Dialog Norm Levels for Music Blu-rays

[jimfisheye]

Does anyone know if the dialnorm value affects the Dolby encoding?

Recap: The encoded renders sound just like the final mixes when I A/B. They still did (at a glance anyway) when I errantly made my first couple test encodes with the dialnorm not set.

Encoded TrueHD+Atmos is in fact lossy it turns out. Even if only slightly. That means we can't even do a quick analytical null test against a mix master. So... no null testing of one dialnorm vs another either.

But the question kind of came up above - or at least this is how I interpret it. The lossy process Dolby uses to phase manipulate the object audio back out of the bed tracks after they mix it into them for the encode... Does using a dialnorm value other than -31 further degrade this?

I think that's a great question!
My guess is no. Based on the couple of test runs where after normalizing the volume between them they sounded the same in an A/B listening test.
It's way too easy a target but there are probably just some rogue hardware disc players with weird behavior and half implemented features.

 

[pat bateman]

You've made this statement before, but after investigating and asking a user with a 2020 Marantz 7015 who has actually tested some of the known problem tracks, this setting has absolutely no effect which is not entirely surprising as this would be huge news. This was also confirmed by another user.

Can you please say exactly which model Denon/Marantz AVR(s) and which known problem atmos tracks have you tested and from which music service(s)?

Note that there has always been a way on Denon/Marantz AVRs to ignore Dolby Dialnorm on a track by track basis which is essentially useless.

Denon X3700H. I use Tidal on an Nvidia Shield. I dont have my Apple TV hooked up right now, otherwise I would try that as well. Lots of people replying to this thread with their successes and failures, and the consensus was that most 2019+ Denon/Marantz AVRs can defeat dial norm. The video attached here should paint the picture in my case. Fantastic Negrito - Have You Lost Your Mind Yet.

 

[dabl]

Denon X3700H. I use Tidal on an Nvidia Shield. I dont have my Apple TV hooked up right now, otherwise I would try that as well. Lots of people replying to this thread with their successes and failures, and the consensus was that most 2019+ Denon/Marantz AVRs can defeat dial norm. The video attached here should paint the picture in my case. Fantastic Negrito - Have You Lost Your Mind Yet.

Thanks, finally some solid info, as I'd expect from an avsforums thread.

Now the reports of a tiny minority of people seeing success makes sense.

At least I know it's possible to buy a solution. Sort of. I haven't really sussed out what is lost when Dialnorm is effectively disabled in this way, for everything, for instance with movies and television. Having to change it back and forth all the time would get old for sure. Also I'd kind of worry about firmware updates breaking this.

As previously discussed Apple of course does not have the problem since they use Dolby MAT to deliver their atmos streams, which mitigates or eliminates the issue completely.

I'd still like to see a test with an actual problem atmos record, one with gross Dialnorm discrepancies between some tracks.

The Dialnorm values for the Fantastic Negrito - Have You Lost Your Mind Yet tracks are all within a -dB or so of each other. None are anywhere near -31 dB which is what your demo is showing.

 

[SeeMoreDigital]

Why Dolby thought there was need to create Dialnorm in the first place is beyond me. If it was originally created for movies then yeah okay but for audio discs I don't get it at all...

 

[dabl]

Why Dolby thought there was need to create Dialnorm in the first place is beyond me. If it was originally created for movies then yeah okay but for audio discs I don't get it at all...

What I'm getting out of the avsforums thread and re-visiting this is, there is no real arguable benefit to it and zero reason to worry about disabling it if possible for everything.

 

[AYanguas]

Why Dolby thought there was need to create Dialnorm in the first place is beyond me. If it was originally created for movies then yeah okay but for audio discs I don't get it at all...

I remember having read in some article on the Internet that there is a tool to, by going through the entire file, automatically calculate the "average level" of the dialog and obtain the value of DialNorm. As designed by Dolby for films, TV commercials, and such. Of course, the average "dialogue" level is meaningless in music and the calculated value can be very arbitrary depending on the possible detection of vocals volume. (And what about an instrumental song?)

It could be that some of the tools that Mastering engineers use may include this tool and be used as a straight forward procedure (from Dolby) to make the final encode. As said, It should be available a switch to use or not use this 'automated calculated' DialNorm, and the possibility for the sound engineer to set manually the DialNorm value?

I cannot understand the apparent "ignorance" that those in charge of the final encoding mastering seem to have, given the arbitrarily different values that are observed in some editions.

I'm sure that somebody at the final QA test to listen the manufactured Blu-Ray (or encoded file for streaming) has had to notice that the Atmos version sounds lower volume than the Stereo version. Really nobody cares at those production releases?

 

[dabl]

I remember having read in some article on the Internet that there is a tool to, by going through the entire file, automatically calculate the "average level" of the dialog and obtain the value of DialNorm. As designed by Dolby for films, TV commercials, and such. Of course, the average "dialogue" level is meaningless in music and the calculated value can be very arbitrary depending on the possible detection of vocals volume. (And what about an instrumental song?)

It could be that some of the tools that Mastering engineers use may include this tool and be used as a straight forward procedure (from Dolby) to make the final encode. As said, It should be available a switch to use or not use this 'automated calculated' DialNorm, and the possibility for the sound engineer to set manually the DialNorm value?

I cannot understand the apparent "ignorance" that those in charge of the final encoding mastering seem to have, given the arbitrarily different values that are observed in some editions.

I'm sure that somebody at the final QA test to listen the manufactured Blu-Ray (or encoded file for streaming) has had to notice that the Atmos version sounds lower volume than the Stereo version. Really nobody cares at those production releases?

Yes, if/when there are actually knowledgeable mastering engineers involved they may see that the Dolby encode step is completed correctly from the final mixed project files.

I suspect the problem is there simply are no mastering engineers involved in the production of atmos mixes destined for streaming services 99% of the time and that when the final lossless mixed project is handed over to be encoded for streaming that the problem is probably introduced.

In the case where there are gross dialnorm value differences between tracks on the same album I kind of wonder if all the individual tracks are simply being processed one at a time, perhaps by different people at some point and the settings on the encoder get set wrong for some.

Obviously all pure speculation on my part.

I do know that if someone associated with the release actually listened to the final project for quality assurances purposes it would be caught but clearly that is not happening. That's not surprising really with the avalanche of albums being encoded.

It seems like this may sometimes happen with projects destined for lossless Blu-Ray too such as the Zappa Waka/Wazoo release, and that in those cases though there aren't gross volume problems between tracks, the dolby dialnorm encoding for all the tracks is set at some level (ex -18 dB) that makes the entire project play back much quieter than some other releases. I remember Joe Travers addressing user comments about this but seemingly not fully understanding exactly why that release came out that way and apparently thinking it was just the way it had to be. My guess is that in the end, for that project, there simply was no *real* mastering engineer involved and/or that these specific technical details about the final encode process were just never discussed or explored.

 

[jimfisheye]

Perhaps a little easy to miss on the screen? I mean, it's not like anyone reads manuals.

 

[dabl]

Perhaps a little easy to miss on the screen? I mean, it's not like anyone reads manuals.

Heh exactly.

I'm curious if there's a default/out of box custom Dialnorm setting and what that is?

Do you know what the custom Dialnorm 'No' setting actually does/is?

I guess I need to RTFM

 

[AYanguas]

...
I do know that if someone associated with the release actually listened to the final project for quality assurances purposes it would be caught but clearly that is not happening. That's not surprising really with the avalanche of albums being encoded.

...

That's what I'm referring I don't understand.

The artist, or the mixer, or the people closest to the project should "listen" to the final crafted product. I think I would do it.

There is an avalanche of albums beeing encoded, but the interested people is "interested" in only one album. They really don't care?


Are we really very different species: 1) the artists, producers and technicians, sellers and 2) we listeners, geeks who realize these things?

Or another reason could be that "real Atmos listening" and its checking, with good discrete speakers equipment, is available only to a few of us?

 

[AYanguas]

https://forums.steinberg.net/t/important-dolby-atmos-and-the-dialogue-level-aka-dialnorm/861632

This is a very short page to read that:

""""""""""""""""""""""
The solution to this problem is very simple:

We set the “Custom dialnorm” or “Dialnorm” in the encoder to “-31 dB”.
This disables this “feature”. If we are using the Encoding Engine (DEE), we have to adapt the corresponding script by adding the following line in the right place:

<custom_dialnorm>-31</custom_dialnorm>

""""""""""""""""""""""

 

[AYanguas]

I'm curious if there's a default/out of box custom Dialnorm setting and what that is?

Do you know what the custom Dialnorm 'No' setting actually does/is?

I don't want to think that the default is "No" for Custom dialnorm, and then the Encoder sets an "automatically" calculated dialnorm.

 

[AYanguas]

https://forums.steinberg.net/t/important-dolby-atmos-and-the-dialogue-level-aka-dialnorm/861632

This is a very short page to read that:

""""""""""""""""""""""
The solution to this problem is very simple:

We set the “Custom dialnorm” or “Dialnorm” in the encoder to “-31 dB”.
This disables this “feature”. If we are using the Encoding Engine (DEE), we have to adapt the corresponding script by adding the following line in the right place:


""""""""""""""""""""""

I have read that again, and there is a final paragraph that says that the external Atmos renderer to encode E-AC-3 JOC (Exactly the file format that Tidal streams):
the external Atmos renderer does not(!) allow you to set your own value for the dialogue level!

Could this be the source of so many Tidal Atmos tracks low volume?

 

[dabl]

That's what I'm referring I don't understand.

The artist, or the mixer, or the people closest to the project should "listen" to the final crafted product. I think I would do it.

There is an avalanche of albums beeing encoded, but the interested people is "interested" in only one album. They really don't care?


Are we really very different species: 1) the artists, producers and technicians, sellers and 2) we listeners, geeks who realize these things?

Or another reason could be that "real Atmos listening" and its checking, with good discrete speakers equipment, is available only to a few of us?

It is a little surprising that nobody that has music out there with this problem or an associated party like an engineer or producer hasn't waved their hands about it, then again most professional musicians I've known have had pretty humble playback equipment and certainly not surround systems.

Also it's key to keep in mind that the gross problem of wildly varying volumes between tracks on the same album effectively doesn't exist on the biggest service provider, Apple Music.

So most people will never hear the problem since it only effectively exists on Tidal and Amazon.

I would love to get word to Steven Wilson that his New Order 'Movement' album atmos mix got ****** up in the encode process and is streaming on Tidal with this problem since he is in a unique position to get somebody to finally ******* listen or at least fix *his* encode.

 

[dabl]

https://forums.steinberg.net/t/important-dolby-atmos-and-the-dialogue-level-aka-dialnorm/861632

This is a very short page to read that:

""""""""""""""""""""""
The solution to this problem is very simple:

We set the “Custom dialnorm” or “Dialnorm” in the encoder to “-31 dB”.
This disables this “feature”. If we are using the Encoding Engine (DEE), we have to adapt the corresponding script by adding the following line in the right place:


""""""""""""""""""""""

You should not be surprised that a post buried on a Steinberg user forum has not made the front page or even an article in widely read trade magazine such as Mix.

99% of the work in the business is not done using a Steinberg product.

 

[jimfisheye]

Heh exactly.

I'm curious if there's a default/out of box custom Dialnorm setting and what that is?

Do you know what the custom Dialnorm 'No' setting actually does/is?

I guess I need to RTFM

'No' is default and you get the calculated dialnorm. Which, as discussed, doesn't apply to and work with music mixes and gives you a random turned down result.

 

[SomethingAnything]

I'm still not quite sure what people are referring to when they talk about Dialnorm issues. Dialogue not kept to a normal level? Which somehow screws up music?

Yesterday I checked out Townsend's excellent PowerNerd on Tidal, and wow, some tracks sure jump around in volume at times. Perhaps this is a Dialnorm issue.

 

[jimfisheye]

The aim was/is to hold dialog in broadcast at a consistent intentional level. Think news broadcast and commercial diatribe. Music mixes don't mix voice like that. Not even lead vocals. Some mixes can get extra creative. Trying to make the news broadcast autopilot triggering off voice levels work with that leads to unexpected and random results.

The instruction for Atmos encodes simply becomes: "Set that to -31"

 

[mandrix]

A friend and forum member who mixes for Atmos once did a test, where he took the same files and encoded to Atmos at different dialnorm levels.
He asked several people with different AVR's to test them and see if volume differences were noted between the different dialnorm levels.
On my Onkyo RZ50 I could detect zero change in volume. Neither could another person with a Denon AVR, don't recall the model. There was one person in the group that the playback volume was very noticeable on his system.

I don't have any special settings I'm aware of for defeating dialnorm on my AVR, it just does apparently?

 

[dabl]

A friend and forum member who mixes for Atmos once did a test, where he took the same files and encoded to Atmos at different dialnorm levels.
He asked several people with different AVR's to test them and see if volume differences were noted between the different dialnorm levels.
On my Onkyo RZ50 I could detect zero change in volume. Neither could another person with a Denon AVR, don't recall the model. There was one person in the group that the playback volume was very noticeable on his system.

I don't have any special settings I'm aware of for defeating dialnorm on my AVR, it just does apparently?

Interesting. I don't know the answer but remember some Onkyo models have been mentioned in the aforementioned avsforum thread which is the most comprehensive exploration of this I've seen in terms of how any given AVR handles dolby dialnorm. Perhaps the default setting is off for that one and possibly the other Denon? This kind of testing is exactly what is documented in that avsforum thread but including many models.

I have to say I'd kill for one of those models that had zero change in volume.

It's becoming clear to me now how someone waving their hands about this might be looked at as a bit daft. If one didn't have the problem one might simply shrug their shoulders and think 'not sure what's up with that poor *******'s system'.

I can tell you if you *do* have the problem and have *any* kind of ear, it will drive you straight out of your mind, or at least to Apple Music and simply be done with it.