In article , Don Pearce
wrote:

It very much depends on exactly how live is live. I don't suppose anyone
would be any the wiser if there was, say, a fifteen second delay in the
broadcast.

FWIW I'm still in the middle of analysis so anything I say at present is
'provisional'. However your point about the meaning of 'live' prompts me to
report the following...

One thing I finally got a 'round tuit' for this time was to write some
simple programs to do time-correlations of different 'versions' of some of
the proms. e.g. 'listen again' versus '320kb live'. This has allowed me to
determine the time-alignments and levels of covariance/difference etc with
a time precision of 1 sample.

I have been surprised by some of the results. I'll say more about that when
I've finished the analysis and perhaps have some idea of causes. But one
comment is perhaps worth making now.

I had assumed that the 1 second 'gaps' in the 320kb live stream were simply
due to some blocks of data going AWOL. Thus, like tape dropouts, were
sections of audio I had not heard in the 320kb version. But my correlations
seem to show that the data stream resumed each time *with an increase delay
of about 1 second*. Hence they may in effect have been more like 'pauses'
than 'losses' - despite the stream being 'live'.

Curious result. Not yet certain of the details or reasons. Say more when I
know more. Currently wading though rivers of data. :-)

Slainte,

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
Audio Misc http://www.audiomisc.co.uk/index.html