum conditions rare, and if so, what the average real (not eff- ective) throughput of PEP? Thanks again!
peter@ficc.ferranti.com (Peter da Silva) (05/29/91)
In article <13225@uhccux.uhcc.Hawaii.Edu> newsham@wiliki.UUCP (Timothy Newsham) writes: > Ok, so can you give me some numbers here so that I can compare it to > other high speed modems? BITS/BAUD/CHANNEL varies from 2-6 for each channel in use. There are 7 bauds/second, and 512 channels. This leads to a maximum 21504 bits per second without compression. Real throughput will be between 12000 and 18000 bits per second, with 14000 typical for a good phone line. PEP does not compress, so that's the real data transfer rate. -- Peter da Silva; Ferranti International Controls Corporation; +1 713 274 5180; Sugar Land, TX 77487-5012; `-_-' "Have you hugged your wolf, today?"