Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[kr4]

1. When playing back multichannel (5.1) DSD/DSF files as DSD in MC31 (31.0.61), all six channels play as expected.
2. When playing back multichannel (5.1) FLAC/WAV files in MC31 (31.0.61), all six channels play as expected.
3. When playing back multichannel (5.1) DSD/DSF files as PCM in MC31 (31.0.61), only five channels as SR is silent.

This is internal to MC because it is confirmed by observing MC's RT Analyzer activity as well as the display on the connected DAC.  I can rearrange routing and connections to utilize RL/RR instead of SL/SR.  When I do that, RR is missing which suggests the problem is earlier in the chain of events. 

However, if I switch back to a really late version of MC30, all channels in DSD and PCM are fine.

I am hoping that this is trivial and ends up in "Weird and wonderful (freak problems and their solutions)!"

[Matt]

I just made a 5.1 DSD track with MC.  Played it and SL and SR show signals in Analyzer.

So I don't think I'm reproducing the problem yet.  Any additional clues?  Thanks.

[kr4]

I am embarrassed to say that it is working now.  If it had been just today, I would not have asked for help but it has been a problem for about2-3 weeks.  Unfortunately, I cannot say what caused it or what cured it. 

<<UPDATE>>  The problem is solved.  It was due to the many changes I was making to expand playback from 8 to 16 channels.  Somehow, I forgot about the necessity to have Output Channels set to 5.1 (or 7.1) in order to account for the missing subwoofer channel in the DSDs that lack it.  Duh.

[kr4]

Yes, the use of 5.1 or 7.1 channel output works fine with 5.0 DSD/DSF files converted to PCM for playback.  Without it, of course, the SL is sent to the sub, the SR is sent to the SL speaker and the SR speaker is silent.   

However, when one adds "Extra Channels" (add 10 to 5.1 or add 8 to 7.1) to have 16 channels for new immersive formats, MC seems no longer to recognize the converted stream properly.  As noted above the SR channel is silent.

[AGAWA]

had similar issue;
changed to Source Number of Channels and used PE to move SL and SR.

[kr4]

had similar issue;
changed to Source Number of Channels and used PE to move SL and SR.

Thanks.  I will try that but I suspect that such a configuration will not accommodate true 5.1 (6 channel) files.

[Matt]

This gets a little deep.  When you pick extra channels, it just adds them like regular channels today.  So stereo with four extra sort of becomes 5.1.

I'll try reworking this for a coming build.  Instead extra channels will show as "Extra" in Parametric Equalizer and Analyzer.

It might take a little back and forth to get everything hammered out, so testing appreciated.  Thanks again.

[kr4]

This gets a little deep.  When you pick extra channels, it just adds them like regular channels today.  So stereo with four extra sort of becomes 5.1.

Not quite.  Stereo with four extra becomes 6, not 5.1. 5.1 (like 7.1) is special in that it leaves a slot for the Sub/LFE channel even if it isn't present.  That is how it accommodates both 5.0 and 5.1 sources without any fiddling around.  The strange thing is that adding any number of extra channels to 5.1/7.1 seems to remove that special ability. 

Unless something changes, the best solution seems to be the creation of two zones:  One for 7.1 and one for 16.

[AGAWA]

1.3.2.2 Channel Mapping for a Multi Channel Area with 5 Audio Channels
For a Multi Channel Area with 5 Audio Channels, the Channel Mapping is defined in Figure 2.
Audio Channel Number
Channel Mapping
1 Left
2 Right
3 Center
4 Left Surround
5 Right Surround
Figure 2 : Definition of Channel Mapping for 5 Audio Channels 1.3.2.3 Channel Mapping for a Multi Channel Area with 6 Audio Channels
For a Multi Channel Area with 6 Audio Channels, the Channel Mapping is defined in Figure 3.
Audio Channel Number
Channel Mapping
1 Left
2 Right
3 Center
4 LFE
5 Left Surround
6 Right Surround
Figure 3 : Definition of Channel Mapping for 6 Audio Channels

[The Computer Audiophile]

I'll add my two cents, and plenty more if anyone is interested.

Much of this configuration requires either remembering how it all works, keeping up to date on any adjustments, or re-looking it all up when needed.

Is it possible to add options for some immersive formats? This would make life so simple.

5.1.2, 5.1.4, 7.1.2, 7.1.4, 9.1.4, 9.1.6

[Matt]

In the next build I'm changing the way extra channels are handled.

Today it simply adds the extra channels to the channel count.  That makes JRSS sometimes mix the channels.

My change is to add the extra channels and label them extra.  They then show in Analyzer and Parametric Equalizer as extra.  JRSS will leave them alone.

We'll build tomorrow and it will be moved to the public later this week.  Testing super appreciated.

Thanks.

[JimH]

Is it possible to add options for some immersive formats? This would make life so simple.

5.1.2, 5.1.4, 7.1.2, 7.1.4, 9.1.4, 9.1.6

Seems like a good idea.

[kr4]

In the next build I'm changing the way extra channels are handled.

Today it simply adds the extra channels to the channel count.  That makes JRSS sometimes mix the channels.

My change is to add the extra channels and label them extra.  They then show in Analyzer and Parametric Equalizer as extra.  JRSS will leave them alone.

We'll build tomorrow and it will be moved to the public later this week.  Testing super appreciated.

Thanks.

Thanks.  Look forward to it. 

FWIW, I did not find JRSS of any use.

[dtc]

I'll add my two cents, and plenty more if anyone is interested.

Much of this configuration requires either remembering how it all works, keeping up to date on any adjustments, or re-looking it all up when needed.

Is it possible to add options for some immersive formats? This would make life so simple.

5.1.2, 5.1.4, 7.1.2, 7.1.4, 9.1.4, 9.1.6

Although much less common, it would also be nice to have formats like 3.1, 4.0, and 6.0. They are not common but do come up from time to time and there is always confusion about now to implement them.

[The Computer Audiophile]

Although much less common, it would also be nice to have formats like 3.1, 4.0, and 6.0. They are not common but do come up from time to time and there is always confusion about now to implement them.

I agree.

Anything that can remove ambiguity and be set without wondering if it's the right or only way to do it, is great.

[mattkhan]

The whole point of extra channels was to enable the jrss mix target to be decoupled from the no of output channels that would be available to the audio device. Adding a bunch of named channel outputs that have ill defined output mappings  and which aren't jrss mix targets and which don't have any other built in support in MC (eg bass management) imv makes zero sense.

In the next build I'm changing the way extra channels are handled.

Today it simply adds the extra channels to the channel count.  That makes JRSS sometimes mix the channels.

Can you be clear on what exactly "sometimes" means because i am pretty it doesn't always do that

Are you also changing how this is represented in the internal way it stores this? ie is this a breaking change to the relevant mcws calls

Is this also a breaking change for existing dsp configs?

Given the potential for things going badly wrong here, i recommend being clear on exactly how, if at all, it will break (I recall the last time, it wasn't communicated and could have caused real problems)

[Matt]

I can make the new method optional if necessary.  To start with I'm going to try always using it.

Thanks.

[mattkhan]

Ok but the question is what exactly is going to break?

[Matt]

Ok but the question is what exactly is going to break?

If you pick extra channels and use Parametric Equalizer, you'll need to update because the extra channels will now be labeled as such.

[mattkhan]

If you pick extra channels and use Parametric Equalizer, you'll need to update because the extra channels will now be labeled as such.

All channels are stored in config via their (int) index internally so why would the labelled name break that?

[Matt]

Thinking more, I'll just add the option right now and default to off.  People can help test and maybe we can default to on at a later date.  Thanks for your help.

[AGAWA]

my suggestions:

1. make speaker config  independent. This will be the first step in the setup.  (My guess is that current complexity is caused by CHANNELS declaration in DSP that is not independent function from the rest)
2. make it possible to declare which channels to use and instruct the mixer what to mix

This should allow full flexibility in configuring whatever is needed.

Take a look at HT devices (be it player or receiver/amp). System setup allows to declare speaker config at first, options are present / not present, small or large, etc.

just my view.

[Matt]

Hi Chris.  Could you help us understand a format like 9.1.6?

What would JRSS do to build all the extra channels?

Today you could do 7.1 with 8 extra channels and mix manually in Parametric Equalizer.

Thanks.

[The Computer Audiophile]

Hi Chris.  Could you help us understand a format like 9.1.6?

What would JRSS do to build all the extra channels?

Today you could do 7.1 with 8 extra channels and mix manually in Parametric Equalizer.

Thanks.

9.1.6 is just another official Dolby Atmos configuration like 7.1.4. In my 7.1.4 system, I extract the 12 channels to WAV files using the official Dolby decoder, then play the WAVs in MC. The channel configuration has always been strange to me. I have to look it up every few months in an attempt to understand what Channels and Extra Channels mean.

It's also possible to use the Auro-3D decoding plug-in for FLAC files that will take an 8 channel file and expand it to something like 5.1.4.

It would be nice to have an exact channel selection for these immersive configurations. Just like there are 5.1 and 7.1 options, I'd love 5.1.2, 7.1.4, and 9.1.6 options. Then there is no guessing or need to wonder if I've done it right.

Removing ambiguity is always good.

[mattkhan]

there are 3 things in here

1) JRSS mix target, if any
2) no of output channels
3) channel names to display in DSP studio

the addition of "extra channels" meant 1 & 2 were decoupled, you could now choose to target (e.g.) 5.1 while running some multiway active setup or having lots of subs

since internally, jriver refers to channels via an index (2-9 for the existing named channels, 11-12 for User1/2, 13-36 for channels 9-32) and since JRSS has no idea how to mix to such formats then, as far as I can see, providing an option like 9.1.6 just means changing the labels which are visible through DSP studio when a particular channel index is referenced in a filter (or whatever).

is there a published list of how those channel names actually map to physical numbered channels btw? I don't recall ever seeing such a thing published in any specs? I see https://www.quadraphonicquad.com/forums/threads/mmh-version-6-3-new-dolby-atmos-helper-tool-updated-multchannel-remix-tool.32969/ says

16 channel (9.1.6): L, R, C, LFE, Ls, Rs, Lrs, Rrs, Lw, Rw, Lfh, Rfh, Ltm, Rtm, Lrh, Rrh
12 channel (7.1.4): L, R, C, LFE, Ls, Rs, Lrs, Rrs, Lfh, Rfh, Lrh, Rrh
10 channel (5.1.4): L, R, C, LFE, Ls, Rs, Lfh, Rfh, Lrh, Rrh)
8 channel (5.1.2): L, R, C, LFE, Ls, Rs, Lrh, Rrh)

but not sure where that comes from

btw, the "problem" with MC is that it supports a wide range of mix operations which render those labels largely irrelevant (for people who use such capability)

[The Computer Audiophile]

is there a published list of how those channel names actually map to physical numbered channels btw? I don't recall ever seeing such a thing published in any specs? I see https://www.quadraphonicquad.com/forums/threads/mmh-version-6-3-new-dolby-atmos-helper-tool-updated-multchannel-remix-tool.32969/ says

but not sure where that comes from

btw, the "problem" with MC is that it supports a wide range of mix operations which render those labels largely irrelevant (for people who use such capability)

That’s the official Dolby channel map for Atmos, the most widely used immersive format. They are numbered 1-16, starting with L.

[mattkhan]

That’s the official Dolby channel map for Atmos, the most widely used immersive format. They are numbered 1-16, starting with L.

and is that what you're after? just having correctly named channels visible in MC such that each names maps to its specific output number?

[whoareyou]

the addition of "extra channels" meant 1 & 2 were decoupled, you could now choose to target (e.g.) 5.1 while running some multiway active setup or having lots of subs

While I agree with this,  and I think this is how it should work, I'm not sure about the relationship between standard audio containers and VSTs i.e. do they need a particular container and is there some type of identifier passed to the VST indicating the audio container type in addition to channel numbering?

For example,  I require 7 channels (5.1 plus 1 extra sub) but my VST3 (HLC) will not support that configuration, and I must use 8 channel container.  This was briefly investigated by the developer, and the answer seemed to be that the framework (JUCE) does not support the 5.1 + 1 configuration.  The solution was to use 7.1 or 5.1 + 2( 7.1 equivalent) standard container configurations and have 1 unused channel.  I have no idea about the API's or what the data looks like, so I don't want to put words into anyone's mouth about how this works programatically, but that was my experience. 

[mattkhan]

While I agree with this,  and I think this is how it should work, I'm not sure about the relationship between standard audio containers and VSTs i.e. do they need a particular container and is there some type of identifier passed to the VST indicating the audio container type in addition to channel numbering?

For example,  I require 7 channels (5.1 plus 1 extra sub) but my VST3 (HLC) will not support that configuration, and I must use 8 channel container.  This was briefly investigated by the developer, and the answer seemed to be that the framework (JUCE) does not support the 5.1 + 1 configuration.  The solution was to use 7.1 or 5.1 + 2( 7.1 equivalent) standard container configurations and have 1 unused channel.  I have no idea about the API's or what the data looks like, so I don't want to put words into anyone's mouth about how this works programatically, but that was my experience. 

I think that's quite normal and it's why extra channels was implemented albeit that particular example was already supported (5.1 in a 7.1 container was a common special case which is now implemented as 5.1 with 2 extra channels).

[The Computer Audiophile]

and is that what you're after? just having correctly named channels visible in MC such that each names maps to its specific output number?

“Just?”

I have music that’s 2.0, 4.0, 5.1, 7.1, 5.1.4, and 7.1.4 and it’s all output to a 12 channel system. MC is mysterious in that it lists things as channels and extra channels. No audio systems have anything called extra channels. Thus, people have to page through forums and documentation to try to play music, when it could be easier.

If people don’t want 5.1.4, 7.1.4, or 9.1.6, then why have 5.1 or 7.1?

We all have enough things to remember in our digital lives. Anything that makes things simpler is good in my book.

Perhaps I’m barking up the wrong tree by asking for simplicity.

[The Computer Audiophile]

I will also add that I was troubleshooting a 4.0 audio issue for weeks, trying all kinds of channel output settings in MC. I ended up outputting as the source number of channels and dealing with everything externally in another app. There’s probably a way to make it work the way I need it to work, but it wasn’t simple and I had no more time to spend.

As immersive formats gain ground and MC is used as one of few apps to support this stuff, it would be great for some simpler items to remain simple.

[Hendrik]

MC primarily tracks channel counts internally, it does not really properly track what speaker a channel represents. With that limitation in mind, those immersive formats are really not something that fits into the concept of MCs audio processing right now, because there is no fundamental difference to MC between 5.1.2 and 7.1, its both just 8 channel.

This should be fixed properly at some point, so that channel are properly accounted for. But for the time being, MC does not even support decoding any "immersive" audio formats.

[The Computer Audiophile]

MC only tracks channel counts internally, it does not really properly track what speaker a channel represents. With that limitation in mind, those immersive formats are really not something that fits into the concept right now, because there is no fundamental difference to MC between 5.1.2 and 7.1, its both just 8 channel.

This should be fixed properly at some point, so that channel are properly accounted for. But for the time being, MC does not even support decoding any "immersive" audio formats.

We all decode it externally to 8, 12, or 16 channel WAV files that are in our MC libraries.

[mattkhan]

Yes just, technically it's just a human readable label for some number. It's not exactly a complicated thing in that sense regardless of how impactful it might be to however many users use MC to playback such content.

Given MC doesn't decode this content then I would think you would also need a way to tell it what the source format is whereas DSP studio is in control of the output format is (otherwise it still won't reliably work as you want as far as I can see).

[The Computer Audiophile]

Yes just, technically it's just a human readable label for some number. It's not exactly a complicated thing in that sense regardless of how impactful it might be to however many users use MC to playback such content.

Given MC doesn't decode this content then I would think you would also need a way to tell it what the source format is whereas DSP studio is in control of the output format is (otherwise it still won't reliably work as you want as far as I can see).

There’s nothing to decode. It’s 12 channel WAV files.

[mattkhan]

There’s nothing to decode. It’s 12 channel WAV files.

I know, that's why I said "given MC doesn't decode" which means it's just some multichannel wav file with no knowledge of what produced it. It means one general purpose way to solve that problem is to add a tag like "channel layout" which then tells MC what those channels mean and it can map to the specified output format as necessary.

In an ideal world, and for a truly comprehensive solution, you'd be able to specify whether PEQ applies to input channels or output channels (e.g. PEQ1 is input, PEQ2 is output) and also be able to understand how that is changed by some other DSP (e.g. convolution).

The ideal world is obviously far more work than just labelling some channels though but still it would be a rather nice, configurable & hopefully easy to use UI for the DSP pipeline.

[AGAWA]

I will also add that I was troubleshooting a 4.0 audio issue for weeks, trying all kinds of channel output settings in MC. I ended up outputting as the source number of channels and dealing with everything externally in another app. There’s probably a way to make it work the way I need it to work, but it wasn’t simple and I had no more time to spend.

As immersive formats gain ground and MC is used as one of few apps to support this stuff, it would be great for some simpler items to remain simple.

Simple with "Channels" set to 4. No upmixing or downmixing. Or SN of Ch, move SL and SR right behind L and R ( when I play 4.0 or 5.1; for 7.1 it probably will be different as 7.1 is not following 5.1 scheme).

This is the reason I suggested complete overhaul of DSP. Make channels selection (speaker setup) independent from logical channels (HW and SW as separate entities), then manipulate logical channels as you wish. Tell the mixer which channels to use in a mix.
In addition in PE, under Order Channels - show channels for each zone configuration called (from 2 upwards, 2, 3, 3.1, 4.0, 4.1 , 5 5.1 and so on). Now there are 10 entries for everything.
I would like to be able to mute channels on input , so they don't get mixed if not necessary. Maybe it is there but they work in a strange way. I mute several  channels and only one (for example) is shown as Mute. After several iterations it can be done. But strange.

Just my view.

[The Computer Audiophile]

Simple with "Channels" set to 4. No upmixing or downmixing. Or SN of Ch, move SL and SR right behind L and R ( when I play 4.0 or 5.1; for 7.1 it probably will be different as 7.1 is not following 5.1 scheme).

With channels set to four, the audio just goes out to channels 1, 2, 3, and four. Rear left is in the center and rear right in the LFE. Not that simple.

[AGAWA]

MDG SACDs are 2+2+2 and i do not want to mix height channels to my 4 speakers (height channels are in C and Sub). And this config allows to do what I want. Simple. But I need to move SL and SR right after L and R (from 6 channels that are in the recording) .
I only suggested what  I think could make it all more flexible. Separating HW from SW (speakers - physical , channels - logical).

[whoareyou]

So, can someone explain how the General/Advanced/New Extra Channels System is different from the old system?  Thanks.

[whoareyou]

Ok, I reread entire thread and realized I missed the following:

Today it simply adds the extra channels to the channel count.  That makes JRSS sometimes mix the channels.
My change is to add the extra channels and label them extra.  They then show in Analyzer and Parametric Equalizer as extra.  JRSS will leave them alone.

But, I still have one change request for all of that I think would help.  Adding output channels will sometimes change the type of output container from one to another type i.e. 5.1 input can become 7.1 output mapping, and I believe these mappings conform to some type of standard?

JRiver must know how the selected output is mapped, so how about a simple indicator, as simple as a text string in UI, that indicates output container being used i.e. none, 5.1, 7.1, others??

Thanks