Re: No ATSC for Brazil

"Jeff Rife" <wevsr@xxxxxxxx> wrote in message
news:MPG.1e41912b66b292c198a2fb@xxxxxxxxxxxxxxxx
> It's pretty easy to see that if Australian broadcasters could send out
> a receivable signal using less error correction (and thus more usable data
> bits), they would, since this might allow them to go back to broadcasting
> actual HD instead of government-defined HD at 576p.
>

> But, none of them are doing this...all are using more bits for error
> correction. If they didn't have to do this, they wouldn't. The UK is
> the same...they are losing as much as 30% of the potential bits to extra
> error correction. I don't think it a coincidence that both these
> countries use COFDM.
>

Doesn't the forward error correction in ATSC = r= 2/3, i.e. 1/3rd of the
data broadcast carries error correction data, the remaining 2/3 (19.4mbit/s)
being available to the broadcaster?

If so then this is actually exactly the same ratio as the modes employed in
both the UK and Australia where the FEC is also 2/3. Both countries also
allow broadcast modes which feature less error correction (FEC r=3/4, i.e.
1/4 of the data broadcast contains error correction data).

> And, since 1280x720 can be sent WITH QUALITY in as little as 11Mbps, just
> how freaking much bandwidth is being wasted by error correction? Even at
> 18Mbps, they should be able to easily send 720p plus two SD sub-channels.
> And, having only 18Mbps data with a 7MHz channel is using a *lot* of error
> correction.
>

The 'loss' isn't with the amount of error correction (the ratio of which is
the same as ATSC) but with other parameters such as the guard interval and
FFT employed (which allows higher spectrum efficiency through the use of
single frequency networking).

Whilst this means each individual channel has less bandwidth it would mean
that more channels could be transmitted in a given area since those
neighbouring areas aren't going to consume as much spectrum with broadcasts
on adjacent channels.

Obviously this model wouldn't necessarily yield any spectrum advantage in
the US where each channel broadcast is generally unique (with only a small
number of low powered 'repeater' stations re-broadcasting the same
programming) however in areas with regional or even national programming
such as Europe or Australia it might make more sense...


.

Relevant Pages

  • Re: FLAC or other uncompressed formats, which is best?
    ... don't think I could figure out the checksumming and interleaving offhand. ... data for each channel isn't in consecutive bytes on the track, the data chunks on the track aren't an exact multiple of the sample size, the channels are interleaved, and there's a lot of error correction redundancy, so it could easily be confused with other types of data, or just random noise. ... The specification includes error correction for gaps of up to 3mm in the track data, which is a *lot* of missing samples. ...
    (rec.audio.pro)
  • Re: dont understand channel coding and capacity
    ... >> channel capacity over and above that required to transmit the message. ... >> transmit additional information used in error correction. ...
    (sci.math)
  • Re: dont understand channel coding and capacity
    ... >> channel capacity over and above that required to transmit the message. ... >> transmit additional information used in error correction. ...
    (sci.stat.math)