Plug-ins with different number of outputs to inputs

Currently, as far as I’m aware, there is no support in Audirvana for plug-ins which may have a different number of outputs to the number of inputs.

Personally, I use a number of these.

The input would typically be stereo, then the plug-ins I’m using would decode and map various surround sound formats embedded in the stereo signal to their full surround sound, multi-channel formats. The input would be 2 in and 4 or more channels out.

Obviously the audio device being used for the output would have to support the number of channels that corresponds with the output.

So you could consider this a feature request. In my view though, if you support VST/Au plug-ins, that should include this category of plug-in.

Otherwise you should state in your documentation or the interface that only stereo in/stereo out plug-ins are supported.

1 Like

As I remember… you are using a macOS platform. The output configuration is managed in Audio MIDI Setup where the system defines the number of speakers/channels in the playback scenario…

:notes: :eye: :headphones: :eye: :notes:

I’m sorry; but you’ve not quite understood the issue here. I’m well aware, of course, of how to use Audio MIDI Setup on MacOS to configure the number of output/speakers.

What I’m talking about is something quite different.

We receive as input to the plug-in, what appears to be a regular stereo signal. In fact though it contains an Ambisonic or QS (Quad) or SQ (Quad) encoded sub-signal.

The plug-in is capable of decoding any of these to Quadraphonic 4 channel output. Or in the case of Ambisonic two plugins can be chained together. 1/ To decode to Ambix B-Format, and another to map the B-Format to whatever speaker configuration the user has (and correctly configured beforehand using Audio-Midi setup), 5.1, 7.1…;etc.

I hope that’s clearer now.

Okay… thanks for that detail…
Obviously, Audirvāna would need to recognize the file format output routing where this can be configured to output to the correct channel(s)… I don’t see many recorded/encoded products produced in these formats as compared to multi-channel PCM and DSD music recordings… The obvious obstacle is in the value of support for a minority of users…

:notes: :eye: :headphones: :eye: :notes:

“Obviously, Audirvāna would need to recognize the file format”

Actually no! Nothing special about the file format at all. Plain old flac streams from Qobuz; etc. that contain these encoded signals. Think of them as kind of hidden easter eggs :grinning:
As for content. Well for classical, including Indian classical, flamenco and other “world” music, practically the entire Nimbus catalog was encoded in Ambisonic UHJ stereo and can be decoded to Ambix/Quad. Yes you used to need hardware decoders for this; but now it can be done with the magic of software.

If you’re old enough to remember Quad LPs from the 70’s, something like 3,000 albums of top artists of the time, in Jazz (John Coltrane, Alice Coltrane) Soul; Temptations, Rock Steely Dan, Deep Purple; just to give a very few examples, were encoded onto stereo LPs and the Quad signal was retrieved using a specific cartridge and, again, hardware decoders. A large number of these encodings made it onto CDs and the same encoded signals are out there on our favorite streaming services.

Here’s a link to a QS playlist I created on Qobuz

… and another, that I believe to be QS encoded.

I had a discrete Quadraphonic 8-Track in my van… I had a Laser Disc and CD store… I have SACD in 5.1 and DVD-A in 5.1… :sunglasses: Looking forward to support for modern multi-channel audio productions… but the older stuff if not in discrete DSD or DVD-A or DXD multi-channel does not cut-above the high-resolution stereo masters of these recordings, from my perspective…

:notes: :eye: :headphones: :eye: :notes:

… well these are CD quality, that’s true. For some of us with immersive systems though, the advantages of 4 or more discrete channels, may exceed the differences of higher rez stereo formats over the CD format. Of course YMMV, as always.

I have had systems built around all of these surround formats… I am currently headphone-centric and utilize HRTF plugins… I have the Waves Audio Abbey Road Studio 3 virtualization for multi- channel surround playback… The Lake TheaterPhone/Dolby/Zoran virtualization system was my design concept… When I say that discrete high-resolution DSD and PCM multi-channel playback is superior to older matrix encodings, this a perspective derived from an intimate relationship with psychoacoustic science and the application of the technology. A expertly designed high performance high-resolution stereo playback system can deliver an immersive experience with contextual harmonic, dynamics and spatiality that a matrix surround production can never convey… The discrete surround experience is a different animal. :wink:

I have no arguments with you there. Where any of these QS or SQ encoded recordings also exist on DVD-A, SACD or Blu-Ray in full 5.1, and when you like the music enough you’d usually be better off with the disc. Having said that, appreciation of any surround sound format is very dependent on the mix. At QuadraphonicQuad.com you’ll see plenty of discussions of the mixes of music on various surround formats. Members there may prefer one of the older Quad mixes to any newer formats including Dolby Atmos, for particular recordings, because of their preference for the older Quad mix. A number of them collect the original Quad LPs and have even put together Quad systems from the 70’s to replay them.

But hey, I’m just pointing out that they’re out there in abundance on the streaming services, and can be decoded using software plug-ins. Of course, I do most of my listening in Stereo and wouldn’t prioritise these over whatever music I happen to be into at the time, which only exists in Stereo. Many of the Jazz, Soul and Rock albums that are on the streaming services, QS or SQ encoded, do correspond with my tastes though. The plug-ins that decode them do a great job, and you will hear separate instruments from the rear channels, than from the front channels (discrete in that sense).

This provides further separation of instruments, which for me personally provides greater insight into the music, and greater appreciation of it.

Ambisonic is another animal. If you’re into classical or any of the other acoustic music that tends to be in Ambisonic, these are on Qobuz; etc. encoded as UHJ stereo and can be decoded to any number of channels you may have. Unfortunately something was lost in this UHJ encoding when compared to original B-Format recordings which can also be mapped to the number of speakers you may have, some of which are truly spectacular.

Ambisonic is a kind of forerunner of Atmos; but is still relevant today. In fact I think it inspired Atmos to some degree at least.

Many of the UHJ encoded Ambisonic recordings from Nimbus, just feature a single instrument. So no instrument separation there. What is the point you may ask? Well when the solo instrument is decoded you have a much better sense of that instrument in the room where it was originally recorded, than you have, if you just play it back in stereo. Of course Nimbus and Ondine also have plenty of multi-instrument recordings, including full orchestra.

So, to not ramble on too much longer, all I’m saying is they’re out there and can be decoded.

For anyone who has at least a 4 channel system, why not enjoy them decoded with the appropriate plug-in(s)?

I have a solution for this using Foobar2000, which can output more channels than it received as input to a plug-in.

I would love to be able to do this in Audirvāna instead though. Hence my reason for starting this thread.

1 Like

@Agoldnear This shortcoming was clearly explained by Mitch Barnett in this thread…

The mix is everything… I’ve had at least 4 different surround format mixes of Santana’s “Abraxas”… The 5.1 DSD/SACD iteration is the best of them… it is not gimmicky and presents the work with respect to the 2-track master(s).

You will find there are folks here on this forum, including myself, that will speak to the quality of the room acoustics management in fulfilling the true surround soundscape. As a practicing sound recording/acoustics engineer, I was a THX certified installer at one point in my career… The dream is to be able to usurp the acoustic environment in which the playback array projects into and recreate the captured or synthetically produced environments imbued in the surround mix. I have seen incredibly stupid surround and stereo speaker arrays implemented in the average home environment… Very few folks have a multi-channel speaker array that is aligned and calibrated properly in the acoustical environment they are projecting…

I have several DTS surround music discs… I would rather listen to these discrete channel products than to listen to any SQ or QS matrix of the same recording. At an AES convention in San Francisco several years ago, I had a discussion with Alan Parsons regarding his surround production strategy and specifically his DTS project “On Air”, (a very entertaining and engaging work…) I wish it were available as 5.1 DSD/SACD (I rip stereo SACD, then decimate to 352.8kHz PCM for HRTF processing and subsequent of the HRTF DSP, these files are modulated to DSD128)

I have listened to great surround mixes that were very engaging and those that take liberty with a lot of movement that completely changes the reference to the original stereo master recording and those that are just giant stereo mixes. I find that single microphone stereo recordings capture a sense of being there, generally equivalent to binaural recordings of the same nature… I have a collection of native DSD binaural recordings, from which to compare to general stereo products… The multi-channel surround speaker array brings a level of entertainment to the listening experience that is different than the stereo experience… However, it is not always more detailed or spatial or even entertaining… They can be just as boring and irritating as some stereo mixes…

What the average Audirvāna user will not do, is build-out a speaker array larger than 5.1… And for excellent playback, this array will be configured with matching speakers of good or excellent quality if one wants to reproduce the high quality of sound presented by Audirvāna… Audirvāna already supports native, discrete multi-channel DSD and PCM recordings.

So in the end, with music productions… there is a steep investment for implementation of multi-speaker surround playback…

So… This is where psychoacoustic multi-channel playback comes to play… Systems like the Yamaha 'Virtual Cinema Front" surround projection, Dolby Virtual Speaker, DTS Virtual-X, and theoretically the u-BACCH plug-in that you employ among other psychoacoustic surround systems.

It’s all entertainment… To what degree does faithful reproduction of the encoded source mean to the individual, is the salient question… Give me the highest resolution encoding that is feasible in a virtualized binaural psychoacoustic presentation.

I love the DSD 5.1 iterations of familiar recordings and the DVD-A 5.1 iterations of familiar recordings… Generally the DTS 5.1 music products are very engaging, however I can easily discern qualitative limitations in comparison to the pure lossless PCM stereo master… Psychoacoustic binaural virtualization is the future… not more speakers… :sunglasses:

Thanks for your valuable insight. I do get a lot of pleasure from the uBAACH and x-Talk Shaper plug-ins.

Despite the limitations of the QS and SQ formats, I find the results often entertaining and even more revealing of the intentions in the music.

If they could encode Quad/Ambisonic into stereo carrier signala way back in the 70s, I can’t help wondering what a similar approach could achieve today, with updated tech.

I guess Atmos is the nearest to what I’m thinking about. From what I can see Apple streams these in HLS containers which have different streams, depending on the choice of the user. If it’s for headphones, essentially it seems to be a binaural stream. If for speakers, an Atmos Dolby Digital stream is decoded and mapped to the number of speakers configured in Audio-Midi set-up.

If an “immersive” mix could be encoded into stereo carrier signals, to be decoded, if desired at the user’s end, and mapped to the number of channels for a multi-channel speaker system, or binaural for headphones, these could be made available on the streaming services. Unlike Apples’ current implementation using Dolby Digital, they could be lossless.

I guess Ambisonic was ahead of its time. I believe Peter Craven was involved in that and it’s interesting that MQA used the term “unfolding”, when too me, this is a term that would be more appropriately applied to Ambisonic (or an updated equivalent) whereby an apparently plain stereo signal could be “unfolded” as desired, using the appropriate decoders.

There is no need for carrier signals to reproduce a virtual multi-channel psychoacoustic surround soundscape via two speakers… I do it all the time with Abbey Road Studio 3 virtualization in 5.1 for headphone playback monitoring of my multi-channel SACD, DVD-A and DTS discs… This is what Dolby Atmos binaural is doing for headphone playback, and APL labs “Virtuoso” is doing for headphone playback, among others. I take the DSD, DVD-A and DTS Music 5.1 signal output from my Marantz multi-disc player into a Focusrite Liquid 56 interface at 24/96kHz, sending the multi-channel signals to Apple’s Logic Pro and apply Abbey Road Studio 3 virtualization HRTF plug-in to listen to those discs in 5.1 surround via my DAC that up-samples this to 192kHz where the output is sent to my HPA…

Apparently you have little experience with 5.1 DTS music products in comparison to matrix surround… At 24/96kHz/192kHz Dolby Atmos seems to be the vehicle of choice for surround virtualization of music production… How to translate this to a pair of speakers is relatively easy… I have experience with sound-projector arrays that convey an extremely compelling surround soundscape. The key in all psychoacoustic presentations is in the management of the ambient room acoustics that interfere with the speaker to speaker sound-field relationships.

I’ve always been impressed with Ambisonic demos…
I personally feel, in the context of Audirvāna playback, the value of any CODEC support requires a majority demographic and a ubiquitous catalog of music product to support the reasoning for implementation… We don’t see a large catalog of Blue-ray music or Dolby Atmos surround music or DTS surround music that reaches a wide spectrum of potential customers…

So I agree it would be nice to be able to coherently manage the multi-channel output so to be symbiotic with any given number of outputs from a multi-channel DSP plug-in… I have been very lazy regarding testing multi-channel PCM 96kHz input through the Abbey Road Studio 3 in Audirvāna, for virtualized multi-channel surround HRTF output being modulated to DSD128… However, when I do get around to it… I’ll provide insight to my experience… but it is low priority for me…:wink:

Well I only have experience of DTS-Master 5.1, Dolby HD 5.1 and SACD multi-channel discs, as a consumer. I’ve ripped these to play back via my 5.1 system.

My position on the QS/SQ and Ambisonic formats is, hey they’re out there on the streaming service and can be decoded to Quad (or 5.1 plus for the Ambisonic). They’re fun and entertaining. I also feel that if a free solution exists in the form of Foobar2000 and certain plug-ins, why can’t Audirvana support different outputs to inputs too? Otherwise it’s not providing complete support for VST/AU plug-ins.

As far as the professional industry is concerned, I know there are composers and artists out there who would love to produce immersive music. In other words music that’s been deliberately created with an immersive experience in mind. They’ve felt held back up to now because few consumers have multi-channel systems.

The Apple approach seems to be the closest solution. Atmos decoded to whatever speaker array the user has, or binaural for headphones.

Tidal also has some Atmos, of course.

Personally I’m not a headphone listener. So I want a solution for speakers. The more I think about it the more I’m convinced that Binaural played through one of the cross-talk cancellation plug-ins is the way forward.

For me a good binaural recording played through these plug-ins via a stereo system provides true immersive sound. I’m always fooled that the rear speakers are playing when I play Binaural through these plug-ins.

So I wish all recordings going forward were produced Binaurally and I would listen to them via uBAACH or X-Talk Shaper. Which, of course, pose no problem for Audirvana.

I wish I could find a way to fool Apple I’m listening on headphones, so I could play the binaural streams from any of their Atmos offerings. Of course though for the time being these are still lossy.

The Apple Spatial Audio music (Dolby Atmos) is encoded as Dolby TrueHD which is a lossless codec… Dolby Labs acquired the MLP intellectual property many years ago.
Check out the Apple Spatial Audio music catalog(s)… There is a wide selection of genre that is available… So it is now just a matter of artists and producers working in a Dolby Atmos flow, all the way to distribution in the format of choice.

The writing is on the wall… multiple speaker arrays supporting different surround codecs and formats are becoming less popular every day… Most folks are not willing to clutter their space and very few have dedicated space for a these speaker arrays… Generally, the price of entry is not compelling, and for many unreachable, and difficult to get them properly aligned and calibrated… Apple Music is defining the consumer the space and is focused on virtual sound reproduction via their products.

The reality today… Surround/spatial/immersive production and reproduction is in the realm of psychoacoustic encoding… which can be supported in a single sound-bar/sound-projector… It is evolving more quickly now than ever before.

Personally speaking as a user… I will never support any added technology that diminishes the established level of performance we now enjoy from Audirvāna.

:notes: :eye: :headphones: :eye: :notes:

Last time I checked Apple Music was using Dolby Digital Plus, which is a lossy format.

Today I experimented to send Apple Spatial audio via the Hang Loose plug-in host through X-Talk Shaper to stereo speakers. This gave the immersive effect normally only enjoyed via Headphones from Apple Spatial audio, through a pair of stereo speakers.

Not quite the same as Atmos decoded to multi-channel speakers; but very effective nonetheless.

For this to work you need a virtual audio output device, such as Blackhole. You set this to be your default output device on Mac in Audio-Midi set-up, and the speaker configuration must be set to stereo.

In the Apple Music app’s preferences, in the “playback” section, set “Dolby Atmos:” to “Always On”, otherwise you’ll just get an ALAC lossy stereo stream at 256kbps.

In the Hang Loose host the input is set to BlackHole and the sample rate to 48khz. This is sent through X-Talk shaper, so the Binaural effect will work well via your speakers. The output is set to your actual audio output device, again at 48khz.

Of course, none of this is relevant to Audirvana as it doesn’t support Dolby Atmos, even though Dolby Atmos is available form Tidal. The reason for this, I believe is that the MacOS has to decode Dolby Atmos to your speaker configuration of choice before sending the output to your output device.

2 Likes

Nope…
https://professionalsupport.dolby.com/s/article/Are-there-ways-to-distribute-Dolby-Atmos-Music-as-lossless-audio?language=en_US

ALAC is lossless… AAC is lossy. :wink:

You could try Eventide’s “Blackhole Immersive” to create a customized up-mixed virtualization for all of your stereo files.

:notes: :eye: :headphones: :eye: :notes:

The link you provided refers to Dolby TrueHD, which is used for Atmos Blu-Ray discs.

Apple’s spatial audio uses Dolby Digital Plus which is lossy.

When you tell Apple your output Audio device is Stereo in Audio Midi set-up and set Atmos to “Always on” in Apple Music you get the Binaural streams normally intended for headphones.

When used with a cross talk cancellation filter you get the binaural presentation (usually only enjoyed on headphones) via speakers.

When you don’t set Dolby Atmos to “Always on” you get a plain stereo stream. I will check again if it’s AAC or ALAC. I might have got that wrong.

The Dolby Atmos Music deliverables specify PCM 24bit/48kHz …

From the Apple Video and Audio Asset Guide

Immersive Audio Source Profile

Immersive audio source must meet the following requirements.

Dolby Atmos music deliverables

  • Dolby Atmos audio files generated from stereo mixes are not allowed. Specifically:
    • A Dolby Atmos track must be created from multitracks or stems created from multitracks.
    • Upmixing from a stereo release is not allowed.
    • Extracting stems (“de-mixing”) from a stereo release is not allowed.
    • A Dolby Atmos track consisting only of a stereo mix placed in the sound field with added ambience or reverb is not allowed.
  • Provide the Dolby Atmos file as a Broadcast Wave Format Audio Definition Model (BWF ADM) file.
  • All tracks within a project must be at the same frame rate.
  • All audio must be 24-bit linear pulse code modulation (LPCM) audio at 48kHz.
  • You must conform and sync the Atmos files with the stereo reference files for the same project.
  • The integrated loudness value should not exceed -18 LKFS measured as per ITU-R BS. 1770-4.
  • True-peak level should not exceed -1 dB TP measured as per ITU-R BS. 1770-4.
  • Full-frequency content should not be present in the Low-Frequency Effects (LFE) channel of the BWF ADM.
  • For albums where gapless playback is intended between tracks:
    • Each album track must be delivered as an individual BWF ADM file.
    • Each track boundary must be no more than half a frame (1,000 samples @ 48 kHz) earlier or later than the same track boundary in the corresponding stereo deliverable.
    • There must be no additional silence at the end of each track when compared to the same track from the corresponding stereo deliverable.

Note: Immersive audio that does not meet the requirements may be removed from Apple Music.