To: topwright who wrote (22369 ) 11/30/1997 11:45:00 AM From: Atin Respond to of 41046
RB: I guess the question is where the technology is now, and where it is going. Where it is now is hybrid networks where traditional telephony networks have to interface with IP networks. The Tempest and the like are targeting this interface point right now. In this space FTEL's lack of density, which has to do with their smaller DSP cards that are used for vocoding etc, and their line interface cards that terminate the T1s is going to keep them in the smaller gateway market. At the moment they do not come anywhere near filling up the bandwidth available on their OC3 datacom card and they might as well have used a 10BaseT card (if they had one with the same proprietary interface for communicating with their DSP card) for all the good their OC3 card is doing right now (besides getting some press from Syracuse, but I hope everyone realizes that Syracuse wasn't interested in IP telephony per se, they were interested in proving that ATM networks can do it, and since FTEL is doing this part well, they would get chosen). Where it is going is that eventually we shall have fiber all over the place. What then? FTEL has a story in that space while many others don't. But FTEL will still require lots DSP power to do the things that are required for carrying voice over IP like vocoding (or video compression etc for that matter) and like I said, at the moment their DSP board is lagging in density and wouldn't be able to provide anywhere near the bandwidth that their OC3 card has. In the time it takes them to develop those densities on their signal processing cards, others will have fiber datacom cards working with their higher density DSP cards. Things will get interesting then if everyone is the same technologically speaking (but I don't think the development times for the two are going to be the same) and I cannot predict a winner 5-10 years out (except that it would be the software if the hardware is the same, and FTEL again a not quite as far along when it comes to developer APIs for supporting customized software applications on top of their hardware). I think people would be well served if they hedged their bets some rather than focusing on any one company but I guess that has to do with one's tolerance for risk and confidence in their predictions about technology. I am sure many here think FTEL's technology is the best out there and if you are sure, go for it! -Atin P.S. Thanks, I much prefer responding to posts of this type. I prefer my arguments couched in technical terms <g>.