Sonar users - how do you mix GPO?

[Christopher Duncan]

Hey, guys.

I use Sonar 3 as my sequencer with GPO inserted as VST instruments. In working on a brass ensemble, I wanted to do a little mixing between the parts as the tune progressed. What I ended up having to do was go into Geek View and insert cc1 events anytime I wanted to change the volume of an instrument. As you can imagine, this isn't the sort of smooth, fader driven mixing process that we're normally accustomed to with audio.

Unfortunately, Sonar appears to have jettisoned the StudioWare panel functionality allowing me to build a mixer of faders controlling cc1. Their Mixer view appears hardwired to cc7.

What I'm looking to accomplish is finding a way to mix using a normal fader based approach rather than inserting random cc events. How do you guys do mixdowns in Sonar? I'm sure I'm not the only person to have struggled with this, so I'd be grateful for any ideas you might have.

Thanks,

[squoze]

You can use the "bounce to tracks" (bouncing one or more GPO channels) and then mix the individual parts (audio tracks created by the bounce).
Or, you can select the "use STD cc#7 / cc#10 volume & pan" option in
Kontakt. This lets you create a sonar volume envelope for each GPO channel (midi track).

...If this is what you're talking about....

[Christopher Duncan]

Hey, Squoze.

Yeah, that's what I'm trying to do. I'd rather not have to handle an orchestra full of individual instrument audio files, so cc is the way to go. However, I haven't turned on the volume / pan option in Kontakt because I don't want to lose the articulations that come with controlling the volume via the mod cc, and that's the real dilemma...

[TobyBrez]

Hey, guys.

What I ended up having to do was go into Geek View and insert cc1 events anytime I wanted to change the volume of an instrument. As you can imagine, this isn't the sort of smooth, fader driven mixing process that we're normally accustomed to with audio.

Thanks,


Hey Christopher,

I use an Oxygen-8 midi controller and assign controllers to one of it's "knobs"
and record volume that way. Depending on your midi controller; you should be able to assign a knob or slider to send cc1 (or any controller) and record into SONAR.

Hope that helps...
Thanks,
Toby
http://www.soundclick.com/bands/4/tobybrezmusic.htm

[squoze]

You can turn on that option and still use cc1 for articulation.
This just lets you add an additional (strictly) volume adjustment to each track.
I do this, but I think its kinda cheating cause you can make, for instance, a oboe solo stand out over an entire orchestra, etc. It does make it easier to manage dynamics, though (for a beginner).

[TobyBrez]

You can turn on that option and still use cc1 for articuation.
This just lets you add an additional (strictly) volume adjustment to each track.
I do this, but I think its kinda cheating cause you can make, for instance, a oboe solo stand out over an entire orchestra, etc. It does make it easier to manage dynamics, though (for a beginner).

Hey Squoze,

Your avatar... Neil Young... On The Beech ! I had it on LP a loooong time ago!
Wow, great album. I looked for that on cd a while back and could not find it.
Do you know if it's still in print?

Thanks,
Toby
http://www.soundclick.com/bands/4/tobybrezmusic.htm

[squoze]

TobyBrez-
Yeah--it's been remastered and reissued! A few others were remastered:
Stars & Bars, Reactor, Hawks & Doves, Old Ways, a few others.
Great album, along with Zuma, my favorites. I actually liked "Landing on Water" (great snare sound).
Sorry for the OT reply but thanks for asking.

[TobyBrez]

TobyBrez-
Yeah--it's been remastered and reissued! A few others were remastered:
Stars & Bars, Reactor, Hawks & Doves, Old Ways, a few others.
Great album, along with Zuma, my favorites. I actually liked "Landing on Water" (great snare sound).
Sorry for the OT reply but thanks for asking.

Cool, thanks squoze...
I was hoping it would eventually be remastered!

Sorry Christopher for hijacking your thread!

Toby
http://www.soundclick.com/bands/4/tobybrezmusic.htm

[SteveMitchell]

Wow. I do all of my expression and volume changes manually. It involves a great deal of hearing parts over and over and over....:)


Stevemitchell

[billp]

Christopher,
SONAR "supports" StudioWare panels. You just can't create/modify them anymore. I have a simple panel on my site that has Mod sliders for 16 tracks. I don't use it much but you're welcome to it. It's HERE (http://www.broadviewnet.net/wylbur/GPOStudioWare.zip).

You load StudioWare panels from the "File | Open" menu. When you have your project open, do "File | Open" and select file type of "StudioWare". It'll open the panel.

[Haydn]

I prefer not to use CC7/CC10 as you get distortion if you set CC7 much above 60. The basic volume levels of the brass are pretty well balanced to begin with. I like to solo just the brass to get the brass levels balanced.

I usually start off mixing CC1 by doing the tuba part first. Then I bring in the bass trombone and then start adding the tenor trombones. You should have a pretty good balance of these instruments at this point. Then I add the trumpets and then the horns. With practice this doesn't take that long.

Once all the brass are balanced, then I can bring in the rest of the orchestra. I always keep the brass in the same instances of Kontakt (or the GPO player). This always me to be able to volume envelopes on the audio track for mixdown.

BTW, I also do the same mixdown on the woodwind and strings sections.

The total mixdown is then done by very the volume envelopes on the audio tracks which act as group tracks. Then I bounce the whole mix down to stereo. Sometimes I have to do extra bouncing if I run out of memory or CPU bandwidth.

[Skysaw]

Believe it or not, I do all my mixing with track envelopes. It's a lot of work, but gives you maximum control.

[beardedone]

Hey Billp! Thanks so much for posting the studioware panel. Very useful indeed.

[Christopher Duncan]

Hey, guys.

Lots of great ideas and help, as usual. Man, I love this neighborhood! :)

Here's the latest. After reading everyone's thoughts, I started wondering why I was having balance problems in the first place. Further listening pointed out that a couple of the ff instruments were really blaring (in particular the trombone - now I understand that qoute about don't look at the trombones, it only encourages them!). But why, given a balanced orchestra, was I having this problem?

The culprit turned out to be the various ff overlays, most notably horn and trombone. In Sonar, I had two tracks for each instrument, e.g. one trombone, one trombone ff overlay. Since they were linked parts, the cc1 setting on both were identical. It was probably a combination of the nature of the ff overlay and having two instruments playing the same thing, but the end result was that these parts blared above the others, hence the need for mixing.

That said, here's what I've done thus far. I moved cc7 down to 40 (I think) on all tracks to eliminate that as a source of distortion. I then unlinked the ff overlays and dropped their cc1 down to around 15 or 20 (in contrast with around 100 for the non overlaid instruments). This helped a great deal and has eliminated most of the distorted sound during the loud passages.

However, doing this was when I started complaining about the tedious nature of entering cc1 values in Geek View to do a mix. Toby's suggestion of using a hardware controller to mix the cc1 was a real "Doh!" moment for me as I have a Mackie D8B mixer which was easy to set up for that.

However, I'm very grateful to Bill for the studioware panel and will be trying that soon. The problem using the D8B is that there's a ton of controller events and the faders have difficulty keeping up. And when they don't, you're initial starting point for the instrument is off. I think the software version will probably be a more responsive solution.

By the way, Bill, I have Cakewalk Pro Audio 9 as well (and many previous versions), so it occurred to me after your post that I can design whatever I need there and simply import it into Sonar. I regret their decision to eliminate the studioware editing capabilities in Sonar, but this is a great workaround.

As always, you guys rock. Thanks!

[billp]

By the way, Bill, I have Cakewalk Pro Audio 9 as well (and many previous versions), so it occurred to me after your post that I can design whatever I need there and simply import it into Sonar. I regret their decision to eliminate the studioware editing capabilities in Sonar, but this is a great workaround.
Christopher,
Yep. Same here.

I think Cake has pushed ahead with audio and left the midi/composers behind a bit. Things like converting envelopes back to CC events, better notation interface, an easier to use controller pane (pain!), on-the-fly input quantization, slider mapping...

[Christopher Duncan]

Christopher,
Yep. Same here.

I think Cake has pushed ahead with audio and left the midi/composers behind a bit. I wish they'd strip all the audio out of one version and sell a product that's just 100% MIDI sequencer again, along with their other offerings. Cakewalk used to be rock solid, but it's never been entirely stable since they started adding audio to it, even after they renamed it to Sonar. Personally, I'd rather have a stable and world class sequencer than a semi-flaky "all in one" package.

[Skysaw]

I find the latest build of Sonar pretty rock solid. I definitely would keep using the MIDI + audio version if they split, but I would also like to see more MIDI editing shortcuts and niftities.

[tradivoro]

I have always first done all the midi work first, made sure that all velocities and relative balances are correct, and then record each individual track to audio... By doing that, I then can use envelope mixing to control all aspects of the mix... Not only that, I can add reverb, additional effects, eq, compression if necessary, and whatever can make the track sound better... I don't think you're going to get those options using just midi... I know that this sounds more laborious, but in the end you get a better product... You can even mix just using the mixer and using automation, if you don't want to use the envelope capability...

[Skysaw]

I only mix to audio when I have to (when I have so many tracks, the computer is choking).

In sonar, you can still put effects and envelopes on virtual synth tracks (like GPO).

[billp]

Christopher -- I'm sorry to hear you're having stability problems w/ SONAR. Like Jamie (Sky), I'm having no problems, at least with 4.01. There were some "issues" with the original 4.0.

FWIW, here's how I use SONAR:

I am primarily a midi composer and realizer, so I have 3 cascading tiers of tracks:

a) midi -- here, I try to create the "final" volume/timbre envelope and instrument dynamics (velocity for example), and panning, that is, a single place where the position and dynamics of each voice is defined. I find it unwieldy to apply dynamics here and then try also to automate a mix further down the chain. Otherwise, you end up going back an forth between the automated mix levels and the midi track dynamics in search of balanced levels.

(If I am using any audio tracks directly, as I do for sound effects and ambient audio in videos, then I do automate the audio track.)

b) soft synth audio outputs -- At least one track for each section, sometimes also separate out for each of the solo instruments. This is where I set the overall levels relative to the mix. And establish mixing groups. For example, I group the individual tracks of the Continuo section (harpsi, basses, cellos) for a baroque piece. This allows me to move the fader on any of the continuo instruments, and the level of the entire group changes. These are grouped relatively, so that they maintain their levels relative to the other members of the group. The primary outputs are routed to the "Master" bus.

This is also where I apply a small amount of eq if necessary to sonically isolate instruments that might be colliding sonically.

This is also where I route the signal to a reverb bus and set the send level to the reverb. This is a secondary output path.

On rare occasion, I use a mute envelope here for complex reasons that I won't describe here.

c) reverb submixes and a master mix -- The reverb(s) submixes are routed to the master output. This technique allows me to blend the reverbed sound with the dry that has been sent directly to the master from the audio tracks. The reverbs themselves are wet signal only, no dry. This way, I can do all of my mixing and blending from the SONAR sliders and do not have to open up the reverbs to blend wet and dry.

And of course the master mix is just that. One bus to control the overall level and to (temporarily) apply and experiment with any mastering effects.

I use Peak + RMS meters on all tracks and busses and never exceed a -6db peak on anything. I record at 24-bit/44.1.

[Christopher Duncan]

Cool stuff, Bill.

By the way, I was looking at the code of some of your CAL routines on your web site. I've never gotten around to writing any routines because I allowed myself to be put off by the Lisp like syntax. However, your code is quite readable (I just mentally substituded {} for () and was back in comfortable territory). Now I'm thinking I should get fuzzy with the language...

[Haydn]

I've found that Sonar 3.1.1 has been quite stable for me. The only issue I have is if I hit Stop and start editing while reverb trails are still going. This is using the Lexicon Pantheon reverb included with Sonar Pro. I usually disable effects while recording parts and this has fixed this issue.

Without audio, you would not be able to do VST instruments. So you need the audio. I've done pieces over 20 minutes in length and with about 120 tracks without problems. Editing starts getting a little slower towards the end of the pieces but it's nothing that slows me down.

[SteveMitchell]

I've found that Sonar 3.1.1 has been quite stable for me. The only issue I have is if I hit Stop and start editing while reverb trails are still going. I'm running 2.2, and there a checkbox under AUDIO somewhere that instructs the driver to allow decays and such.

SteveMitchell

[tradivoro]

I only mix to audio when I have to (when I have so many tracks, the computer is choking).

In sonar, you can still put effects and envelopes on virtual synth tracks (like GPO).

Well, yes, but you'd have to have a different instance of the kontakt player per midi track... At how many tracks do you start to notice drop outs and stuff??

[Skysaw]

Well, yes, but you'd have to have a different instance of the kontakt player per midi track... At how many tracks do you start to notice drop outs and stuff??

I run with a GB of memory on my PC, and can get quite a lot going on at once, without dropouts. I generally have one instance for winds, one for brass, one for piano/harp/percussion, and 2 for strings, though I sometimes break that up further. I also often run with several instances of Reaktor and other virtual instruments and/or audio tracks, and with several to many effects all at once.

Once I'm done with live recording, the latency slider goes way up, and the dropouts are usually not a problem.

[tradivoro]

I run with a GB of memory on my PC, and can get quite a lot going on at once, without dropouts. I generally have one instance for winds, one for brass, one for piano/harp/percussion, and 2 for strings, though I sometimes break that up further. I also often run with several instances of Reaktor and other virtual instruments and/or audio tracks, and with several to many effects all at once.

Once I'm done with live recording, the latency slider goes way up, and the dropouts are usually not a problem.

I see, you treat sections as one kontakt player... Well, that's definitely a good way to go about it... I guess I'm so conditioned over the years of recording the midi to audio, that's I didn't even think of that... Something to consider.. :)

[StrangeCat]

Sonar 3.1.1 Producer) I just started to use GPO for production and I do the same thing by having once instance of it going for a group ensemble that way I can have different busses before I even record the audio. Generally I will have Kontakt open as well and have samples from that. like i said I am new to GPO. I could start getting drop outs if I had efx on every track as audio, just the midi tracks I haven't ran into any drop outs that way. Doing orchestraion I am just importing a midi file from Sibelius 3 that I have as a score. I haven't quite got the hang of the whole mod wheel for velocity, I mean what a pain in the ~~~...each instrument individual in the whole ensemble? I working on it though ;) oh and i use GPO as a dxi not vst in sonar.

[billp]

Well, yes, but you'd have to have a different instance of the kontakt player per midi track... At how many tracks do you start to notice drop outs and stuff??
Tradivoro,
If you're referring to the application of effects to the synth's audio track(s), remember that each Kontakt instance can drive up to 8 separate stereo output tracks, each pair corresponding to a separate stereo audio track in SONAR. In effect, you can load 8 voices in the player, drive each voice with a separate midi track, and send each one of the player instruments to its own separate output audio track--8 in, 8 out. So you can apply different effects, envelopes, etc to each slot in a player instance.

[tradivoro]

Tradivoro,
If you're referring to the application of effects to the synth's audio track(s), remember that each Kontakt instance can drive up to 8 separate stereo output tracks, each pair corresponding to a separate stereo audio track in SONAR. In effect, you can load 8 voices in the player, drive each voice with a separate midi track, and send each one of the player instruments to its own separate output audio track--8 in, 8 out. So you can apply different effects, envelopes, etc to each slot in a player instance.

Provided you have a soundcard with that many outs... But I can see applying a kontakt player for each section, even though when you start applying eq, that's going to affect everything coming out of the contact player, all parts included, if you just have a soundcard with 4 outs, 2 stereo outs... However, applying a kontakt player to each section is a technique that can come in handy in a lot of instances...

[billp]

Provided you have a soundcard with that many outs... But I can see applying a kontakt player for each section, even though when you start applying eq, that's going to affect everything coming out of the contact player, all parts included, if you just have a soundcard with 4 outs, 2 stereo outs... However, applying a kontakt player to each section is a technique that can come in handy in a lot of instances...
Tradivoro,
Just want to clear this up, and I'm certain someone will correct me if I'm wrong :D . This is not a soundcard issue, at least for SONAR. My soundcard has 4 stereo outputs (my old one had 1 stereo out!), but I can create an unlimited number of audio tracks in SONAR to which I can send the outputs of the players, up to 8 pairs from each player. I have at least one project with 13 synth audio tracks and several audio mixdown tracks. These tracks are sent to a submix master stereo bus, which is routed to the soundcard. That's where any soundcard limitations come into play.

Just don't want anyone to get the idea that the number of synth output tracks is some how limited to the number of physical audio outs on your soundcard.

[tradivoro]

Tradivoro,
Just want to clear this up, and I'm certain someone will correct me if I'm wrong :D . This is not a soundcard issue, at least for SONAR. My soundcard has 4 stereo outputs (my old one had 1 stereo out!), but I can create an unlimited number of audio tracks in SONAR to which I can send the outputs of the players, up to 8 pairs from each player. I have at least one project with 13 synth audio tracks and several audio mixdown tracks. These tracks are sent to a submix master stereo bus, which is routed to the soundcard. That's where any soundcard limitations come into play.

Just don't want anyone to get the idea that the number of synth output tracks is some how limited to the number of physical audio outs on your soundcard.

Well, the problem is, I still don't have GPO and this is why I can't sit there and experiment... I'm thinking ok, you have this one kontakt player and you have 8 instruments, you select a midi track, select the relevant kontakt player, until you max out the 8 intruments on that kontakt player, midi wise... So, I'm thinking that one kontakt player is using a stereo audio output and that audio output is being shared by 8 different instruments... If that's not the case, that's great...

[Joseph Burrell]

That is not the case. Each instrument in each Kontakt player can be routed to a seperate audio output so that each can be mixed seperately.

[tradivoro]

That is not the case. Each instrument in each Kontakt player can be routed to a seperate audio output so that each can be mixed seperately.

But my question is, yes, I can understand each instrument can be routed to Kontakt player... Here's where I get confused... You have a kontakt player... Let's say you assign your woodwinds to this... So now you load 2 flutes, 2 clarinets, 1 oboe, 1 english horn, 1 basson, 1 contrabasson... All coming out of out 1 kontakt player... You have your midi tracks in Sonar for each instrument in that kontakt player, and you've assigned a different midi channel to each instrument in the kontakt... Now, with midi you can raise and lower the volume of each of these and pan them...

But now, let's say you want you want to eq the flutes different from the basson... Sonar is seeing the same Kontakt player for the flutes as for the basson... Any eq I apply to the that kontakt player is going to affect all the instruments equally, it's not going to differentiate the flutes from the basson... Am I right or am I wrong??

[billp]

But my question is, yes, I can understand each instrument can be routed to Kontakt player... Here's where I get confused... You have a kontakt player... Let's say you assign your woodwinds to this... So now you load 2 flutes, 2 clarinets, 1 oboe, 1 english horn, 1 basson, 1 contrabasson... All coming out of out 1 kontakt player... You have your midi tracks in Sonar for each instrument in that kontakt player, and you've assigned a different midi channel to each instrument in the kontakt... Now, with midi you can raise and lower the volume of each of these and pan them...

But now, let's say you want you want to eq the flutes different from the basson... Sonar is seeing the same Kontakt player for the flutes as for the basson... Any eq I apply to the that kontakt player is going to affect all the instruments equally, it's not going to differentiate the flutes from the basson... Am I right or am I wrong??
Tradivoro,
I see where the confusion is. In the player itself, you can select different OUTPUT channels. In the instrument configuration window in the player--where the name of the selected instrument is displayed--in the lower right-hand corner is a little "pitchfork" looking icon to the the left of a couple of numbers--"1/2", for example. This is a drop-down list of output channels. This is where you set the output channel of each instrument. They can all be the same or not, as you choose.

In your sequencer--SONAR for example, I'm sure CUBASE also--you can set up separate synth audio tracks for each one of these output channels:

Garritan Personal Orchestra VST 1 -- 1/2
Garritan Personal Orchestra VST 1 -- 3/4
Garritan Personal Orchestra VST 2 -- 1/2
Garritan Personal Orchestra VST 2 -- 3/4
Garritan Personal Orchestra VST 2 -- 5/6

etc.

Normally, when you add a synth to the synth rack in SONAR, it automatically creates 1 default audio track for the synth, which is monitoring tracks 1/2. Clone this audio track, and use the drop-down list on the cloned track's input selector to choose "3/4" (for example). Now you have two audio tracks "listening" to the single Kontakt player, but one is listening to the player's "1/2" outputs and one is listening to its "3/4". In this example, maybe you've routed your flutes to "1/2" and your bassoon to "3/4" in the player. Now you have separate audio tracks in SONAR for the flutes and the bassoon.

These tracks can then have their own EQ, volumes, etc.

[Skysaw]

I never thought about splitting them up this way. Great tip, Bill!

[tradivoro]

Thanks for that great explanation Bill... I still have to get the GPO, and this is why I'm asking these questions in the abstract... But I understand perfectly what you're saying and yes, it makes the kontakt player a very versatile instrument in that you don't have to record the sounds as I normally do with my other synths... Thanks once again...