Strategies for surround templates when working with sample libraries

[Olympum]

I have recently come to truly enjoy listening to my sample instruments in surround (both quad and binaural) while writing. I am routing the surround mics to the rear speakers, where the library allows it, as well as sending close mics to a surround reverb.

Personally, the only “sane” method I have found so far is using VEPro and returning stereo pairs for main (tree), room (surround), and all the close mics, one for each instrument or instrument group. This setup keeps the DAW sane enough to navigate it and reason about it.

But … although VEPro is fine, it has an overhead to setup, run, manage, maintain … and I would love to see if I there is a clean solution to do this without VEPro. I can see how a setup without VEPro would work for 10, 20, 50 … instruments tracks, but I fear this would become too complex to manage when every single instrument has 3 stereo outputs. I can easily see the template going beyond 1000 aux channels very, very, quickly.

How are others dealing with this? Is practically VEPro the only game in town when going up to surround for large orchestral projects?

Click to read more...

 

[Glen Brown]

What DAW are you using?

Surround makes all the sense in the world, especially with the ever growing previlence of dolby atmos.

I have several templates for downmixing 3rd order ambisonics to 7.1.2 (atmos), 7.1 and quad (all down mixed to binaural).

 

[JohannesR]

If you are listening in surround while writing, the disabled tracks template is the only method I can think of to keep the track count and complexity at a minimum (disabling all the tracks you don’t use).

If using VE Pro, why not do a decent quad mix for each instrument group inside VE Pro? Then apply reverb in your DAW (to get from Quad to 7.1/5.1). I can of course see the benefit of having easy access to all the separate mics inside your DAW, but the performance hit will be extreme for very little gains - at least in my book (I’m assuming you have a separate VE Pro computer). You can also remote control the faders in VE Pro from your DAW, so you do actually have control over every mic.

 

[JohannesR]

A workaround that doesn’t involve VEPro is to have your surround mix pre-routed, but only work with the Tree mics when writing. Then enabling the rest of the mics when bouncing.

It doesn’t quite solve your problem if the point is to write in surround, but nevertheless it can be a good strategy to have in mind once a project gets unmanageable; on that forte tutti cue, maybe you don’t need 6 mics for the harp when writing.

 

[Olympum]

What DAW are you using?

I am on Cubase. But since C13, at least on Apple silicon, I have noticed high CPU on idle when I have over 80-100 tracks enabled, even if they are not playing. As a result I have also started to investigate Logic, since it switch’s the audio engine off unless the track is playing.

If you are listening in surround while writing, the disabled tracks template is the only method I can think of to keep the track count and complexity at a minimum (disabling all the tracks you don’t use).

That’s what I am leaning towards right now, since the CPU overhead with VEPro of all the returns, even in a separate host, is rather visible. I am afraid though of going down this disabled-tracks road and that after hours re-building a disabled template with all the additional outputs I discover that it does not work well because it puts too much pressure on the CPU.

If using VE Pro, why not do a decent quad mix for each instrument group inside VE Pro? Then apply reverb in your DAW (to get from Quad to 7.1/5.1).

In VEPro I have each instrument output three stereo pairs: main, room and close. Then I group all the main, room and close mics, so that I have for example MAIN, ROOM, WW, HRN, TPT, … I am returning stereo pairs, not quad. I want to keep the close with as little room information as possible, so that when I finish writing and move to Atmos mixing I can use them as objects.

A workaround that doesn’t involve VEPro is to have your surround mix pre-routed, but only work with the Tree mics when writing. Then enabling the rest of the mics when bouncing.

That’s a really good hint. Once pre-routed, I would mean writing with only tree mics and put a surround reverb for writing to get some rear feedback. Obviously not the real deal, but maybe good enough to get a sense of space, without all the headaches.

BTW, and maybe I should clarified this at the start, the reason I am finding it super valuable, at least for me, to use surround while writing, is that I have hearing loss in my right ear which reduces my perception of position. I find that in stereo I struggle a lot with depth and positioning, whereas as soon as I move to surround my brain somehow compensates.

 

[muk]

@Guy Rowland is the right person to help decide between VE Pro vs disabled tracks in Cubase. I hope he will chime in.

I am switching to working in Atmos at the moment. My Atmos template is almost finished now. I am using VE Pro, with the same approach as you for the mics. For each library, I have three mic positions (close, room, ambient). The close mic gets a return for each individual instrument, while the room and ambient mics are summed per library. I am emulating a real recording session with this, where you get a close mic for every solo instrument, but the tree mics and ambient mics record all the instruments playing together, naturally. Mihkel Zilmer has a nice set of videos about that approach:

https://www.youtube.com/user/mihkelzilmer/videos

I am also using Mihkel's approach to audio tracks. Each vst instrument has a corresponding set of audio tracks laid out and correctly routed. These audio tracks are disabled in Cubase. With a single click I can enable the audio tracks I need, record the vst instruments to the audio tracks, and then unload and disable the vst tracks. That way I can work and mix in the same template. Everything is also routed to stems, so that I can deliver all the necessary stems to libraries right from my one template. No need to switch between composing templates and mix templates.

Finally, I have duplicated this routing now for my Atmos setup. I have a stereo out bus, and an Atmos out bus. the Atmos routing is necessarily different from the stereo, what with beds and objects, and various speakers and reverbs. Still I managed to create a routing that allows for Atmos stems too.

Both paths work simultaneously. I can compose in Atmos, but then mix in stereo if I want to and disable the Atmos branch when composing is finished. Or the other way around. Or stay in Atmos for everything.

It's complex to work out the routing scheme at first, and an unholy amount to set up the template. But once it's done, it's amazing to work with it. I can do everything I want in just one template. Everything is laid out correctly for the finished delivery right from the start, with all stems in stereo and atmos as needed. Writing in Atmos is also an amazing experience. When I switch back to stereo, everything sounds flat.

 

[Olympum]

@Guy Rowland is the right person to help decide between VE Pro vs disabled tracks in Cubase. I hope he will chime in.

Ahhh ... yes, I have watched Guy's videos, and I'd love to work with VEPro and CC automation to switch the instances. I just can't get over the stress of setting up the automation in VEPro for this system to work well. It feels so tedious upfront. So right now I keep everything on chewing CPU and RAM, which is a limiting factor. VSL should really make this part of the core product or at least give us an empty instance with 48 MIDI ports and 16 channels each with the automation already defined that we could import.

I am emulating a real recording session with this, where you get a close mic for every solo instrument, but the tree mics and ambient mics record all the instruments playing together, naturally.

That's exactly the same reasoning I applied when setting up the VEPro template, emulating a real recording session.

I am also using Mihkel's approach to audio tracks. Each vst instrument has a corresponding set of audio tracks laid out and correctly routed. These audio tracks are disabled in Cubase. With a single click I can enable the audio tracks I need, record the vst instruments to the audio tracks, and then unload and disable the vst tracks. That way I can work and mix in the same template.

I have seen folks in Logic do this, but this is a great idea, it does make a lot of sense to do this in Cubase to have a single template for writing and mixing. I will copy this idea.

Both paths work simultaneously. I can compose in Atmos, but then mix in stereo if I want to and disable the Atmos branch when composing is finished. Or the other way around. Or stay in Atmos for everything.

I am very much aiming for the same, so that I can produce downmixes (stereo, but also binaural for uploading to YouTube since most folks there use headphones). In my case, I am working in quad for writing, without Atmos, and I plan to move to Atmos only in the mixing stage.

I tried the opposite route of working in stereo and then upmixing, and it just does not work, it feels weak and fake, which is why I prefer to start in surround. And once you start writing in surround, opening an older project in stereo will leave you disappointed.

 

[muk]

I have seen folks in Logic do this, but this is a great idea, it does make a lot of sense to do this in Cubase to have a single template for writing and mixing. I will copy this idea.

I can recommend it. The advantages of rendering in place are that the naming and routing is already in place. The downside is that you record it in real-time, obviously. And you have to remember to record all the instruments of a library together, because we summed the room and ambient mics. If you only record one instrument, then the room and ambient mics will also only contain the sound of this one instrument.

I am very much aiming for the same, so that I can produce downmixes (stereo, but also binaural for uploading to YouTube since most folks there use headphones). In my case, I am working in quad for writing, without Atmos, and I plan to move to Atmos only in the mixing stage.

