VSTs at different levels out of the box

[CDK]

Hi all,

I wanted to ask a question about mixing VST levels within the one project.

Some of my VSTs are at very different volume levels ‘out of the box’, (e.g. Mojo Horns, Session Horns, Broadway Lites all seem to be quite soft compared to some others), meaning I have to raise their volume to match other VSTs in the mix. I obviously want to avoid audio clipping but it’s sometimes hard to know what the end result is going to be like throughout the mixing process.

Does you have any advice/tutorials for mixing/balancing out a range of different VSTs? I hope this makes sense!

Thanks :)

[Weekend Astronaut]

As long as the levels in your VST instrument's internal meters and in the associated audio track's output meters are not clipping then you should be fine if you don't care to go into a discussion about intersample peaking. If you're not clipping at the audio output track (and anywhere down the chain from there) then you're fine. It's no different than just balancing levels with normal audio files. If they're not already clipped, you just manage the gain along the chain.

If you mix your VST instruments so they sound right then all you need to do is make sure their output levels aren't clipping anywhere just like anything else. That's just you mixing.

The VST caveat regarding gain-staging is more for effects which take an audio input. Different VST effects behave differently and many digital effects can sound bad if you overload their input going in (or cumulatively through a series), but I'm not sure that's related to your question.

[chuckebaby]

Its all about gain staging.

What you want is each individual signal path to be respectfully within its range (db wise). This means following/analyzing/adjusting  the path from input to output. When your at this point in the mixing stage don't worry too much about cohesiveness (one track being louder than the other). Be more concerned with gain staging. Don't red line any of your track levels. I've had projects with 30+tracks that require some tracks to be set at -24db to avoid blowing up my master bus.

Once proper gain staging is accomplished, then you can move on to adjusting track faders by both using visual inspection and more importantly...Ear inspection. Some tracks may read 0db and another -10 db and sound the same room level. This is where EQ and a good ear will help.

[CDK]

24 minutes ago, pwalpwal said:

this seems more like "just mixing"

Yep I know.. haha

[CDK]

Thanks for your replies.

A general question... I usually do my mixing all in the MIDI stage/initial project stage; however should I be bouncing all my tracks to audio then mixing/mastering them in a separate project? I can never seem to get my tracks sounding as punchy as commercial tracks do and I wonder if it's because I'm staying in MIDI the whole way.

[CDK]

2 minutes ago, pwalpwal said:

I think you should be mixing the audio out of your midi tracks, so don't mix the levels via the midi track volume setting but rather via the audio output from those instruments

And if I'm using a multi-timbral instrument like Kontakt?

[David Baay]

If a VST's default output level is consistently lower or higher than you want with audio Gain and Volume controls at 0dB, start by adjusting the MIDI Volume fader. This controls the synth's Master Volume with CC7 (or channel volume of a multitimbral synth if the track has an output CHannel assigned).

By default, the MIDI Volume fader is disabled as indicated by parentheses around the default level of 101 (see the MIDI tab at the bottom of the Track Inspector in the case of a  Simple Instrument track).  When you move it, it automatically becomes enabled, and that value will be sent to the synth to initialize its output level every time you start playback. This is especially important with hardware synths that may have their volumes left at some other level by a previous project or alteration from the front panel. Softsynth states are stored in the project so that won't happen, but it's still good practice to set that initial volume to have a reliable starting point. Conversely, if a MIDI volume control was inadvertently enabled at some point, you can double-click it to reset the value to 101, and then right click and choose 'Disable Control'. Then you can control the synth's volume from its GUI. 

Also, be aware that if two or more tracks point at the same synth/channel, and both have their MIDI Volume faders enabled, the highest numbered track will have the final say in what that level is, as Cakewalk process the track volume messages (and other MIDI controller messages like Pan) from top to bottom.

[Robert Bone]

Some VST instruments really peg their preset volumes way louder than they should be - there is no standard, so from one instrument to the next initial plugin instrument volume can be all over the place.  Lots of plugin authors like their presets to sound 'punchy/loud' and they are set to initial values that are way too loud.

Whenever I load a plugin instrument or live instrument/mic, I always set the combo of the plugin output volume and the assigned/associated audio track(s) gain setting to a peak of around -18, so that I have lots of room to work with.  

Bob Bone

 

[CDK]

7 minutes ago, Robert Bone said:

Some VST instruments really peg their preset volumes way louder than they should be - there is no standard, so from one instrument to the next initial plugin instrument volume can be all over the place.  Lots of plugin authors like their presets to sound 'punchy/loud' and they are set to initial values that are way too loud.

Whenever I load a plugin instrument or live instrument/mic, I always set the combo of the plugin output volume and the assigned/associated audio track(s) gain setting to a peak of around -18, so that I have lots of room to work with.  

Bob Bone

 

So how would I set the plugin instrument's associated audio track/s gain setting to a peak of around -18? Is there an actual setting? Or would you just adjust the gain knob?

 

[CDK]

10 hours ago, pwalpwal said:

Each multi has its own audio output? Otherwise you're mixing that inside kontakt, but same, mix the audio rather than the midi

Do you mean each multi has its own audio output for each MIDI channel/instrument?

[lapasoa]

Instead to speculating about every single VST gain, it would be much better in your mixing to use your ear and your creativity and you'll be fine.

[Robert Bone]

20 hours ago, lapasoa said:

Instead to speculating about every single VST gain, it would be much better in your mixing to use your ear and your creativity and you'll be fine.

I do agree that using one's ears is a critical part of the whole music production, from start to finish, however I have seen many sources give -18 dBFS as a good target peak for digital audio recording.

From a Sound On Sound article on Gain Staging: "If you take the sound with the highest peak levels and set it so that it peaks at between -12 and -18 dBFS, you shouldn't run into problems with plug-ins or summing on the mix bus".

In any case, I initially set instrument volume and gain to a target of around -18 dBFS for my digital audio recording projects, without involving faders at that point.  (Avoid using faders for initial gain staging, because they affect signal after the plugins, and by adjusting instrument volume and gain, you are setting appropriate levels for sending signal to the plugins).

Bob Bone

[Robert Bone]

On 10/5/2019 at 8:55 PM, CDK said:

Do you mean each multi has its own audio output for each MIDI channel/instrument?

In Kontakt, you can route output of each loaded instrument too its own stereo output channel.  What I do, once I have loaded up to 4-5 instruments in a single instance of Kontakt, is to run one of the batch functions (available via drop-down menu in the Output Section), which is a function to clear out whatever is in the Output Section, and then it creates a separate stereo output channel for each loaded instrument - and properly assigns them in the same order as the instruments were loaded. 

Please note that when it does that, it mashes each instrument name into its associated stereo output channel name, so after running that batch function, I then double-click on each output channel in the output section, giving each a cleaner looking name, like Piano, Bass, Organ, Rhodes, etc., instead of the mashed up names it gives them. It just takes a minute to rename the resulting output channels, and is easier on the eyes, to see reasonable names at a quick glance.

Bob Bone

[CDK]

On 10/6/2019 at 12:43 PM, Robert Bone said:

Some VST instruments really peg their preset volumes way louder than they should be - there is no standard, so from one instrument to the next initial plugin instrument volume can be all over the place.  Lots of plugin authors like their presets to sound 'punchy/loud' and they are set to initial values that are way too loud.

Whenever I load a plugin instrument or live instrument/mic, I always set the combo of the plugin output volume and the assigned/associated audio track(s) gain setting to a peak of around -18, so that I have lots of room to work with.  

Bob Bone

 

So just to clarify, for a VST instrument for example, you set the associated audio track/s gain so it's peaking around -18dB? Or do you lower the fader the on the synth audio track?

Amongst the the things I've read/watched this week one suggestion has been to set the audio track/s level at -10dB either in the composing (MIDI track) stage or at the mixing stage (in a new project after exporting all MIDI tracks to audio); then I'll have plenty of room to play with when mixing/mastering the final product. Does that make sense to you?

[Robert Bone]

Just to be clear - I am NOT an audio guru - I am a keyboard player that does his best with the music production process.  All I can do is tell you what I do.  I do try to confirm things I write about, but still sometimes get things wrong.

I suggest looking at some articles/videos about the topic of Gain Staging, which is the process of managing gain through the path of the signal from the input at the top of the channel strip, through the output at the bottom of the channel strip.

Here is a pretty good one, from Sound On Sound: https://www.soundonsound.com/techniques/gain-staging-your-daw-software 

It also helps to have an idea of how signal flows through Cakewalk.  Here is a link to that (the basic flow through a channel strip is from top to bottom): 

https://www.cakewalk.com/Documentation?product=Cakewalk&language=4&help=Mixing.07.html

Gain is the level of the input signal and volume is the level of the output signal, for a gain stage, and the term Gain Stage refers to every point along the signal path, where the gain can be set, so Gain Staging is the process of setting/managing gain at each gain stage along the signal path. 

This is true as well, through each effect in an effect chain - each effect has an input gain, the effect processing, possible make-up gain, and output, and on through the next effect om the chain, etc.  Gain Staging is best done in a top-down fashion, following the signal path.

Basically, I want to end up with a clean mix, with no clipping or noise, so I try to properly set gain at every point along its path (  If the signal is too high, it can cause clipping, or not get the best result from an effect (even if not clipping it can be too hot for best results when going through an effect), and if the signal drops too low, I may have to add a bunch of gain back in, which can introduce noise.  (not like back in the analog/tape days, because the digital noise floor or balance of signal to noise, is just amazing these days in the digital world).

I start with faders at Unity, and I leave them alone when setting gain staging.  They control volume after the signal has already gone through any channel effects, and they are most precise within the vicinity of unity, so I do not involve them in gain staging - they are for balancing relative volumes of the different channel strips. .  start by looking at the loaded instrument's own main output volume - this is often set really hot, so I start by dialing that down in the instrument, if needed, sometimes by a bunch - to get it down to somewhere around -14 to -15, and then dropping it down some more with the track gain knob, to get it initially set to about -18.  This usually leaves me plenty of room for using additional mix buses without bringing the signal too high by the time it gets to the master bus.

I do an A/B (before and after) comparison of what affect an effect has on the signal, by turning that effect off/on and making sure I haven't either raised the signal or dropped the signal, as a result of going through that effect.

The earlier gain is set properly, the better, (meaning as high in the signal chain as possible), and that starts at the instrument output volume.    If you try to manage gain using your faders, you are too far down the signal chain, as your faders come in after your signal has already gone through and returned from your effects.

Others can do a far better job than I, at explaining the above, I also haven't had any coffee yet, so apologies for any confusing or ignorant statements above.

Bob Bone

[David Baay]

2 hours ago, CDK said:

On 10/5/2019 at 7:43 PM, Robert Bone said:

Whenever I load a plugin instrument or live instrument/mic, I always set the combo of the plugin output volume and the assigned/associated audio track(s) gain setting to a peak of around -18

So just to clarify, for a VST instrument for example, you set the associated audio track/s gain so it's peaking around -18dB? Or do you lower the fader the on the synth audio track?

Note that Bob said he sets the combination of instrument Output and audio track Gain to get the desired input level to the track. Ignoring  live hardware inputs for the moment since they aren't affected by track gain, and  this thread is about soft synths, the Instrument output level will be determined either by a fader/knob in the synth GUI, or - as I described earlier - by the MIDI track Volume.  Gain is the knob at the top of the track Inspector or the fader labeled 'Gain' in and audio/synth track header (when showing Mix or All controls). Note that on a Simple Instrument track that hasn't been split into separate MIDI and audio 'Synth' tracks, you have to use the Audio tab of the track Inspector to access audio Gain.

Usually when people say 'fader' without any reference, they mean the output Volume fader.  What Bob and others are advocating is to get the input Gain level  (what you see when an audio/synth track is armed for recording) to some nominal value like peaking at -18dB), and then use the output Volume level to control the level of that instrument in the mix.

There is no fixed rule for where to set the track Volume fader to start. The more tracks you have, the lower each one has to be for the sum to stay under 0dB on the Master bus with Gain and Volume controls at unity (0dB). In a DAW, buses effectively have unlimited headroom at the input so, if you want, you can control the summed level by simply pulling the input Gain on the bus down, and not worry about the fact that somewhere in the DAW's floating-point math inner workings, the summed level is theoretically above full scale.  But in order to get a good feel for how it all works, it's a best practice to  pull your track output Volumes down to keep the sum peaking  under ~6dB at the input to the Master bus.

More food for thought. I would suggest you close your browser at this point, and go play with all the controls in Cakewalk to see what they do.  ;^)

[Robert Bone]

25 minutes ago, David Baay said:

Note that Bob said he sets the combination of instrument Output and audio track Gain to get the desired input level to the track. Ignoring  live hardware inputs for the moment since they aren't affected by track gain, and  this thread is about soft synths, the Instrument output level will be determined either by a fader/knob in the synth GUI, or - as I described earlier - by the MIDI track Volume.  Gain is the knob at the top of the track Inspector or the fader labeled 'Gain' in and audio/synth track header (when showing Mix or All controls). Note that on a Simple Instrument track that hasn't been split into separate MIDI and audio 'Synth' tracks, you have to use the Audio tab of the track Inspector to access audio Gain.

Usually when people say 'fader' without any reference, they mean the output Volume fader.  What Bob and others are advocating is to get the input Gain level  (what you see when an audio/synth track is armed for recording) to some nominal value like peaking at -18dB), and then use the output Volume level to control the level of that instrument in the mix.

There is no fixed rule for where to set the track Volume fader to start. The more tracks you have, the lower each one has to be for the sum to stay under 0dB on the Master bus with Gain and Volume controls at unity (0dB). In a DAW, buses effectively have unlimited headroom at the input so, if you want, you can control the summed level by simply pulling the input Gain on the bus down, and not worry about the fact that somewhere in the DAW's floating-point math inner workings, the summed level is theoretically above full scale.  But in order to get a good feel for how it all works, it's a best practice to  pull your track output Volumes down to keep the sum peaking  under ~6dB at the input to the Master bus.

More food for thought. I would suggest you close your browser at this point, and go play with all the controls in Cakewalk to see what they do.  ;^)

One slight addition to the notion of why it is suggested to do all the Gain Staging with faders set to Unity and left alone at that point. 

If you look at a fader, they are purposefully created so that the closer it is to Unity, the more precise the adjustments are, as the fader is moved.  In other words, the scale of movement is quite different, near Unity, than it is if you look at, say, -30 dB.  SO, it is generally recommended to get all your Gain Staging set while leaving the faders at Unity, so that when you DO start to use them to actually adjust relative volumes between tracks, very precise adjustments can be made.  They build them this way for that very reason.

Bob Bone

 

[John Vere]

To solve the issue of different VST's having this crazy all over the map outputs I use custom presets for the ones I use the most. I also have track templates saved where I've already tweaked them to my liking. The default levels are normally a bit loud. I think they do this because of the "Everything will sound better if it's louder" rule.  

[Robert Bone]

11 hours ago, Cactus Music said:

To solve the issue of different VST's having this crazy all over the map outputs I use custom presets for the ones I use the most. I also have track templates saved where I've already tweaked them to my liking. The default levels are normally a bit loud. I think they do this because of the "Everything will sound better if it's louder" rule.  

EGGGGGZACTLY!  Yup - I absolutely LOVE track templates, and in addition to setting up one for every drum kit I use, I also have some standard ones for individual synths with specific presets, including associated audio and midi tracks and settings, and some commonly grouped instruments I use with Kontakt - like Piano, Strings, Bass, and Organ.  It is nice to have all the levels, tracks, sounds, and assignments set up and brought into projects as needed, with just a couple of mouse clicks.

I also have some standard Project Templates set up, that I can quickly select from and launch one, when I get some noodle running around in my head, and want to capture it in some fashion - prior to it evaporating from my brain.  These would include the whole raft of stuff connected to a Battery 4 drum kit or AD2, as well as a few flavors of instruments - all done to speedily get me up and running in a pre-fab project to record my noodles.  (I often tap out odd-metered rhythms on the steering wheel when out and about, and will launch one of these noodle templates as soon as I get back home, before my brain switches to another channel).

Life is SO much easier when levels are saved with already reasonably set output levels - so rather than having to do identical gain staging anytime I load the same Kontakt piano, I will - (ahead of time) - get the desired one loaded up and its output volume dropped down in the neighborhood of -18 dB, and then save it in the same folder as it would normally come from, with a new name - (nothing fancy - I just add "00-GS", without the quotes, to the front of its original instrument name).  That way, whenever I want to open the folder for that piano, to see the available instruments to choose from, any of the ones I already set levels for show up alphabetically at the top of the instrument list, so I see those first.

I just LOVES Me those Track and Project Templates in CbB. 

Bob Bone

[arlen2133]

3 hours ago, Robert Bone said:

EGGGGGZACTLY!  Yup - I absolutely LOVE track templates, and in addition to setting up one for every drum kit I use, I also have some standard ones for individual synths with specific presets, including associated audio and midi tracks and settings, and some commonly grouped instruments I use with Kontakt - like Piano, Strings, Bass, and Organ.  It is nice to have all the levels, tracks, sounds, and assignments set up and brought into projects as needed, with just a couple of mouse clicks.

I also have some standard Project Templates set up, that I can quickly select from and launch one, when I get some noodle running around in my head, and want to capture it in some fashion - prior to it evaporating from my brain.  These would include the whole raft of stuff connected to a Battery 4 drum kit or AD2, as well as a few flavors of instruments - all done to speedily get me up and running in a pre-fab project to record my noodles.  (I often tap out odd-metered rhythms on the steering wheel when out and about, and will launch one of these noodle templates as soon as I get back home, before my brain switches to another channel).

Life is SO much easier when levels are saved with already reasonably set output levels - so rather than having to do identical gain staging anytime I load the same Kontakt piano, I will - (ahead of time) - get the desired one loaded up and its output volume dropped down in the neighborhood of -18 dB, and then save it in the same folder as it would normally come from, with a new name - (nothing fancy - I just add "00-GS", without the quotes, to the front of its original instrument name).  That way, whenever I want to open the folder for that piano, to see the available instruments to choose from, any of the ones I already set levels for show up alphabetically at the top of the instrument list, so I see those first.

I just LOVES Me those Track and Project Templates in CbB. 

Bob Bone

+ 10,000

All that right there! ?