DTS-CD DTS - "lossy"?

[Wunlow]

Again, no complaints for either one! I'd be none the wiser just going by ear.

The mix is a big deal. Hopefully obvious!
Non destructive mastering (that means no volume war or bionic treble blast) is a big deal.
These delivery formats are solid.

Agreed 100% on this.

It's somewhat of a given everyone in this group would prefer lossless. But sometimes what is done in the mastering stage can make a bigger difference. This happens with multi-channel audio or even mono film soundtracks more often than most realize. Sometimes it is a new transfer of the mag tracks that now have issues that were not present in a previous release. Or excessive noise reduction or dynamic range compression is applied on the latest reissue. Some at first may find this blasphemous but there indeed are occasions where that 384kb/s Dolby Digital track on that old crusty laserdisc is better than the latest lossless DTS-MA track.

 

[atrocity]

If I do an audio only project now days I will just save to FLAC. But FLAC is not compatible with video standards (as far as I know) so DTS is still a good option.

This makes me non-sarcastically wonder if in a streaming world "standards" even really matter any more...they certainly don't matter as much.

You're right that FLAC would not be an acceptable option on a "professional" Blu-ray or DVD, but it generally works fine in things like an MKV played on an Oppo (maybe not older models?) or using Kodi.

 

[Owen Smith]

Did you look at the white paper? Quote from the Introduction.

And they chose 20/44.1 for CD and 24/48 for DVD. Going to 24/96 was done later as an extension.

Glad we agree lossless is better.

 

[Owen Smith]

While my 76-year-old ears miss a lot that my 25-year-old ears used to hear, everyone whom I’ve demonstrated my system to is wowed by the sound. So someone with less ringing in their ears might well hear deficiencies, I think most of my DTS recordings sound pretty good, which has always been my goal. “Pretty good” mostly means that it makes ME happy. I have a handful of MP3s that I downloaded in the 00s that sound like crap, so my ears aren’t totally tin, but I believe a younger critical ear could hear some issues that I don’t.

Actually crappy ears can sometimes hear more lossy differences. Lossy encoding is based on throwing away things we can't hear, sometimes because they're masked by something else that is eg louder. But if your hearing loss means you can't hear that louder something eg because you've lost that frequency, you may then be able to notice that something else appears to be missing or is poorly resolved. Lossy algorithms model normal human hearing not damaged hearing.

 

[ar surround]

Actually crappy ears can sometimes hear more lossy differences. Lossy encoding is based on throwing away things we can't hear, sometimes because they're masked by something else that is eg louder. But if your hearing loss means you can't hear that louder something eg because you've lost that frequency, you may then be able to notice that something else appears to be missing or is poorly resolved. Lossy algorithms model normal human hearing not damaged hearing.

Most definitely. Over time, married men definitely develop hearing loss of frequencies in the female vocal range.

 

[Mr. Afternoon]

I've found recordings with higher dynamic ranges are those that suffer more from lossy compression. You try to encode a jazz or classical release to lossy and compare, the lossy will crap the bed. The higher the complexity, the worse the lossy codec will sound.

 

[ar surround]

I've found recordings with higher dynamic ranges are those that suffer more from lossy compression. You try to encode a jazz or classical release to lossy and compare, the lossy will crap the bed. The higher the complexity, the worse the lossy codec will sound.

PM, I'm wondering if you have ever tried a lossless vs lossy DTS comparison with a track or two from The Golden Bonana, and what you have noticed. Did it 'crap the bed?' That DVD-A of yours is a reference quality recording.

 

[barfle]

I've found recordings with higher dynamic ranges are those that suffer more from lossy compression. You try to encode a jazz or classical release to lossy and compare, the lossy will crap the bed. The higher the complexity, the worse the lossy codec will sound.

Agree 100%! Those crappy MP3s I mentioned were all classical, scraped from mp3.com when they were previewing various artists for free, with the intent that you would buy the real thing. I haven’t visited that site for a decade or more, but as I understand, they’re not doing that any more.

 

[Mr. Afternoon]

PM, I'm wondering if you have ever tried a lossless vs lossy DTS comparison with a track or two from The Golden Bonana, and what you have noticed. Did it 'crap the bed?' That DVD-A of yours is a reference quality recording.

I have never tried it! I never buy or keep any albums in a lossy format so I never thought to do a DTS vs Lossless comparison on my own material. The Golden Bonana DVD-A had a DD stream on the video layer for backwards-compatibility, as I didn't have enough space for DTS on that single-layer DVD-R. Maybe I should do a comparison...however I disagree with your statement on reference quality recording. There's a lot of masking I did due to various reasons (mainly the fact I was using a cheap USB mic) and the programming on the drums/bass was less than stellar. Furthermore, as my first mix job, I didn't actually know how to mix, so there's still a lot of clarity that can be brought out of the multitracks. I think a comparison between some tracks off of Album #3 would be a better use case, as I feel there are some elements that would probably get lost in lossy compression (e.g. Pan Drum in Track 20). I'll do a comparison and let you guys know.

 

[mandrix]

Atmos is lossy compression? That's news to me. If you're talking about streaming Atmos, then yeah, lossy.
The base MLP codec is absolutely lossless. The extended MLP codec used in Atmos is supposed to be as well, I thought?
Or is it? I could not find a quick reference source.
I'll bow to greater wisdom on this...

 

[stoopid]

Atmos is lossy compression? That's news to me. If you're talking about streaming Atmos, then yeah, lossy.
The base MLP codec is absolutely lossless. The extended MLP codec used in Atmos is supposed to be as well, I thought?
Or is it? I could not find a quick reference source.
I'll bow to greater wisdom on this...

This is my understanding as well. The only 'trick' with atmos (the dolby true-hd spec) is using metadata to steer the various tracks to the speakers during playback. For any releases I've seen the audio files used as the source are lossless 24bit at 48khz or better resolution.

That's one reason bluray has become the defacto format -- these high resolution audio files (8+ channels worth) can be quite large and need all that disc space, exceeding DVD dual layer storage.

 

[mandrix]

This is my understanding as well. The only 'trick' with atmos (the dolby true-hd spec) is using metadata to steer the various tracks to the speakers during playback. For any releases I've seen the audio files used as the source are lossless 24bit at 48khz or better resolution.

That's one reason bluray has become the defacto format -- these high resolution audio files (8+ channels worth) can be quite large and need all that disc space, exceeding DVD dual layer storage.

That link was a pretty good read. Well it blew one of my explanations away.
Me: well see here the software is encoded to tell the processor where to place the objects, which I reckon is where tha sound goes?
Wiki: metada=spatially encoded objects.....not matrixed....

But seriously, thanks to @stoopid , it really was a good read, all joking aside. I picked up a few things.

 

[ar surround]

...however I disagree with your statement on reference quality recording. There's a lot of masking I did due to various reasons (mainly the fact I was using a cheap USB mic) and the programming on the drums/bass was less than stellar. Furthermore, as my first mix job, I didn't actually know how to mix, so there's still a lot of clarity that can be brought out of the multitracks.

'Reference Quality' is in the ears of the beholder...

 

[4-earredwonder]

I think the word LOSSY says it ALL!

 

[Sal1950]

IMHO we'd be so much better off with lossless LPCM at 24/48 than DTS or any lossy compression scheme at 24/96 or above. In fact I'd happy roll with lossless 24/44 but would prefer 24/48 just for a little elbow-head room.
The numbers game being played by many in the audio world has gone beyond ridiculous.
2L is now pushing their fully discrete multich recordings in DXD 24/352.8 at a premium cost of course, but for what audible gain? No one has yet definitively proven that we can actual hear anything better than 16/44.1 Redbook? In todays world I understand that bandwidth is rapidly becoming a non-issue but DXD?, that's an obscene waste of space.

 

[SeeMoreDigital]

I always wondered if lossy DTS 24/96 would be worse, better or equivalent to the lossless 24/48 streams found on Blu-ray.

The '96/24' element of a lossy dts audio stream is nothing more than a meta-data flag!

Indeed, if you take a dts 96/24 audio stream and strip out all of its meta-data, its file size will be the same as a basic dts 48.00kHz audio stream, encoded at the same bit-rate!

 

[JediJoker]

The '96/24' element of a lossy dts audio stream is nothing more than a meta-data flag!

Indeed, if you take a dts 96/24 audio stream and strip out all of its meta-data, its file size will be the same as a basic dts 48.00kHz audio stream, encoded at the same bit-rate!

I don't think that's correct. It sounds like you're only able to decode the core DTS stream without the 96kHz extensions with whatever software you're using to inspect and alter the files.

 

[Mr. Afternoon]

The '96/24' element of a lossy dts audio stream is nothing more than a meta-data flag!

Indeed, if you take a dts 96/24 audio stream and strip out all of its meta-data, its file size will be the same as a basic dts 48.00kHz audio stream, encoded at the same bit-rate!

It's not a metadata flag, IIRC the 24/96 extension data is stored is 384kbps of the full rate 1509 kbps, which comes up as inaudible noise without a 24/96 decoder...or gets ignored. I don't remember which

Regardless I think DTS 24/96 is a sham because it's less bitrate for the audible audio in total.

 

[SeeMoreDigital]

It's not a metadata flag, IIRC the 24/96 extension data is stored is 384kbps of the full rate 1509 kbps, which comes up as inaudible noise without a 24/96 decoder...or gets ignored. I don't remember which

Regardless I think DTS 24/96 is a sham because it's less bitrate for the audible audio in total.

Whatever the dts 96/24 maybe...

Around ten years ago when I had access to the official dts encoder suite I created some basic lossy dts encodes at 48kHz (from lossless 5.1 channel sources) and barring a few bits they came out at the same file size as the dts 96/24 flagged sources.

Anybody who has access to the official dts encoder suite can perform their own comparisons...

 

[Owen Smith]

Whatever the dts 96/24 maybe...

Around ten years ago when I had access to the official dts encoder suite I created some basic lossy dts encodes at 48kHz (from lossless 5.1 channel sources) and barring a few bits they came out at the same file size as the dts 96/24 flagged sources.

That's because DTS has a fixed bit rate(s), so it is encoding to fit within that bit rate budget. It has to come out the same size, which means the 96Khz extension (which we can't hear) is taking bits away from the lossy encoding of the stuff that we can hear.