That's entirely possible as well. In my template I have rough premixes in place for both stereo and Atmos. That way I will not be downmixing an Atmos mix to stereo, but have a different routing paths for both. The audio renderings are routed to both paths, stereo and Atmos. Any mixing decisions that I want to apply to both formats happen at this audio track stage. EQ, exciters, tonal shaping basically. Things like compression, and reverb, will often be different for stereo and
Atmos. So these I add downstream in either the stereo or the Atmos routing paths.

If it is only for a quick demo I can also use the Dolby downmix from Atmos to stereo. For publication files I want more control than that. That's why having a separate stereo path in place really helps. I can work in Atmos, and then create the stereo mix as I like with all the routing already correct, the mics folded to the right channels, etc.

And once you start writing in surround, opening an older project in stereo will leave you disappointed.

That's my experience as well. Working in 7.1.4 is more engaging and satisfying for me than working in stereo.

 

[Symphonichrome]

Each vst instrument has a corresponding set of audio tracks laid out and correctly routed.

Much of this thread is above my proverbial pay grade but am hoping I am correct in my interpretation of the statement quoted herein, being:

As an example, say I click on a VSL Synchron Violins patch that uses their surround mix...
You would compose with all mics in the patch turned off except for the close mic
Then, once you have it the way you like, bounce the close mic to it's own audio channel

Then, each additional non-close mic used by each instrument in the project are each summed to one channel and that channel is bounced to audio
Repeat until each individual microphone (non-close) has it's own bounced audio as an entire orchestra

Fingers crossed I am even half-right about this

Thank you for helping me learn

 

[Olympum]

Much of this thread is above my proverbial pay grade but am hoping I am correct in my interpretation of the statement quoted herein, being:

As an example, say I click on a VSL Synchron Violins patch that uses their surround mix...
You would compose with all mics in the patch turned off except for the close mic
Then, once you have it the way you like, bounce the close mic to it's own audio channel

Then, each additional non-close mic used by each instrument in the project are each summed to one channel and that channel is bounced to audio
Repeat until each individual microphone (non-close) has it's own bounced audio as an entire orchestra

Fingers crossed I am even half-right about this

Thank you for helping me learn

I prefer to compose in surround, but it's rather taxing resource-wise (specially the surround panners). The more traditional route would indeed be to write using the tree mics, and when mixing print the tree, close and surround for each instruments into audio tracks.

With orchestral music, I see orchestration as part of the mixing process, and I find composing in surround a better strategy to get the sound I am looking for. I also find writing in surround to be more inspiring, and perhaps ... musical?

 

[Aymara]

I prefer to compose in surround, ...

Because you also distribute to Apple Music, why not compose in Atmos and then downmix to Stereo to be able to deliver both. From the distribution point of view everything else doesn't make much sense to me.

At least that's what I do now.

Tip: LUFS check is done in 5.1 as recommended by Dolby.

 

[Olympum]

Because you also distribute to Apple Music, why not compose in Atmos and then downmix to Stereo to be able to deliver both. From the distribution point of view everything else doesn't make much sense to me.

At least that's what I do now.

Tip: LUFS check is done in 5.1 as recommended by Dolby.

I do mix in Atmos, but I write in quad. I find it a good compromise especially in regard to computer resources.

 

[muk]

As an example, say I click on a VSL Synchron Violins patch that uses their surround mix...
You would compose with all mics in the patch turned off except for the close mic
Then, once you have it the way you like, bounce the close mic to it's own audio channel

Then, each additional non-close mic used by each instrument in the project are each summed to one channel and that channel is bounced to audio
Repeat until each individual microphone (non-close) has it's own bounced audio as an entire orchestra

Fingers crossed I am even half-right about this

Thank you for helping me learn

It's close, but not entirely correct. In your example, I'd have VSL Synchron Violins patch loaded. I have The close mic routed to a track called VSL Sychron Strings Violins 1. The tree mic, mid mic, etc. are routed to a track called VSL Synchron Strings Room. And finally, the surround mics, ambient mics, gallery mics etc (all the ambient mics), are routed to the track VSL Synchron Strings Ambient.

In this example, I have three tracks: Synchron Strings Violins 1 (the close mic), Synchron Strings Room, and Synchron Strings Ambient.

I create the mockup with all these mics active. Meaning I create the mockup in Dolby Atmos. When I am finished with a part, I can then render the the track to audio.

This video explains the setup that I have (except that I work in Dolby Atmos):

 

[Symphonichrome]

I prefer to compose in surround, but it's rather taxing resource-wise (specially the surround panners). The more traditional route would indeed be to write using the tree mics, and when mixing print the tree, close and surround for each instruments into audio tracks.

With orchestral music, I see orchestration as part of the mixing process, and I find composing in surround a better strategy to get the sound I am looking for. I also find writing in surround to be more inspiring, and perhaps ... musical?

Ahhhh ok - I only have a traditional stereo monitor setup but I know that Apple Music loves receiving Atmos mixes

It's close, but not entirely correct. In your example, I'd have VSL Synchron Violins patch loaded. I have The close mic routed to a track called VSL Sychron Strings Violins 1. The tree mic, mid mic, etc. are routed to a track called VSL Synchron Strings Room. And finally, the surround mics, ambient mics, gallery mics etc (all the ambient mics), are routed to the track VSL Synchron Strings Ambient.

In this example, I have three tracks: Synchron Strings Violins 1 (the close mic), Synchron Strings Room, and Synchron Strings Ambient.

I create the mockup with all these mics active. Meaning I create the mockup in Dolby Atmos. When I am finished with a part, I can then render the the track to audio.

This video explains the setup that I have (except that I work in Dolby Atmos):

Thank you for that clarification - I will definitely check out that video

 

[Symphonichrome]

It's close, but not entirely correct. In your example, I'd have VSL Synchron Violins patch loaded. I have The close mic routed to a track called VSL Sychron Strings Violins 1. The tree mic, mid mic, etc. are routed to a track called VSL Synchron Strings Room. And finally, the surround mics, ambient mics, gallery mics etc (all the ambient mics), are routed to the track VSL Synchron Strings Ambient.

In this example, I have three tracks: Synchron Strings Violins 1 (the close mic), Synchron Strings Room, and Synchron Strings Ambient.

OK I watched the video and it was excellent

However, I am still a tad confused (shocker, I know lol)

In the example you posted re: Synchron Violins...are you stating that I would have 3 instances of Synchron Strings Violins loaded in my template?

1. Close (all other mics in the patch turned off);
2. Room (all other mics in the patch turned off);
3. Ambient (all other mics in the patch turned off).

I thought there might be a way to route those separate mics to their own channels out of the Synchron Player into Logic but, I have not found a way yet

Ultimately, would that mean that, for each instrument, I would need 3 instances of the instrument patch loaded in my DAW?

Thank you for your advice (and patience with my newbie questions)

 

[muk]

No problem, I am happy to help.

In the example you posted re: Synchron Violins...are you stating that I would have 3 instances of Synchron Strings Violins loaded in my template?

1. Close (all other mics in the patch turned off);
2. Room (all other mics in the patch turned off);
3. Ambient (all other mics in the patch turned off).

That's not necessary. You can route each microphone channel to a separate output. To do so, in Synchron Player go to the mix tab. Immediately above each channel name you can set the output channel that you want to use (1 (Stereo), 2 (Stereo) etc.). So it's only one instance per instrument, but three output channels. I'm on Cubase 13, and I use VE Pro to host the instrument players. I can't help you with Logic. But it shouldn't be too difficult to configure so that you get the necessary input channels from the Synchron Player.

 

[Symphonichrome]

No problem, I am happy to help.



That's not necessary. You can route each microphone channel to a separate output. To do so, in Synchron Player go to the mix tab. Immediately above each channel name you can set the output channel that you want to use (1 (Stereo), 2 (Stereo) etc.). So it's only one instance per instrument, but three output channels. I'm on Cubase 13, and I use VE Pro to host the instrument players. I can't help you with Logic. But it shouldn't be too difficult to configure so that you get the necessary input channels from the Synchron Player.

Thank you again

I have been changing the 1(Stereo), 2 (Stereo), etc. in the Synchron player but, cannot yet figure out how to have Logic recognize them

I will have to do a deep dive (I will report my findings back here)

 

[Symphonichrome]

UPDATE:

I figured it out (for the most part - very close to the proverbial finish line)

I need to use Synchron as a multi-output in Logic

I then load up, for example, Synchron Duality Strings which provides the following microphone positions (I have the Full/Extended library):

• Room Mix – Stereo
• Solo – Mono
• Mid– Stereo (L/R)
• Main/Room Mic – Decca Tree Stereo (L/R)
• Main C/Room Mic – Decca Tree Mono (Center)
• “Dark” microphones – Stereo (L/R)
• Ribbon – Stereo (L/R)
• Main Surround – Stereo (L/R)
• High Stereo (3D) – Stereo (L/R)
• High Surround (3D) – Stereo (L/R)
• “Bright” microphones – Stereo (L/R)

So, if I have this right, I would turn off the "Room Mix" microphone and then make a separate output for each of the other microphones or, would I combine some of those microphones on certain outputs?

I feel like my eyes have finally opened to the possibilities

I am beyond thankful for your guidance

 

[muk]

It depends on what you want. You can create a separate output for each mic position if you want that much granular control. I don't need it, and it would slow me down and use more tracks than necessary. So I sum the microphones of every library into three 'positions': close, room, and ambient. The close microphones are on individual tracks for every instrument. So I'd have Duality Strings Vl 1 (close mic), Duality Strings Vl 2 (close mic). Room and Ambient mics are summed. So the room mics of Vl 1, Vl 2 and the other sections all go to the same track 'Duality Strings Room'.

This is the layout used in typical recording sessions. If you were to record an orchestra live, you would have individual spot mics (close mics) on the soloists. But the room mics and ambient mics necessarily pick up the whole orchestra at once, not only individual instruments/sections.

How exactly you group the microphones is up to you. The close mics are clear. The decca tree is usually in middle distance, so it goes to 'Room'. The surround mics, gallery mics and similar are further away, so they go to 'Ambient'. Others might be fringe cases. Up to you to decide. I go by my ears. What sounds far away and ambient goes to ambient. Mics that are a bit closer, but not spot mics, go to room.

Also you don't have to use all the microphones in a library simply because they are there! I usually try q opt of options until I settle on a mix that works as my go-to starting point in my template. This might, for example, not include the mid mics, or the dark mics. Or it could just be the close mics, decca tree (Room), and surround mics (Ambient). Whatever works for you.

 

[Symphonichrome]

It depends on what you want. You can create a separate output for each mic position if you want that much granular control. I don't need it, and it would slow me down and use more tracks than necessary. So I sum the microphones of every library into three 'positions': close, room, and ambient. The close microphones are on individual tracks for every instrument. So I'd have Duality Strings Vl 1 (close mic), Duality Strings Vl 2 (close mic). Room and Ambient mics are summed. So the room mics of Vl 1, Vl 2 and the other sections all go to the same track 'Duality Strings Room'.

This is the layout used in typical recording sessions. If you were to record an orchestra live, you would have individual spot mics (close mics) on the soloists. But the room mics and ambient mics necessarily pick up the whole orchestra at once, not only individual instruments/sections.

How exactly you group the microphones is up to you. The close mics are clear. The decca tree is usually in middle distance, so it goes to 'Room'. The surround mics, gallery mics and similar are further away, so they go to 'Ambient'. Others might be fringe cases. Up to you to decide. I go by my ears. What sounds far away and ambient goes to ambient. Mics that are a bit closer, but not spot mics, go to room.

Also you don't have to use all the microphones in a library simply because they are there! I usually try q opt of options until I settle on a mix that works as my go-to starting point in my template. This might, for example, not include the mid mics, or the dark mics. Or it could just be the close mics, decca tree (Room), and surround mics (Ambient). Whatever works for you.

Very true - it will be up to how much granular control I want

I thought the "Room Mix" mic channel in Synchron summed the Deca Tree mics, that's why I thought I would mute the "Room Mix" channel, is my interpretation incorrect?

Finally, is using an Atmos setup required for outputting the stems to send to a mix engineer or could I do it all in stereo outputs for each mic and then the mix engineer could expand to Atmos?

This thread has really opened my eyes to what I can do - once I figure out the answers to the two questions herein I can start making my new template WOOHOO!!!

Thank you again