Path: utzoo!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!olivea!uunet!hayes!tnixon From: tnixon@hayes.uucp Newsgroups: comp.dcom.modems Subject: Re: PEP vs. v.32 Message-ID: <3825.27d2404a@hayes.uucp> Date: 4 Mar 91 12:40:42 GMT References: <6895.27CDE02A@zswamp.fidonet.org> Organization: Hayes Microcomputer Products, Norcross, GA Lines: 64 In article <6895.27CDE02A@zswamp.fidonet.org>, root@zswamp.fidonet.org (Geoffrey Welsh) writes: > For instance, I've noticed that the trend in 9600+ bps modems is to > increase the resolution in the coding constellation. PEP still uses sixteen > point QAM; has DSP technology not improved enough to enable us to code more > than four bits per baud in a multicarrier scheme? Telebit has proposed adding trellis coding to acheive better performance. The coding would go "vertically" across the frequencies, rather than "horizontally" from symbol to symbol. I believe the contraint on more constellation density has to do with the amount of time available to process each symbol more than anything else. > Bottom line: I can imagine multicarrier technology benefitting from > the application of improved technology, but there seems to be no move on > Telebit's part to do so. From this I can only conclude that the future of > multicarrier technology lies with maintaining compatibility with existing PEP > installations... I do not think that we will see improvements in DAMQAM from > Telebit. PEP is coasting. > > >Other companies that promote multicarrier modulation have > >eliminated ISI through other methods, allowing throughput > >well over 20kbps. > > I was not aware that any companies were promoting, in a large way, general > purpose multicarrier modems... In presentations before the CCITT and TIA TR-30, Intelligent Modem Corporation appears to have some innovative techniques to improve the performance of PEP. For example, they propose using much fewer characters (64 or less), sending more bits per carrier, sending each symbol for much less time, and sending no guard period (dead time) between symbols. The result is (supposedly, since we haven't seen actually test results) throughput up to 24,000 bps synchronous, or even higher. They also propose echo cancellation for full-duplex. I've always thought that Telebit tried to make multicarrier appear much more "magical" and "obscure" than it needed to be, in presentations before standards committees. They've never been really willing to "open the kimono", if you know what I mean; have avoided showing real test results; and have not been at all forthcoming with proposals to handle synchronous transmission or full duplex. They killed the "asymmetrical modem" standards proposal in the process. I'm very pleased that there's now an additional player in the multicarrier field who is willing to explain it adequetely and work with established experts to improve it, without being (1) constrained to maintaining backward compatibility with existing devices, and (2) so concerned about public image, maintaining venture-capital investor confidence, and trying to tie up the technology with existing patents. IMC hasn't shipped their multicarrier modem yet, because they want to see what happens with the V.fast standard. But I expect it will be a good product, if it is at all consistent with the descriptions we've been given. -- Toby Nixon, Principal Engineer | Voice +1-404-840-9200 Telex 151243420 Hayes Microcomputer Products Inc. | Fax +1-404-447-0178 CIS 70271,404 P.O. Box 105203 | UUCP uunet!hayes!tnixon AT&T !tnixon Atlanta, Georgia 30348 USA | Internet hayes!tnixon@uunet.uu.net