Vocal Effect Meters Peaking

[musikman1]

Hi

I've got a couple of lead vocal tracks, and a couple of harmony vocal tracks feeding one vocal sub mix. In the sub mix FX bin I have the following, in order top to bottom...EQ, Waves Doubler, CW VX-64 (just using delay and de=esser sections), and Waves TruVerb Reverb.   The individual lead vocal tracks faders are set to -1.0db, the harmony vox tracks faders are set to -4.0db, the Vox Sub Mix fader is set to roughly -2.4db.  When playing the track, none of the individual tracks meters or sub mix meters of the vocal are peaking over 0.0db, and the vocals are sitting nicely just above the music, and the Master channel is staying well below 0.0db.  When I check the individual plugin meters, it's a different story...

All of the FX gain meters are showing just under 0.0db as well, except for the Waves Doubler.  For some reason that one is hitting over 0.0db several times throughout the track.  Not by a lot, it's peaking to around 1.1db to 2.5db.  The 3 gain controls in the Doubler plugin can be set individually for the Direct signal, and the added/doubled Voices 1 & 2.   I have the Direct gain set at -4.0db, and both the added voices set at -5.5db, but I'm still seeing the meter in the red on and off throughout the track. I can get it to stay below 0.0, however I'm losing too much gain when I do that, which then makes it necessary to raise the entire vocal sub mix in order to get the vocals back to sitting above the music where they belong. 

I'm leaning towards just ignoring the peaking going on with the Doubler plugin, since it's not that much over, and I'm not hearing any distortion at all.  It's just strange that all the other plugs in the chain are fine.  I am thinking that because the Doubler plug is "adding" two extra voices to get that "fattening" effect is why it may be peaking.  So the dilemma is, if I start lowering the gain controls for the added voices in the Doubler, while keeping the Direct gain where it is, I lose too much of the effect I intended to get in the first place.  In turn, if I lower the added voices gains, AND the Direct gain,  I lose too much gain and now I'm reaching to raise the vox sub mix slider because the vocals are too soft in the mix.......which in turn, raising any of the faders will again begin peaking the Doubler plug's meters.  I even tried backing off on the output gain of the EQ plug, since it sits in line before the Doubler, and that helped a little, but again, I can't go too far, otherwise I lose too much signal. 

Any thoughts?  Thank you.

 

Edited May 17, 2022 by musikman1

[bitflipper]

Short answer: turn everything down.

You're right, the doubler is effectively adding two more tracks, which are summing with the main vocal. And because they are essentially copies of the main vocal, they're going to experience constructive interference, meaning a bigger jump in peak values than you might expect.

You probably won't hear any distortion, because you're working with floating-point data and that means there's no clipping even when you go into the red. However, it doesn't mean distortion (specifically digital overs, the nastiest-sounding kind of distortion) won't haunt you later because the real world isn't floating-point. This is even more likely if you end up exporting to MP3. Just pull back all the vocal tracks to give yourself 8-12 dB of headroom. Easiest way to do that is to group them and use the Gain slider. If you have compression on the vocal tracks, you'll have to go back and tweak the threshold setting, lowering it by the same amount.

[musikman1]

1 hour ago, bitflipper said:

Short answer: turn everything down.

Hey bitflipper, thanks for the reply.  I assume by everything, you also mean the music tracks. It seems in this situation in order to keep the vocals above the music I'd have to turn down the music tracks too if I take down the vox to avoid clipping the plugin. .  The submix faders for the rest of the instrument mixes are ranging from -4.0db to -9.0db.  (This would include AC Gtr, Elec Gtr, Bass, Drums, Keys.)  There's always that question of once I have the entire mix running with no peaking on anything, should I reach for the individual track faders or the submix faders to make overall mix adjustments...I do have a general method I follow...see below..

1 hour ago, bitflipper said:

Just pull back all the vocal tracks to give yourself 8-12 dB of headroom.

Pull back just the individual vocal tracks, or also the vocal submix fader?  What I usually do initially is try to get a fairly strong signal on the individual tracks (no higher than -3.0db) then once I have all the individual music tracks balanced between themselves, and then balanced with the other instruments the way I want them, I bring them all down equally even further to give plenty of headroom in the Master bus.  If I need to take any instrument mix down even further in the overall mix after that, I'll generally just use the submix faders.  I'm not sure if this is 100% technically the optimal way to do it or not, but until now I haven't had any issues with peaking or not having enough headroom.  Please advise if you have any suggestions.  Thanks!

Edited May 18, 2022 by musikman1

[Byron Dickens]

2 hours ago, bitflipper said:

Short answer: turn everything down.

👍

Some problems really do have simple solutions.

[Bristol_Jonesey]

I would also run any de-esser on individual tracks, not submixes

[musikman1]

Thanks, I will give all the suggestions a try, very much appreciated!   I like when the solution is simple! :-)

2 hours ago, Bristol_Jonesey said:

I would also run any de-esser on individual tracks, not submixes

That de-esser is part of CW's vocal strip, and I also  have the delay section turned on.  So I can  just turn off the de-esser section and keep the delay on the submix.  I don't think the de-esser is imperative anyway on this particular vocal track.  I should be good with just the delay.   I will make note of your suggestion for future reference though, thank you.

[bitflipper]

20 hours ago, musikman1 said:

What I usually do initially is try to get a fairly strong signal on the individual tracks (no higher than -3.0db)

For a long time that was standard practice, and for good reasons. With digital audio, however, gain-staging becomes less important. (Not unimportant, but less important). There is really no benefit to maximizing individual tracks, not like when we were battling noise in the analog domain. In a DAW It's perfectly OK to have your tracks peaking at -12 dB or lower. It preempts other potential problems down the chain.

And yes, if you need to turn the vocals down you'll likely also have to turn instruments down by the same amount. That's OK, even if the signal going into the master bus seems low, even if you have to turn up your monitors. Don't worry about the full mix volume while you're mixing. Wait until the mix sounds good, then as a final step use a limiter to bring the overall volume up to your final target levels.

[musikman1]

Thanks for clarifying bitflipper. Makes sense. Yeah I'm an older retired guy so I guess I still have that mindset of the days when we were gain staging with all that analog equip.  I remember watching my band's sound engineer, and picking up pointers from him as he was gain staging the board.  The good ol' days! lol

5 hours ago, bitflipper said:

Don't worry about the full mix volume while you're mixing. Wait until the mix sounds good, then as a final step use a limiter to bring the overall volume up to your final target levels.

What I've been used to doing is once I get the mix sounding good, and a good level balance between all the instruments and vocals, I then usually bring everything up equally until I've got the Master channel meters hitting no higher than between -3.0db to -5.0db, and then I add a limiter at that point to get a little punch without ever peaking anything.  So let me get this straight, are you saying to add the limiter to the Master channel while it's still down at say -12.0db without ever bringing up the levels of all the tracks once the mix sounds good?

Edited May 19, 2022 by musikman1

[bitflipper]

Yeh. What I'm saying is don't worry about the overall levels while mixing, as long as they're reasonable (e.g. peaking at -12 dB). If it sounds too quiet, turn up your monitors to hear what it ideally should sound like at commercial levels. If you subsequently master with a gentle hand, it'll retain much of the punch and dynamics in the pre-mastered mix. If it still sounds weak with the limiter, disable it and revisit the tracks, which may be in need of additional compression to make them sound fuller.

This, btw, is not universal advice. Many insert a limiter early in the process so that they can hear its effect on the mix in progress. I prefer to hold off on that step until late in the mix process. If the mix doesn't stand on its own pre-master, then it's not ready for mastering.

[John Vere]

I have developed a workflow that involves using the gain on the tracks to set a target of around—8.0db to -12db depending on my mixing goal for that track. I leave all faders at unity. 

The vocals are the exception where I will go a bit higher  -6.0 db to no higher than-4.0db. 
like you I will combine lead and harmony to a sub bus. I also treat  the gain of all sub buses like tracks. Faders at unity. 
If adding effects pushed the level past my target I will turn down the gain. Using the faders won’t solve the issue. The gain is before the effects and the faders after. So if an effect is being over driven you need to use the gain. I will only use the bus faders a little bit in the final mixing stage. 
If you’re interested in my workflow watch #22

Oh and reading this again, most De-essers have a preview where you can listen to what it's actually removing. I know the Izotope Nectar has this and when I listen I usually hear nothing so I remove it. 

Edited May 19, 2022 by John Vere

[musikman1]

11 hours ago, bitflipper said:

If you subsequently master with a gentle hand, it'll retain much of the punch and dynamics in the pre-mastered mix. If it still sounds weak with the limiter, disable it and revisit the tracks, which may be in need of additional compression to make them sound fuller.

So what you mean here is to add the limiter while keeping the overall levels unchanged at -12.0db, as opposed to first bringing up the overall levels closer to 0.0db and then applying the limiter. (which is what I have typically done in the past to get more punch).   I will give that a try, thank you.  I have noticed that some of the presets in my limiter add +3.0db, or +6.0db, or +9.0db, and some have choices between 16bit dithered or 24 bit dithered, Harmonics 1 or Harmonics 2, Intersample, etc..so I have choices at least). So with an overall level of roughly -12.0db, add in the +6.0db with the limiter, and that would bring everything up to -6.0db. Export that out to wav and then use my mastering software to bring it up closer to 0.0db.   That sound about right?

10 hours ago, John Vere said:

If adding effects pushed the level past my target I will turn down the gain. Using the faders won’t solve the issue. The gain is before the effects and the faders after. So if an effect is being over driven you need to use the gain. I will only use the bus faders a little bit in the final mixing stage.

John, this is very interesting to me, I guess because I typically have thought of the Gain knobs as useful only if I can't quite get a strong original recorded signal and it needs a little boost.  So using the Gain control "instead" of the faders is a concept I'm going to need to experiment with.  Something you said is key to my current situation with the vocal track FX.  "The gain is before the FX and the Faders after".   It does make perfect sense to use the gain if one of the FX is peaking.  It's crazy, I just never thought of reaching for the gain knob, I usually grab for either the track or submix faders.   Also, the fact that only one of the FX in the chain was peaking at first kinda threw me, although after more thought I figured it had to be because the doubler was adding 2 extra voices, thus adding more level. 

I will be tuning in to #22. 

All great advice here, thanks so much guys. I'm a musician first, sound engineer second, but I've recently made a decision to continue to learn how to achieve better mixes.  Sometimes I get a little buried in the technical aspect.  I just need a good basic method that I can start with for every project.  In fact I'm about to start another thread to try to learn how other users approach getting a nice in-your-face bass track.  Something that has eluded me now and then, depending on the kind of bass source I use. 

 

[John Vere]

Glad that my info helped. Understanding Cakewalks signal flow is paramount to good mixing skills. It’s simply top to bottom more or less. So the Gain is very critical as it sets the level for the whole track channel strip. This includes pro channel, the effects bin. So keeping the faders at unity is the only way to see what is really going on. 

[musikman1]

All makes sense and every time I think I'm getting somewhere, I realize I have much still to learn!  Thanks again, it's a process I intend to improve on, and with all the good friends' advice here, I don't see any reason to fail!

Continuing my effort to learn how to get better mixes, here's the thread I just posted regarding bass....

 

[Craig Anderton]

On 5/18/2022 at 5:41 PM, bitflipper said:

For a long time that was standard practice, and for good reasons. With digital audio, however, gain-staging becomes less important. (Not unimportant, but less important). There is really no benefit to maximizing individual tracks, not like when we were battling noise in the analog domain. In a DAW It's perfectly OK to have your tracks peaking at -12 dB or lower. It preempts other potential problems down the chain.

+1. However, there's one reason I would recommend aiming for a relatively consistent level - e.g., not have some tracks peaking  at -3, and others at -18, etc. Compressors, amp sims, and other level-sensitive effects are affected by input level, because that determines how often the signal level goes over the threshold or starts saturating. For example, my amp sim presets assume an input level that peaks around -3 dB. That way if I normalize the guitar to -3 dB before mixing, then I know the amp sim preset will sound as intended. 

With compressors, I include the intended amount of gain reduction in the preset name, like "Warm vocal -6." Then I know to adjust the level coming in to provide maximum gain reduction of 6 dB. That doesn't mean I won't tweak the compression threshold, but it gives a consistent starting point.

[musikman1]

On 5/21/2022 at 11:20 AM, Craig Anderton said:

However, there's one reason I would recommend aiming for a relatively consistent level - e.g., not have some tracks peaking  at -3, and others at -18, etc.

Craig, are you referring to the individual tracks, or the submixed tracks? ...or both?  I've also been thinking about what John had mentioned about keeping all his faders at 0.0 and just using the gain control to cut or boost as needed.  As I said before that is a different approach than how I've been used to doing things, using the faders to set mixing levels, instead of the gain.  I know my brain is fatigued from reading all these helpful suggestions, LOL🤪, but I guess what I'm trying to figure out is if you're saying to keep all the faders on both the individual tracks and submix tracks at 0.0 and use the gain to level them all out roughly so that all the meters sit at a consistent given number, like -12.0db for example sake, (or whatever number you choose), then....well  I don't think I'm understanding this right, because if I set everything the same, I won't be able to mix all the instruments so that they blend correctly without one or some of them going out of the desired meter range.  Setting everything the same there's going to surely be some tracks that are too loud or too soft in comparison to the rest, no?  Unless I'm likely misjudging how using the gain control will affect the meters, which given my tired brain, I probably am!

Edited May 26, 2022 by musikman1

[musikman1]

On 5/19/2022 at 11:07 AM, John Vere said:

I have developed a workflow that involves using the gain on the tracks to set a target of around—8.0db to -12db depending on my mixing goal for that track. I leave all faders at unity. 

The vocals are the exception where I will go a bit higher  -6.0 db to no higher than-4.0db. 
like you I will combine lead and harmony to a sub bus. I also treat  the gain of all sub buses like tracks. Faders at unity. 
If adding effects pushed the level past my target I will turn down the gain. Using the faders won’t solve the issue. The gain is before the effects and the faders after. So if an effect is being over driven you need to use the gain. I will only use the bus faders a little bit in the final mixing stage. 

   After watching your video again John (#22), I noticed that you do pretty much have all the faders for both individual tracks and submixes set at 0.0.  This is what I was referring to in my previous post.  With all the faders set at 0.0, John is setting his peak meter target between -8.0db and -12.0db.  I assume this means ALL the peak meters, both individual tracks and submixes. At least that's what I noticed on the video.  All the meters were roughly between that selected -8.0 to -12.0 range, although in the video, the Master fader was not at 0.0, more like -18.0, which I'm thinking may have been just for the purpose of balancing the mic level with the music level for making the video?  

In the video you mentioned some tracks can be below that target range, but what if a track needs to be above that range in order to be heard in the mix? An example of what I'm trying to ask is, what if you set everything so the meters are all between that predetermined -8.0 to -12.0db range, and you find that the acoustic guitar is buried in the mix at that -8.0 to -12.0db level?  Do you raise the acoustic guitar track or acoustic gtr submix gain so that the meters go to -3.0db in order to balance it so it's no longer too soft? If every other track and submix is in the -8.0 to -12.0 range, you can't turn everything else down or then it puts everything else out of the target range. So how do you compensate?  I know it's not cut and dry, but just curious about this kind of scenario when trying to keep everything balanced.

I'm going to have to start the mixing process from scratch if I want to try using this method and see what comes of it, because I've been using faders mostly, so setting the faders all to 0.0db is going require me to remix from scratch using the gains. I'm guessing many of my questions will be answered in the process of trying it. We'll see.

[John Vere]

You got that right Yes I had to turn the master down or the video would blow your computer speakers up!  I usually mention this but might have forgot.

Yes for sure if something is not cutting through in the mix with my method turn it up.  Actually I don’t turn it up I will use a compressor and set the peak at around-3.0 db. Vocals always get this treatment. Acoustic guitar I’ll use EQ and compressors in that order. 
I have way to many compressors right now and still experimenting to make a solid recommendation 

[musikman1]

Thanks for clarifying John. Ok so you're basically setting the master fader at 0.0db as well.  Makes sense.  I have an old Joe Meek Comp/EQ (hardware unit), and I love how it sounds on vocals.  Last time I did a vocal track I used it. Problem is it gets recorded with the Compression on it, so if I want to add FX in later, I'll be adding whatever FX on top of the Joe Meek compressed vocal. I try to use a mild setting.  According to some of the comments I've noticed many are saying they use more than one compressor on the same track.   It's a really old unit, but gives the vocals a lot of nice presence.

Edited May 28, 2022 by musikman1