[opendtv] Re: OTA
- From: Craig Birkmaier <craig@xxxxxxxxx>
- To: opendtv@xxxxxxxxxxxxx
- Date: Wed, 14 Mar 2012 08:04:02 -0500
At 3:55 PM -0500 3/13/12, Manfredi, Albert E wrote:
Craig Birkmaier wrote:
[ATSC to WiFi box]
2. Codec efficiency - MPEG-2 is simply not efficient enough for mobile
applications, especially over wireless broadband networks;
The part of the path in which you need the highest possible coding
efficiency is the broadband pipe. In this instance, the "broadband
pipe" is the ATSC channel, so there's NO PROBLEM with coding
efficiency. You have plenty of bandwidth within the WiFi network,
assuming a reasonably up to date WiFi.
Excuse me Bert. Why did the ATSC add h.264 support for the M/H standard?
The ATSC channel, and ANY other distribution pipe carry bits. If you
want higher image quality or more sub channels with good quality you
need more efficient codecs.
Now, the box itself MAY recode the stream to H.264, if there are
whatever licensing reasons to do this. But even if the indoor stream
retains MPEG-2 compression, you can still fit maybe 6 ATSC
multiplexes, at least, in a typical 802.11n WiFi.
It is not licensing reasons that require support for h.264, although
this was a factor in moving from MPEG-2 to h.264. You need h.264
because that is the codec that is now being used by most of the
popular mobile devices. So your Antenna to WiFi box will need to
support both h.264 and MPEG-2, and pay the licensing fees for both.
And you will need a rather sophisticated ship to do the transcoding
in real time.
Here is one such chip that cost $216 each in quantities when it was
released in 2007. If there was a market, such a chip would probably
cost less than $100 today...
http://www.eetasia.com/ART_8800477251_499501_NP_ce9cb44b.HTM
> 3. Hardware acceleration - again, licensing costs make it expensive to
support MPEG-2.
I'd be surprised if MPEG-2 isn't already supported in accelerators,
because that's all there was for a long time.
There is a good chance that the chips used in many mobile devices
"could" decode MPEG-2. Most GPU vendors do support both MPEG-2 and
h.264. But the $2 per unit license (uncapped) is a major barrier to
support.
For example: Apple sold more than 37 million iPhones and 15 million
iPads in the first fiscal quarter of 2012. To support MPEG-2 the
license fees for this quarter alone would have been $104 million. The
royalty for h.264 is capped at $6.5 million annually.
http://www.erg.abdn.ac.uk/future-net/digital-video/mpeg2-dec.html
In any event, once again, the box can always transcode to H.264, if
that's really necessary.
See above.
NO BERT. It is because ATSC does not work for mobile and ATSC M/H is
too inefficient.
Nope. Define "too inefficient," and provide a point of comparison.
It isn't hard to do, Craig. Work it out, so maybe you'll convince
yourself.
You missed it, Craig. What "broadcaster"? Let's say Sinclair builds
out this LTE infrastructure for the Baltimore market area. They may
be the only "broadcaster" transmitting OTA TV signals in Baltimore,
or perhaps there may be one other LTE TV net in that market.
I should let Mark answer this, and hope he will. But broadcasters are
not throwing in the towel. What is likely to happen is that we will
see many second tier broadcasters take the money and run, or choose
to sell their spectrum and rent a sub-channel from a station that
does not sell.
But the broadcasters who remain are likely to stay with ATSC 1.0 for
some period of time until ATSC 2.0 is finished and transmission
hardware is available for deployment. My educated guess is that
broadcasters will begin the planning for a new transmission business
model in parallel with the development of the standard. This planning
will include discussions/negotiations with the existing cellular
infrastructure providers AND studies to determine if broadcasters can
build out the new distributed infrastructure on their own.
In the end, we are going to see the industry move to a
shared/distributed transmission infrastructure, and I strongly
suspect that this infrastructure will be paid for based on a usage or
carriage model. What is carried, and whether these bits will be
"free" or paid for is very much "up in the air." But we can already
see what is happening with Internet streaming, so broadcasters are
likely to have little choice but to follow a similar model.
So, my guess is, the other content owners, now no longer
"broadcasters," would either pay Sinclair for carriage (using ad
revenues?), or Sinclair would pay them to carry their signals, and
keep the ad revenue? Or some combination of this.
They would pay for carriage, just as Sinclair would pay for carriage.
I would assume that the licenses used for this carriage would be
pooled and that the license owners would get some form of royalty to
offset their own carriage charges. And I would assume that at some
date in the future the politicians would say these royalty fees
should go to the government.
This new utility is more labor intensive than the previous big
sticks, though. So unless the ad revenues go up somehow, that extra
cost will be covered by a subscription fee. *Very much* like the
original CATV systems. And we all know what they morphed into.
No it is not more labor intensive. You are simply trading off the
costs of a big stick infrastructure for more smaller sticks. Big
sticks are very expensive to build and to operate. And with the
distributed model you have more industries sharing the physical
infrastructure.
At this point, you HAVE to ask, what do CBS, Fox, ABC, etc. do to
get their content on the air in Baltimore? If Sinclair owns the
transmission infrastructure, then why would these congloms go to
anyone other than Sinclair? So, the other "broadcasters" will be
competing only for local news, weather, and whatever community
events they cover.
First, it is unlikely that the government would allow ANY one company
to own all of the TV spectrum in one market. In large market it is
highly likely there will still be at least five broadcasters, if not
more. As for the business model for mobile broadcasting all I can say
is this:
Local broadcasters will need to create a compelling argument for
their continued existence, as the congloms (and major content
franchises, especially sports) will be able to play broadcasters
against the telcos for the best carriage deals.
So the REAL question here is WHO is going to build the LTE broadcast
infrastructure?
Existing broadcasters?
The telcos?
A partnership of both?
Regards
Craig
----------------------------------------------------------------------
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:
