On 16/11/2010 22:42, Sean Breheny wrote: > On Tue, Nov 16, 2010 at 11:39 AM, Herbert Graf wrote: >> The max theoretical bandwidth from analog end to analog end is about >> 35kbit/s. > That sounds right to me as a practical limit, but not a theoretical > limit. Where did you get that from? If it were a purely analog line > then the throughput would be governed by the bandwidth (fixed at about > 8kHz) and the signal to noise ratio (which could always be improved > and so does not impose a theoretical limit). That leaves the > digitization step as the theoretical limit and that is 64 kbit/sec. > > Sean Sample rate is 8KHz So practical bandwidth is < 4KHz, hence the 3200 symbol rate or 3429=20 symbol rate limits. SNR, uLaw/ALaw and 8 bit companded resolution limit=20 the bitrate (bits per symbol) The V90/V92 mode by pass the digitisation with uLaw or Alaw and 8kHz=20 sampling. Either for Downstream only or Downstream and Upstream. If V90/V92 had been designed for Europe only, it would have been 64k Max=20 rather than 56k Max Actual ISDN in Europe, entry level is always been 2 x B + D =3D 2 x 64 +16= =20 =3D 144kbps Some Telcos did offer 9,600bps X.25 over the D (signalling=20 channel) as the user data only uses the B channels. --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .