My quest for Quadraphony -- journal and audio samples

[Ben Asaro]

Back in November, I posted a thread with some questions about quadraphonic sound.

That started my very deep dive into into the world of immersive, multichannel audio. During the intervening months, I did as much research as I could and I placed an order through Perfect Circuit with some modules with the idea of doing quadraphonic performances and mixes.

I suspected this was going to be a very long-term project and it looks like that suspicion was correct.

Now that Jamuary is over, I am back at it. The first step in all of this is getting a grip on SIG+, but that was just the first kernal of the idea. I thought, oh, 4 voices, 4 channels -- done! Little did I know how much my perspective would change on the subject.

Once I had the modules in hand, I started doing my research online about immersive audio.

I should back track a bit and explain what modules I purchased. They are the Shakmat Aeolus Seeds surround panner, Shakmat Aeolus Seeds mixer, and Doepfer CV panner.

Of all of the web sites I found, the one I enjoyed reading the most was Wendy Carlos' awesome web series of articles, What Comes Around Goes Surround, https://www.wendycarlos.com/surround/surround.html. She does a fantastic job of breaking immersive audio going backwards, and then forwards again, through time, explaining the different methods and technology used to create immersive music formats.

Now, this may sound a little sacrilege, but I actually agree more with Carlos' philosophy and approach to immersive audio than Suzanne Cianni's. It was Carlos' basic model that really stuck out to me the most, and will be my personal starting point with this project.

Simply put, if you take the sound stage and mic it across with surround channels, you start off with something like this:



But upon playback, that sound stage now curves around the listener, like this:



I found this so much more helpful and instructive towards plotting out my own mixes over just arbitrarily moving things around as a gimmick; well, not purely as a gimmick lol -- I'm not opposed to some gimmickry.

[Ben Asaro]

Once I had the basic idea down, my next step was to make a test mix in Logic. THIS singlehandedly was the thing that a) convinced me that binaural/multichannel audio is probably going to be the way I make music going forward and b) was very instructive in finding the hidden pitfalls and expenses in doing immersive audio as an independent artist.

Here's what I learned from that little test mix:

1. Even though Logic will enable you to mix in surround, once you have that mix completed, there's not much you can do with it! You will need an additional piece of software from Dolby to take your multichannel mix (ADM wav) and re-encode it to a format that people can use (mp4). The kicker is that the software costs $300.

2. For an end user to hear the mix in surround, not only do they need to have the necessary speakers, but their source may not support the multichannel mix and will still only play it in stereo! For example, although my PS5 has Apple Music, and music supports immersive audio including Dolby Atmos and binaural, the app will only play in stereo. In order to get surround from Apple Music, I will need a dedicated Apple TV.

[Ben Asaro]

The first thing I am trying to sort out is how to get a source to move forward and backwards across the soundstage. Now, this is quite easy in Logic, you just make that track a 3D Object, and drag it across using automation. But that would be doing it the easy way, and we all know how I feel about that!

I want to be able to do it using hardware and built it into the actual performance. In my mind, three things should happen as a source moves further away:



1 - the high frequencies roll off
2 - the source diminishes in level
3 - the sound becomes more reverberant

This will be my first actual experiment in immersive audio. I started working on it today. My thinking is that if I took a stereo panner and turned it on its side, "left and right" can become "front and back". Now instead of moving a source across the X plane (left-right), I can move it along the Y plane (front-back).



The catch is going to be sorting out how to set up the modular so that a single source will do all of these things from a single point of control, ideally.

[Ben Asaro]

First test was a success!

Using the precision adder algo on disting, I was able to have a single knob simultaneously reduce the volume and roll off some of the high frequencies at the same time!



The patch is remarkably simple, though it took ages to get my brain to wrap around how to do it.

disting precision adder (set to divisor/12) > MULT
> low pass filter CV input w attenuator
> cv panner

cv panner output 1 > mixer
cv panner output 2 > mixer

low pass filter > MULT
> panner input 1
> panner input 2

At the bottom of its range, the precision adder is taking the audio from the filter and rolling off the highs; this signal path is going into the mixer at a set level. As you increase the voltage, the precision adder opens up the filter and the audio going from the corresponding output on the panner is going into another mixer channel at a higher level.

Thus, the sound gets louder/brighter and softer/darker depending on where I have the precision adder set.

[Ben Asaro]

Hmmm ... last thought for tonight: I wonder if I could do this more simply using a VCA instead of the panner? (Adds it to the to-do list...)

[Eddy Deegan]

This is outside of my expertise/experience. I'm stereo man myself but I applaud your aspirations and will be watching future developments with interest.

[The Elf]

I find reverb pre-delay critical for creating a sense of depth, in both my stereo and surround mixes.

[Arpangel]

I'm getting more into "multi-speaker presentations" be it surround or quadrophony, I just think they add something good.
We did some experiments a few years back, we bought eight Novation A-Stations, and fed them into custom made surround mixers, we had a 9.1 system, including height, we recorded the synths onto discreet tracks using an Alesis ADAT, the trouble is no one but us was able to hear our mixes! they were stuck in our studio! as no one had the gear to play them back apart from us, that's the catch with all this.
I still have my friends custom surround "matrix mixers" trouble is I haven’t worked out how to use them, they are so complex. He died and I never got a chance to ask him to show me how to use them properly, it's a shame, as they are very powerful tools, they just sit on my shelf as a memory of my friend, and his wonderful exploits.
Ben, you have a Disting, you are indeed a very brave man, I take my hat off, if that was here it’d end up a pile of dust ground into the floor as I'm taken away by the men in white coats.

[Ben Asaro]

I find reverb pre-delay critical for creating a sense of depth, in both my stereo and surround mixes.

Great advice! Thx!

[Ben Asaro]

Thx, Eddy!

If you’re working purely ITB, I think switching up to at least binaural sounds is a worthwhile experiment. In hardware is where things start to get chewier.

[Ben Asaro]

I'm getting more into "multi-speaker presentations" be it surround or quadrophony, I just think they add something good.
We did some experiments a few years back, we bought eight Novation A-Stations, and fed them into custom made surround mixers, we had a 9.1 system, including height, we recorded the synths onto discreet tracks using an Alesis ADAT, the trouble is no one but us was able to hear our mixes! they were stuck in our studio! as no one had the gear to play them back apart from us, that's the catch with all this.
I still have my friends custom surround "matrix mixers" trouble is I haven’t worked out how to use them, they are so complex. He died and I never got a chance to ask him to show me how to use them properly, it's a shame, as they are very powerful tools, they just sit on my shelf as a memory of my friend, and his wonderful exploits.
Ben, you have a Disting, you are indeed a very brave man, I take my hat off, if that was here it’d end up a pile of dust ground into the floor as I'm taken away by the men in white coats.

Lol I rarely use disting for that exact reason! The menu system is maddening, but 100 utilities in 4hp means I always seem to find a use for it.

Dolby has gone all-in on immersive audio and if you have the software, making surround mixes is easier than ever. Best part of Dolby Atmos is that it will downmix to whatever the end user has, so even in stereo, all of your elements are preserved. More speakers = more discrete panning, but you don’t have to make multiple mixes, per se.

[Arpangel]

Hmmm ... last thought for tonight: I wonder if I could do this more simply using a VCA instead of the panner? (Adds it to the to-do list...)

Thx, Eddy!

If you’re working purely ITB, I think switching up to at least binaural sounds is a worthwhile experiment. In hardware is where things start to get chewier.

Binaural only really works on headphones though, with varying reults.

[Martin Walker]

The first thing I am trying to sort out is how to get a source to move forward and backwards across the soundstage. Now, this is quite easy in Logic, you just make that track a 3D Object, and drag it across using automation. But that would be doing it the easy way, and we all know how I feel about that!

I want to be able to do it using hardware and built it into the actual performance. In my mind, three things should happen as a source moves further away:



1 - the high frequencies roll off
2 - the source diminishes in level
3 - the sound becomes more reverberant

As Eddy mentioned, this thread looks to be another of your must-read ones Ben.

You're off to a flying start, and I know you want to perform your forward/backward moves using hardware, but you might nevertheless find some useful thoughts from the design and discussion in the manual of this freeware Proximity plugin from well-respected Tokyo Dawn Labs:

https://www.tokyodawn.net/proximity/



Several models can be combined:

Distance signal delay by speed of sound
Distance gain loss
Absorption of high-frequencies in air
Stereo width manipulation
Proximity effect of virtual microphone
Distance based early reflections

Hope this helps!

Martin

[Ben Asaro]

Thanks, Martin! I will definitely give this a serious look and listen, even if for inspiration, it looks like a fantastic tool/resource! Creating a predelay is pretty simple in modular, believe it or not. I think a voltage controlled delay line can be added to the output of the precision adder… (Adds this to the to-do list)

[Ben Asaro]

Binaural only really works on headphones though, with varying reults.

No argument there, though I would still argue it’s a worthwhile experiment and most ppl are using some form of in-ear listening these days.

[Hugh Robjohns]

Best part of Dolby Atmos is that it will downmix to whatever the end user has, so even in stereo, all of your elements are preserved.

This is semi-true, and was the same for all previous Dolby multichannel formats, too.

However, there's no free lunch — meaning there's no guarantee that a wonderful surround or immersive mix will sound any good as a stereo downmix.

The biggest danger when mixing in surround is being seduced by the lovely spatial separation, which allows conflicting and overlapping sounds to coexist happily in their own space.

This makes mixing in surround so easy, because nothing masks anything, and if it does you just move it somewhere else so the problem goes away.

But a stereo (or even mono) downmix brings back all the familiar issues of frequency masking, comb-filtering, and so on... so if you're not checking your surround mix in stereo and mono regularly, and correcting all those downmix problems in the traditional ways, you dont actually have a valid dowmix at all.

In other words, just because the hardware can perform the function doesn't mean what it creates is worth listening too! There is no free lunch, and you still have to put all the work in, even though you won't hear that effort in the surround mix... or maybe you will because things needed to make the stereo work well might compromise the surround mix (just like optimising a summed mono mix can compromise the stereo mix).

[Arpangel]

Binaural only really works on headphones though, with varying reults.

No argument there, though I would still argue it’s a worthwhile experiment and most ppl are using some form of in-ear listening these days.

That's a good point, about headphones, it is weird, we made a few binaural recordings, and if you listen to them in the place where they were recorded, they are frighteningly realistic, theres something psychological going on, the brain is matching the sounds to the place, sounds great though.

[Ben Asaro]

Best part of Dolby Atmos is that it will downmix to whatever the end user has, so even in stereo, all of your elements are preserved.

This is semi-true, and was the same for all previous Dolby multichannel formats, too.

However, there's no free lunch — meaning there's no guarantee that a wonderful surround or immersive mix will sound any good as a stereo downmix.

The biggest danger when mixing in surround is being seduced by the lovely spatial separation, which allows conflicting and overlapping sounds to coexist happily in their own space.

This makes mixing in surround so easy, because nothing masks anything, and if it does you just move it somewhere else so the problem goes away.

But a stereo (or even mono) downmix brings back all the familiar issues of frequency masking, comb-filtering, and so on... so if you're not checking your surround mix in stereo and mono regularly, and correcting all those downmix problems in the traditional ways, you dont actually have a valid dowmix at all.

In other words, just because the hardware can perform the function doesn't mean what it creates is worth listening too! There is no free lunch, and you still have to put all the work in, even though you won't hear that effort in the surround mix... or maybe you will because things needed to make the stereo work well might compromise the surround mix (just like optimising a summed mono mix can compromise the stereo mix).

Thanks for the insights, Hugh!

One thing I do like about the Atmos monitoring app in Logic is that you can play back the mix in various downmixed statuses to check for issues.

The funny thing is -- the multichannel mix is not my primary goal, it's more of a convenient byproduct of my actual project, which is a piece written to be played on multiple playback devices simultaneously.

But that's a very, very long way off! I still have a mountain of education to get through first!

[Ben Asaro]

With regard to resources, I have found the following to be very helpful:

https://youtu.be/P6Wst_im_WQ?si=tza81pjkzgtE2Dhu

This video is great if you have Logic Pro. However, they neglect to inform you that the software required to downmix is an additional $300. >.<

https://immersiveaudioalbum.com/how-to- ... zFuINHLx10

This article was very helpful to me in sorting out file formats and how to get test mixes from my DAW to my hifi.

[Arpangel]

which is a piece written to be played on multiple playback devices simultaneously.

That’s sounds great, and it’s when things start to get "really" interesting, you can get a piece that is constantly changing, and never repeats itself, the thing to try and overcome in pieces like this is a good variation in content, sounds, so that it doesn’t all seem to be in the same textural area.

[Ben Asaro]

That’s sounds great, and it’s when things start to get "really" interesting, you can get a piece that is constantly changing, and never repeats itself, the thing to try and overcome in pieces like this is a good variation in content, sounds, so that it doesn’t all seem to be in the same textural area.

Totally agree! But that's a LONG way off at the moment. I've not even got started yet, let alone thinking about the ending!

[Ben Asaro]

Another one of those incredibly 'fun' skiff reorgs this morning, keeping in mind what Martin and Elf suggested --



I was able to create some space in the big case to add in the Eduemonia and Nostalgia to the skiff. The Eudemonia will be mixing the audio from the panner, instead of the Listen Four. I moved the AA.1 into the skiff as well since that now has all of the main output modules in it; that just makes sense from a workflow perspective to my mind.

And finally, the Nostalgia will be my predelay for the 'near/front' output of the panner.

So the signal path will be:

NEAR/FRONT
Low pass filter> mixer > predelay > reverb

FAR/BACK
Low pass filter > mixer > reverb

[Ben Asaro]

Small update: I couldn't figure out why I was getting crosstalk from the panner. Turns out that +/- 12 semitones isn't enough voltage to get the panner to fully go fully left/right. I fixed that by changing the precision adder to +/- 10V, but then it was way too much voltage for the low pass filter.

I ended up pulling the mmMidi from the big case and replacing it with my trust Doepfer attenuator, and now the filter is smooth as butter AND I have full control over the panner! Happy days!

Whew! Now to sort out the Nostalgia; it may not be able to do what I want, which is a shame, but I'm digging through the 'manual' (it's more like a tear sheet) now.

I also had to swap all of the modules out of the Make Noise skiff and put them into my other MN skiff, which is powered, because I needed that extra 4hp from the uZeus to squeeze in the AA.1.

Oh, the joys of modular!!

[Ben Asaro]

I spent an entire day working out how to do this via modular, and in the end, it works! Using a precision adder, mixer, and line delay, I can get a sound that gets brighter/louder/more present (increased predelay to the reverb) to one that's darker/quieter/more reverberant.

HOWEVER -- the cost in modules is high; high enough to make me question whether the juice is worth the squeeze vs lowering the source level and adjusting the filter cutoff manually. For a single voice, this technique requires a precision adder, VC panner, attenuator, two mults, and a mixer with two inputs and discrete level control for each.

If I was going to pursue this in earnest, I would need another skiff with four of each module.

I don't know if I will end up there, to be honest, as I haven't even started writing the music yet, so I don't know how important it will be to the final outcome.

However, I don't feel that the time was wasted in the least, and now I have a record of it should I decide later that it's something I want to pursue.

[Ben Asaro]

This is for Future Me, in case I decide to go down this road again. I only have two low pass filters in my case, so this is what I would need to have full Y-plane control over those two voices.



The Good News is that I can also build this up over time as I am not in any particular hurry for this project.