[opendtv] Re: Thomson readies solutions for U.S. Digital TV broadcast transition
- From: "Allen Le Roy Limberg" <allimberg@xxxxxxxxxxxx>
- To: <opendtv@xxxxxxxxxxxxx>
- Date: Wed, 10 May 2006 10:09:34 -0400
The terminology may vary depending on what group of engineers you are
talking to.
DTV receiver designers apply the term MPEG-2 decoder to the packet decoding
apparatus that decompresses video packets to recover video, and the term
AC-3 decoder to the packet decoding apparatus that decompresses sound
packets to recover multi-channel sound. The terms refer to the packet
coding protocols prescribed by ATSC in A/53. The data for additional
decoding protocols are contained in the data windows of MPEG-2 packets
The demodulator recovers 8-level baseband 8VSB signal which after
equalization and echo-suppression filtering issupplied to the symbol
decoder. The symbol decoder is a trellis decoder, customarily of Viterbi
type. After convolutional de-interleaving of the data bytes, Reed-Solomon
forward error correction is done to recover randomized data packets. This
is followed by de-randomization, performed by exclusive-ORing the data
packets with a prescribed pseudo-random noise sequence. The data packets
are sorted to packet coders according to packet identifiers (PIDs) in their
headers.
E-8VS introduces a number of further considerations, but does not appear to
be being taken up by anyone. The halving or quartering of code rate costs
too much channel capacity in the opinion of one broadcast engineer I spoke
to is concerned.
----- Original Message -----
From: "John Willkie" <johnwillkie@xxxxxxxxxxxxx>
To: <opendtv@xxxxxxxxxxxxx>
Sent: Tuesday, May 09, 2006 3:18 PM
Subject: [opendtv] Re: Thomson readies solutions for U.S. Digital TV
broadcast transition
> It's quite sad, Bert, that your argumentive and ignorant positions have
brought this list to the point where you are the only person posting.
>
> Ignorant sounds harsh, no? It's appropriate, especially when contemplating
your comments (quoted verbatim below) on this item.
>
> Over a month ago, you made the same mistake I am going to address in this
message. Mark Aitken gently tried to disabuse your lack of knowledge as to
what an MPEG-2 decoder does and THAT IT HAS ABSOLUTEY NO ABILITY TO DECODE
VIDEO!
>
> So, let's go back to basics. A Demod discerns an RF signal and outputs a
data stream that can be processed by an MPEG-2 demodulator. What a
demodulator does is to enable an audio or video decompressor to select ONE
AND JUST ONE packet id to decompress and to pass on for audio or video
rendering.
>
> Therefore, only a fool or the abjectly ignorant (or both) would talk about
an MPEG-2 demodulator doing H.262 or H.264 processing. Some folks might
have put both functions on a single die, but they charge extra for it.
>
> ANYONE who has ever contemplated THE FIRST THING about REALLY processing
MPEG-2 packets (or who has examined the licensing situation) LEARNS about
this in just the first few hours.
>
> Still with the basics: MPEG-2 means ISO/IEC 13818. That is divided into
three sections: -1 for systems, -2 for video and -3 for audio. (-3 has ZERO
relevance in the ATSC world; it isn't even referenced in the standard ATSC
specification suite. Offhand, I can't recall the number right now for
MPEG-2, but it's much higher than 13818.
>
> MPEG-1 is ISO/IEC 11172.
>
> The one chip solution that you are looking for has tuner, MPEG-2 demod,
AC-3 audio and MPEG-2/MPEG-4 video decompressor. It doesn't exist, but you
think you find it everywhere.
>
> Here's a tip: comment on published specs, don't speculate on press
releases. If there isn't a published spec, it's just spin.
>
> John Willkie
>
> -------
> Bert wrote:
> Thomson is one of the two companies selected by the NAB, for low-coast and
high perfromance ATSC STBs.
>
> For the ATSC market, they have developed an IC, the 4300A, that
incorporates both ATSC demod and MPEG-2 decoder. Interesting. I'm assuming
that when they say MPEG-2, they mean H.262 algorithm only, not H.264.
>
> Next step is to also incorporate the tuner in a single chip, to further
reduce cost, footprint, and power requirement. Althought they are already
calling this 4300A a "one-chip solution."
>
> No mention of this at their own web site, though, that I could find. I was
looking for more specifics.
>
> Notice that the low-cost STB they are building for the NAB/MSTV does not
use this 4300A chip. It will have an NTSC output, composite video, and
audio. So it's just going to be a stripped-down SDTV box. However, they are
also planning an HDTV STB, it seems, and a receiver with USB output for PCs.
Even though this STB does not use the 4300A, it is said to meet all the
stringent NAB/MSTV requirements. Not bad.
>
> Bert
>
>
>
> ----------------------------------------------------------------------
> 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: