[opendtv] Re: Ride in the Harris-LG MPH bus

  • From: "John Willkie" <johnwillkie@xxxxxxxxxxxxx>
  • To: <opendtv@xxxxxxxxxxxxx>
  • Date: Tue, 17 Apr 2007 22:55:31 -0700

Bob;

I'll let Mark answer for himself, but I think you are confusing codec
efficiency (which is irrelevant here) with the efficient use of the
transport stream.

A-VSB has two inefficiencies: the SRS payload (2.2mbps/sec, or about 12% of
the 19.39 mb/sec in the transport stream), and it also requires (I have to
phrase this carefully to not violate an NDA) most packets in the transport
stream to have adaptation field headers with training signals.

The adaptation field imposes a bit of overhead, and the training signals
within the adaptation field also decrease the bits available for a/v content
and data essence.

I'm not aware of the size of the training signal, and if I were, I wouldn't
be able to tell, but the bit isn't insignificant.

Since I've been told this from non-ATSC sources, I can say that the sticking
point seems to be just 'how many" packets can be transmitted in a sequence
of a few dozen packets that DON'T have the training signals.

Samsung, before this MPH came about, had an upper limit that they feel, if
exceeded, they will walk away from the proposal.

Sorry I can't be more specific, but not having these requirements, I
suspect, will make it easier for the MPH proposal to pass T3/S8. Also,
there's the fact that LG has 'some' sway at the ATSC.

This, I bet, will be discussed at the ATSC annual meeting next month.
Assuming your annual revenues are less than $8 million, you can join for
less than $2000. :-)

That is, if the U.S. market is important to you.

John Willkie

----- Original Message ----- 
From: "Bob Miller" <robmxa@xxxxxxxxx>
To: <opendtv@xxxxxxxxxxxxx>
Sent: Tuesday, April 17, 2007 6:38 PM
Subject: [opendtv] Re: Ride in the Harris-LG MPH bus


> Mark,
>
> I had heard the claim by Harris that their method was much more
> efficient than A-VSB.
>
> Your numbers suggest otherwise. Seems that A-VSB is far more
> efficient. Am I reading this right?
>
> Also if more of these 2.2 Mbps channels were devoted to A-VSB in a 6
> MHz station would the SRS signal penalty remain at 2.89 Mbps or would
> it rise proportionately? If it remains the same then A-VSB becomes
> even more efficient the more of the channel you use with it.
>
> Bob Miller
>
> On 4/17/07, Mark Schubin <tvmark@xxxxxxxxxxxxx> wrote:
> > Transmission was from Sinclair's KVCW, channel 29, broadcasting from
> > Black Mountain at 750 kW ERP at 1086 feet HAAT.  Two 2.2 Mbps
> > sub-channels were used for the robust transmission leaving 15 Mbps for
> > the main transmission.  One sub-channel, using "half-rate" transmission,
> > had a payload of 557 kbps; the other, with "quarter rate," had 299
> > kbps.  MPEG-4 AVC coding was used for those sub-channels.
> >
> > The antenna was one 6-inch untuned whip split four ways
> >
> > It was interesting to me that the main channel failed when the bus was
> > barely creeping.  FYI, it was using a 5th-generation LG receiver.
> > Sometimes it seemed okay at speed.  The "half-rate" sub-channel took a
> > few hits in the worst reception conditions.  I take their word for it
> > that the "quarter-rate" sub-channel never took a hit (the decoder,
> > however, froze).
> >
> > The white-noise threshold is said to be 8 dB for the half-rate channel
> > and 4 dB for the quarter-rate.
> >
> > I'm told the A-VSB demo also uses two 2.2 Mbps robust sub-channels.
> > Their "half-rate" is about 1 Mbps and "quarter-rate" about 500 kbps, but
> > they take a his of another 2.89 Mbps for the SRS signal.
> >
> > TTFN,
> > Mark
> >
> >
> >
>
>
> ----------------------------------------------------------------------
> You can UNSUBSCRIBE from the OpenDTV list in two ways:
>
> - Using the UNSUBSCRIBE command in your user configuration settings at
FreeLists.org
>
> - By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word
unsubscribe in the subject line.
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>
>

 
 
----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:

- Using the UNSUBSCRIBE command in your user configuration settings at 
FreeLists.org 

- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word 
unsubscribe in the subject line.

Other related posts: