Marantz AVR issues with DSOTM Atmos {Pls read first post)

[barfle]

While Atmos as an object based system, the location of any object is still ultimately defined by the relative volume levels of the individual speakers. Therefore, a test tone should be able to be defined and located by the volume of a certain speaker with all of the other speakers having -0- volume.
Also, when playing the Demolandia test tone file, the AVR clearly sees them as Dolby Atmos.

Well, the relative volume and phase of several speakers can determine a phantom location of an object. Again, this is all “as I understand it.”

If an object is intended to be, for example, in a location slightly to the left of the room and slightly forward of the center, the decoding may well determine that it’s supposed to go only to the left front overhead speaker, IF THAT’S WHERE THE SPEAKER HAPPENS TO BE LOCATED IN THAT PARTICULAR ROOM. But every room is different (well there might be some duplicates), so a discrete channel selection would be a contradiction to the object orientation of Atmos.

I would usually expect some bleeding from overhead speakers simply because the volume and phase relationships are necessary to locate an object.

 

[JohnN]

Back heights test tone comes from the front heights if you don't have back heights. Side test tone comes from the rear speakers when you don't have side speakers. All others match.

 

[ar surround]

Well, the relative volume and phase of several speakers can determine a phantom location of an object. Again, this is all “as I understand it.”

If an object is intended to be, for example, in a location slightly to the left of the room and slightly forward of the center, the decoding may well determine that it’s supposed to go only to the left front overhead speaker, IF THAT’S WHERE THE SPEAKER HAPPENS TO BE LOCATED IN THAT PARTICULAR ROOM. But every room is different (well there might be some duplicates), so a discrete channel selection would be a contradiction to the object orientation of Atmos.

I would usually expect some bleeding from overhead speakers simply because the volume and phase relationships are necessary to locate an object.

If what you are saying is true, then someone needs to explain to me why this Demolandia test tone file that I am using is picked up by my Marantz SR7013 as a Dolby Atmos stream, and then able to to put each test tone exactly in where it belongs in each of the 11 speakers.

 

[himey]

Again, as I understand it, the idea that an object could be placed in an individual speaker where all objects are placed in phantom positions is a contradiction. Although Dolby has recommended speaker positioning, they are not exact locations, so positioning an object in an xy plane would have to be calculated from the locations in a particular room. It’s possible that one or more of those positions would be in the same location as a speaker, but I don’t see that as a hard-encoded channel in an object oriented system.

We’re used to mono, stereo, 5.1, 7.1, and those are all channel-based systems. Many recordings have test tones that go to a specific speaker, but Atmos is not channel-based, so a discrete placement IN A SINGLE SPEAKER seems contradictory, at least to my retired A/V engineer brain.

Are you taking into account the fact that Atmos decodes to specific layouts like 7.1.4, 5.1.4, 5.1.2 ect? There is no reason test tones shouldn't work across the board.

 

[Albert Dagger]

Has someone tried ripping the test tones and decoding them to a 7.1.4 wav file? I think it’s important to exclude any issues with the test tones themselves before believing it has to do with the AV/AVR.

As far as I understand, it’s an isolated issue with the test tones of this disc, and might not even affect the album mix.

 

[César]

Has someone tried ripping the test tones and decoding them to a 7.1.4 wav file? I think it’s important to exclude any issues with the test tones themselves before believing it has to do with the AV/AVR.

As far as I understand, it’s an isolated issue with the test tones of this disc, and might not even affect the album mix.

It also affects to the album playback. Someone described in the album thread how the voices in the heights are played (discretely) and in my case are duplicated by pairs.
But you are right, I need to compare the ripped files (don't think I'll see anything, as I bet they will be properly decoded by the laptop application)

 

[Albert Dagger]

It also affects to the album playback. Someone described in the album thread how the voices in the heights are played (discretely) and in my case are duplicated by pairs.
But you are right, I need to compare the ripped files (don't think I'll see anything, as I bet they will be properly decoded by the laptop application)

When I tried the test tones with my TML and TMR speakers, both front and rear height played back correctly in the top middle channels, as it happens for all other mixes. With this configuration, I can’t have auro enables, I think.

I had the feeling that the rear heights were lower in volume but this can be speculation as I didn’t measure the spl. So this problem doesn’t affect my specific setup.

It would be good to compare playback of the disc vs the ripped file.

 

[barfle]

If what you are saying is true, then someone needs to explain to me why this Demolandia test tone file that I am using is picked up by my Marantz SR7013 as a Dolby Atmos stream, and then able to to put each test tone exactly in where it belongs in each of the 11 speakers.

Well, that confuses me, too. Of course, I’ve repeatedly stated, “as I understand it,” which may well mean that my understanding is limited. I’ve looked (a little) for some technical papers on the format, but it seems like Dolby wants to keep the details to themselves.

 

[barfle]

Are you taking into account the fact that Atmos decodes to specific layouts like 7.1.4, 5.1.4, 5.1.2 ect? There is no reason test tones shouldn't work across the board.

Those layouts (x.y.z) are hardware specifications, and the first two digits are certainly channel specific, and have been since DVDs came with sound encoded that way. When the “immersive” format hit, Dolby was touting their ability to place “objects” around the space without the overheads being “channel” based. A discrete tone from a single speaker is the definition of a channel, and Dolby claims that Atmos is NOT channel-based as 5.1.0 certainly is.

 

[César]

Got a little spare time today, saved to USB my current configuration and started another one from scratch, including Audyssey calibration.
Physically, I have 5.1.4 with rears around 5 degrees behind me and top rears right on top of the surrounds, so the logical configuration for me is SHL and SHR (and not RHL and RHR that is prepared for having your rear heights more on your back and in line with Auro config).
So setup done, Audyssey ok and test tones from AVR ok too. But then, any Atmos test tone or Atmos content fails to be played from the rear height speakers. Nothing!
The only way for the rear heights to behave is to configure them as RHL and RHR. And in this particular scenario is where all Atmos content is routed fine, apart from DSOTM.
I give up, as it's only a single disc that seems to fail and all the rest are ok, I'll leave this configuration. Yet the SHR/SHL setup should work too, so this leads me to strongly believe it's a Marantz bug.
End of the story for me, I'll not get obsessed with this, got enough complications on my life. If anyone finds a solution, pls let me know.
And thanks for helping so far!

 

[JohnN]

https://www.quadraphonicquad.com/forums/threads/marantz-7706-7015-atmos-decoding-problem.32677/

 

[César]

https://www.quadraphonicquad.com/forums/threads/marantz-7706-7015-atmos-decoding-problem.32677/

thanks, but completely unrelated issues

 

[JohnN]

thanks, but completely unrelated issues

Marantz bug search of Google for that. So famous, QQ.

 

[himey]

Those layouts (x.y.z) are hardware specifications, and the first two digits are certainly channel specific, and have been since DVDs came with sound encoded that way. When the “immersive” format hit, Dolby was touting their ability to place “objects” around the space without the overheads being “channel” based. A discrete tone from a single speaker is the definition of a channel, and Dolby claims that Atmos is NOT channel-based as 5.1.0 certainly is.

Once you start messing around with decoding Atmos into different speaker configurations, I am sure you will come to a different realization. The objects get moved, due to the different speaker layouts. There is a fixed number of layouts.

 

[ar surround]

Marantz bug search of Google for that. So famous, QQ.

Yes, there have been several times where a Google search has directed me to a QQ thread...and sometimes my own post.

 

[JohnN]

Yes, there have been several times where a Google search has directed me to a QQ thread...and sometimes my own post.

I blame all on the Marantz noise gate.

 

[César]

Well, I found a (not so good) solution. I ripped the disc, decoded the Atmos into wav files using MMH (5.1.4 setup) and reencoded into Atmos again. I did it first for the test tones and for once they played fine and then for the whole record. Quick check at 25:10 and now everything plays as per the post below:

"mrcond said:
I have the Denon AVR-x3700H, and as I indicated previously, the channel ID sounds and pink noise all mapped perfectly on 7.1.4 setup. Here is where the voices at the end of Money and beginning of Us and Them are coming from:

(Starting at time 25:15 - using the speaker naming conventions from the DSOTM Atmos setup menu)
"(laughter) I was in the right" = Left Front + Right Front
"yes absolutely in the right" = Left Rear Surround
"I certainly was in the right" (#1) = Right Top Front
"that geezer was cruising for a bruising" = Right Rear Surround
"why does anyone do anything" (#1) = Right Side
"why does anyone do anything" (#2) = Left Side
"I was really drunk at the time" = Left Top Rear
"...yelling and screaming..." = Left Top Rear
"I certainly was in the right" (#2) = Left Front
"I was really drunk at the time" = Left Top Front"

Now don't ask me why reencoding solves the problem...

 

[ar surround]

Well, I found a (not so good) solution. I ripped the disc, decoded the Atmos into wav files using MMH (5.1.4 setup) and reencoded into Atmos again. I did it first for the test tones and for once they played fine and then for the whole record. Quick check at 25:10 and now everything plays as per the post below:

"mrcond said:
I have the Denon AVR-x3700H, and as I indicated previously, the channel ID sounds and pink noise all mapped perfectly on 7.1.4 setup. Here is where the voices at the end of Money and beginning of Us and Them are coming from:

(Starting at time 25:15 - using the speaker naming conventions from the DSOTM Atmos setup menu)
"(laughter) I was in the right" = Left Front + Right Front
"yes absolutely in the right" = Left Rear Surround
"I certainly was in the right" (#1) = Right Top Front
"that geezer was cruising for a bruising" = Right Rear Surround
"why does anyone do anything" (#1) = Right Side
"why does anyone do anything" (#2) = Left Side
"I was really drunk at the time" = Left Top Rear
"...yelling and screaming..." = Left Top Rear
"I certainly was in the right" (#2) = Left Front
"I was really drunk at the time" = Left Top Front"

Now don't ask me why reencoding solves the problem...

Nice work, César. Thanks for your efforts as it took quite a bit of work to get to this point. However, the Atmos mix of the album sounded just fine to me with the voices coming from the general locations noted above without perfect decoding. Without this thread, I never would have known.

 

[ar surround]

Hey @barfle , if we may have seemed confused before about Atmos, can you imagine how a dude who has this layout must feel? I know I've been mad...I've always been mad...

 

[JohnN]

Hey @barfle , if we may have seemed confused before about Atmos, can you imagine how a dude who has this layout must feel? I know I've been mad...I've always been mad...






View attachment 97858

Love/laugh.

This is comfortable for now: