Mike Henry <{$mrtickle$}@nospam.demon.co.uk> wrote:

>So then, if we say that the Denon DVD-1920 is in the "bad" camp of devices
>which are downsampling the audio from 96kHz to 48kHz, when there is
>apparently no need to do so, my questions are
>1. why does it do this, it is really just copyright rules or is there some
>other limitation that makes this necessary?
>1. do they all do this?
>2. if not, which amps/dvd players don't do this?

I'd guess that multi-channel audio can't simply be extracted from a
received signal and passed essentially unchanged over an HDMI link, but
that there's a non-trivial encoding task for the HDMI output chip, and
that the encoding task would be more complex for DVD-audio quality or
multi-channel audio, so it would be down to the chips available on the
market.

I don't know the answer to your second and third query.

-- 
Dave Farrance

"Dave Farrance"  wrote in message 
news:c3o5445ecpl02v0uj40q0rj1v8jrfvrcfu@4ax.com...
> Mike Henry <{$mrtickle$}@nospam.demon.co.uk> wrote:
>
>>So then, if we say that the Denon DVD-1920 is in the "bad" camp of devices
>>which are downsampling the audio from 96kHz to 48kHz, when there is
>>apparently no need to do so, my questions are
>>1. why does it do this, it is really just copyright rules or is there some
>>other limitation that makes this necessary?
>>1. do they all do this?
>>2. if not, which amps/dvd players don't do this?
>
> I'd guess that multi-channel audio can't simply be extracted from a
> received signal and passed essentially unchanged over an HDMI link, but
> that there's a non-trivial encoding task for the HDMI output chip, and
> that the encoding task would be more complex for DVD-audio quality or

The encoding task would be more complex down-sampling a 96kHz LPCM stream to 
48kHz and re-encoding it than leaving it intact as a 5.1/6.1/7.1 LPCM stream 
and letting the amp decode it raw. After all if no copyright flag is set the 
96kHz LPCM stream goes straight though to the amp unchanged anyway.

> multi-channel audio, so it would be down to the chips available on the
> market.
>
> I don't know the answer to your second and third query.
>
> -- 
> Dave Farrance

"Agamemnon" <agamemnon@hello.to.NO_SPAM> wrote:

>"Dave Farrance"  wrote
>> I'd guess that multi-channel audio can't simply be extracted from a
>> received signal and passed essentially unchanged over an HDMI link, but
>> that there's a non-trivial encoding task for the HDMI output chip, and
>> that the encoding task would be more complex for DVD-audio quality or
>
>The encoding task would be more complex down-sampling a 96kHz LPCM stream to 
>48kHz and re-encoding it than leaving it intact as a 5.1/6.1/7.1 LPCM stream 
>and letting the amp decode it raw. After all if no copyright flag is set the 
>96kHz LPCM stream goes straight though to the amp unchanged anyway.

The down-sampling would have to be included anyway for the situation
where the downstream unit was non-HDCP -- and so it would be simpler for
the silicon of the HDMI encoder if it didn't have to handle top speed.

-- 
Dave Farrance

In , Dave Farrance
 wrote:

>"Agamemnon" <agamemnon@hello.to.NO_SPAM> wrote:
>
>>"Dave Farrance"  wrote
>>> I'd guess that multi-channel audio can't simply be extracted from a
>>> received signal and passed essentially unchanged over an HDMI link, but
>>> that there's a non-trivial encoding task for the HDMI output chip, and
>>> that the encoding task would be more complex for DVD-audio quality or
>>
>>The encoding task would be more complex down-sampling a 96kHz LPCM stream to 
>>48kHz and re-encoding it than leaving it intact as a 5.1/6.1/7.1 LPCM stream 
>>and letting the amp decode it raw. After all if no copyright flag is set the 
>>96kHz LPCM stream goes straight though to the amp unchanged anyway.

How can you be sure though; do you own any titles (do any exist) with
5.1/6.1/7.1 LPCM streams at 96kHz without the copyright flag set that are
indeed sent straight to the amp unchanged? Just so that we can pin down
that it is the copyright flag that is causing the downsampling, and it
couldn't be anything else.

>The down-sampling would have to be included anyway for the situation
>where the downstream unit was non-HDCP -- and so it would be simpler for
>the silicon of the HDMI encoder if it didn't have to handle top speed.

Yes but simpler that doesn't mean impossible or not allowed, which is all
I'm trying to ascertain :-)

Mike Henry <{$mrtickle$}@nospam.demon.co.uk> wrote:

>Dave Farrance wrote:

>>The down-sampling would have to be included anyway for the situation
>>where the downstream unit was non-HDCP -- and so it would be simpler for
>>the silicon of the HDMI encoder if it didn't have to handle top speed.
>
>Yes but simpler that doesn't mean impossible or not allowed, which is all
>I'm trying to ascertain :-)

This 2005 article says that it was a limitation of the products then in
the marketplace and that it would take time for the audio companies to
include full-blown HDMI switching capability.  That was over three years
ago, so hopefully things have changed.

http://news.digitaltrends.com/talk-back/45/hdmi-falls-short-on-audio-for-now

-- 
Dave Farrance