[EAS] National test won't be shown on AT&T U-verse?

[Harold Price]

Sean,

You mentioned "ENDEC" and daisy-chain in your message, so let me 
address this from the point of view of the Sage ENDEC, and broadcast 
stations that make up the daisy-chain.

This is not the way the Sage ENDECs work.  We don't join the program 
already in progress.  We record from the start of the alert, and 
begin the playback from that record point, using the record buffer as 
a delay line, playing back the alert while it is still coming in.

Each listener to a Sage ENDEC output will hear the complete alert as 
it was heard at that Sage ENDEC's input.  If all the links in the 
chain do the same thing, every listener will hear every word spoken 
as part of the alert.  Your local station might start the alert 40 
seconds later than some other station, listeners to that station will 
hear the entire alert.

The thought that the alert will be heard from every speaker in the 
land, simultaneously, like it was a 1950's SciFi movie, just isn't 
true, of course.  With all the various delays from realtime in 
multiple satellite hops, compression delays, cough/profanity delays, 
there is no "real time" any more.

Joining the alert in progress, when the delivery method is multiple 
hop with 15 second (or more) delay at each hop, is always less than optimal.

If the President can't begin speaking until the last station in the 
link is listening to the "live" feed, when does he or she start to 
talk?  Thirty seconds?  One minute?  Three?

And think about the people at the start of the chain, they hear some 
sort of vamp until ready warmup, "stand by for a message of critical 
importance from the President".  Now repeat that for a minute.  You 
want to get everyone's heart rate up, start a panic, and flood 911, 
that's the way to do it.

If some parts of the system need to force tune to a "details 
channel", then that details channel needs to be delayed until the 
force tune is complete.  Otherwise, the audience on that system is 
going to miss the first part of the alert, and on a message of this 
presumed importance, the President had better not bury the lead.

I can't speak to cable products from other vendors.  I know that 
force tune solutions from systems that use the Sage ENDEC do go to a 
details channel, and that channel is delayed until the force tune is complete.

Harold

At 08:30 AM 11/9/2011, Sean Donelan wrote:
>On Wed, 9 Nov 2011, suzanne at mab.org wrote:
> > This puzzles me.  It's the tag of a story from Broadcasting & Cable
> >about the FCC order not to use the alert tones in news stories.  If the
> >duration of the test is the issue, does that suggest that U-verse can't
> >even run RMTs because they're "too short"? --SG
>
>Its actually a potential issue for all systems.
>
>RMTs are recorded store&forward; EANs are re-broadcast "live."
>
>Because RMTs are store&forward, i.e. the ENDEC records the incoming
>message, and then re-generates the recorded message on the downstream
>feed, its not a problem.  But with EANs, there is a delay through the EAS
>daisy-chain as each repeater generates the tones/quakes and force tune to
>the "live" feed already in progress.  If you are monitoring a PEP
>directly, the delay may be only 5-15 seconds until the ENDEC puts the
>"live" feed on the air after it re-generates the EAS tones; but if you are
>several steps down the daisy-chain, it takes a small amount of time for
>each station in the chain to switch to the "live" feed, the feed could be
>over.
>
>Add in other delays such as digital TV encoding/decoding; things can get
>interesting as engineers say.
>
>_______________________________________________
>This is the EAS Forum Discussion List
>
>Please invite your friends to join our Forum!
>http://lists.radiolists.net/mailman/listinfo/eas
>
>And, remember the main page: http://eas.radiolists.net