Page 1 of 1

A question about DTS-HD Master Audio tracks.

Posted: Sat Jan 03, 2015 2:36 am
by NickNewb
Mr. Woodstock, the resident guru, I have another question for you! (or anyone else who understands audio codecs)

Hopefully it is so elementary, it will be easier for you to just answer it instead of typing up an annoyed response. :P

I'm curious (again) to know how something works with regards audio tracks, so here's an example:

I've put a Blu-ray into my player and on the back of the box, it advertises "English 2.0 DTS-HD Master Audio." When I go to rip one of the main titles (it's a TV show on Blu-ray, so there are multiple episodes) with MakeMKV, there are 2 audio tracks: "DTS-HD MA English" and "DTS Stereo English." MakeMKV says that the codec for the former is "DTS-HD Master Audio" and simply "DTS" for the latter. I also noticed that MakeMKV flags the "DTS Stereo English" audio track as "Core audio." (See Image)

I was just curious to know, why are there 2 audio tracks? Is the DTS-HD Master Audio the 'better' of the two? If so, then why is the 'crappy' one flagged as 'Core Audio?' Would you mind schooling me for a minute on audio codecs and why there might be a fallback included (if that's why there are 2 tracks on here)?

I have noticed this on other movies that advertise something on the box, and then there are multiple audio tracks, only slightly different (one being 7.1 surround, and another 5.1 surround, with the 5.1 surround being flagged as 'Core Audio' etc.)

Image

Re: A question about DTS-HD Master Audio tracks.

Posted: Sat Jan 03, 2015 6:03 am
by matt198t
The DTS-HD Master Audio track is lossless.
The DTS track is lossy.

Re: A question about DTS-HD Master Audio tracks.

Posted: Sat Jan 03, 2015 5:33 pm
by Romansh
DTS-HD uses a core + extension system where a lossy core is encoded for compatibility with older setups and the HD extension(s) contain the difference between the lossy core and the lossless (or higher-quality, in case of the still lossy DTS-HD High Resolution Audio) representation of the source audio.

Re: A question about DTS-HD Master Audio tracks.

Posted: Sat Jan 03, 2015 7:12 pm
by NickNewb
Thanks for the responses, much obliged.

Re: A question about DTS-HD Master Audio tracks.

Posted: Tue Jan 06, 2015 8:36 pm
by NickNewb
Ok, so one more explain-like-I'm-five question...just to make sure I'm understanding correctly.

I recently tried to play one of these .mkv files using the Xbox One's Media Player app which supports .mkv playback.

I did not hear any sound. Are either of these a correct understanding as to what happened?:

A. The Media Player software played the video track just fine and also tried playing *only* the DTS-HD Master Audio track, which is not enough to actually generate sound. Somehow, even though the lossy DTS Stereo English track is flagged as 'Core audio' by MakeMKV, apparently that's not enough, and I would still need to somehow flag that track as being the default audio track...? But if I did that, then when I play this file using software that DOES support DTS-HD Master Audio playback, will it take the extension and add it in?

or

B. The Media Player software saw this core + extension stuff and said 'what is this crap? I don't understand. You get nothing!' In this case, would I still need to somehow flag the lossy DTS Stereo track as being the 'default' track? (And again, by doing that, when I do eventually play this file using software that does support the core + extension stuff, will it ever apply the extension?)

Finally, I'm a little confused as to how an extension can take something that is LOSSY and covert it to LOSSLESS. I thought lossy meant that you have compressed something and permanently lost data.

Re: A question about DTS-HD Master Audio tracks.

Posted: Tue Jan 06, 2015 9:36 pm
by Romansh
NickNewb wrote:Finally, I'm a little confused as to how an extension can take something that is LOSSY and covert it to LOSSLESS. I thought lossy meant that you have compressed something and permanently lost data.
The extension is encoded at the same time as the lossy core, so the encoder has access to the lossless/source audio and can generate an extension substream containing the exact difference between the source and the decoded lossy core.

This works because the decoding process is deterministic, i.e. if all compliant decoders output the same data given the same input lossy stream, thus it's a matter of combining both signals as per the specification.