Back to basics - simple leveling understanding?

B

benherron.rrr

New member
Ok, so I understand the use of panning, effects an dynamic processing and have a good grip on them. My current trail of thourght lies with the simple leveling/balancing of the mix. To help you help me im going to try and paint a picture of a scenario in which my question exists:

Lets take a simple alternate rock song, with a simple Verse, Chorus, Verse, Chourus, Mid 8, Chorus. With simple instrumentation: Drum kit, bass guitar, Electric guitar (rhythm, lead and the bits inbetween), Vocals and backing. Hypotheticly speaking: The first verse has the kit, bass, rhythm guitar and a single lead vocal. The mixing is fine the Kit is singing with the bass, the vocals nicely sitting on top with the guitar nicely blended in all peaking at lets say -3 on the master fader. But when we get into the chorus we have the addition of backing vocals, a lead guitar part. and relative doubleing.

This is where my question lies: Obviously with the addition of new parts the over all volume of the part is louder, it would begin peaking. And the difference between the verse and chorus volume wise becomes very noticable. So the next part is asking the question; How do I level out the volumes? If you begin messing around with the volumes of each part via automation, it ends up throwing the balencing out, creating different dynamics between the trasitions, when you were perfectly happy with what you had. with the same effect if you start turning down cirtain parts. You could then place a compressor on the master fader to squash everything down to level out the track, but then you begin killing the dynamics, again which is not what you want. Would the best idea be to automate the master fader? so you have the verse peaking around -3 and the chorus doing the same, without changing the balance or killing dynamics, but keeping the overall listening level around the same.

I hope you kinda know what I am trying to explain, and you opinions would be very welcome.
 
I think you're over-thinking this. You really have two choices..

  1. Pull down all the levels so the verses are quieter, giving headroom for the increase in levels during the choruses.
  2. Dampen down (or kill off) some of the dynamics... mix choruses so you don't have a big jump in levels (addition of backing vocals shouldn't need cause a big raise peak levels)... generally homogenise everything together, etc.

Though #1 would probably be the better idea, #2 is generally what people do.

Remember that peak levels don't give a direct indication of perceived volume/loudness... you can have both your verse and chorus peaking close to full scale, yet the average/RMS levels (and resulting perceived loudness) can vary hugely.

Have an experiment with compression and see if you can make your choruses seem more "dense" and have a higher average level, without any increase in peak level over the choruses.
 
benherron.rrr said:
But when we get into the chorus we have the addition of backing vocals, a lead guitar part. and relative doubleing.

This is where my question lies: Obviously with the addition of new parts the over all volume of the part is louder, it would begin peaking.
...
So the next part is asking the question; How do I level out the volumes? If you begin messing around with the volumes of each part via automation, it ends up throwing the balencing out, creating different dynamics between the trasitions
Click to expand...
In addition to what mattr already said, when you add parts, the dynamic relationships between the tracks*do* change and probably *should* change. One should not expect by default that the tracks that fill the verses will fill the choruses in the same way, or even that the way they fill verse one is the same way they fill verse three.

Part of this at least should be reflected in the song's arrangement. One cannot and should not expect every instrument to sound like a lead all the time at the same time. If you have a wall of sound in the verses meaning that you have no room for more in the chorus, then something has probably got to give. Either you have to remove some of that wall or remove some of the extra stuff you want in the chorus; you can't fit eight pounds of sand into a five pound bag.

If you add some lead guitar in the chorus, the rhythm guitars may need to recede a bit while the lead is actually there (for example). Sometimes some of the stuff used to fill the verse should just disappear during the chorus (and vice versa). If you add some backup vocals such as a chorus to layer with the lead vocal, then *layer* them, meaning the lead may have to come down and meet the chorus rather than lay on top of them.

Dynamic changes in the pan space are sometimes called for also. It's easier to avoid clipping when you put this in the left channel and that in the right channel, even if this was panned center during the verse. That kind of dynamic panning is not uncommon.

G.
 
Thanks all for your help. My question isn't really regarding arrangement or really the dynamics of the song as a whole its more like: Imagin having a mix you are more than happy with, but due to the arrangement of the song, it can go from say, a piano and voice in a verse to a sudden transition to a full band making up a chorus. The mix is fine your happy with it but the transition just makes you jump the difference in volume is too severe. if you follow me? just saying that if I wanted to level that out to releave listening fatigue or to get rid of the physical element of surprise. what would be the most appropriate/ simplest way to get around this?
 
That might be something you want to stick in the MP3 clinic so folks can hear exactly what's going on, but it seems pretty straightforward the way you describe it - i.e. remove most of the difference in volume.

Bring the full band in, but bring them in combined at roughly the same perceived level (more RMS-influenced than peak-influenced) that the piano was at. Maybe a couple of dB more - it's not uncommon to have the chorus be 2-3dB louder than the verse - though I think that's a cheap technique most of the time myself.

Either way, again, if you have just piano and vocal in the verse, the piano can stand to be dominant, but when you bring everything else in along with it in the chorus, each individual track, including the piano, probably, probably needs to be reduced somewhat so that the sum of them isn't all that much louder than the verses.

Exact numbers are not predictable, but sometimes all it takes as 1.5-2dB off of each track to make the sum of them fit OK.

G.
 
Last edited:
benherron.rrr said:
Ok, so I understand the use of panning, effects an dynamic processing and have a good grip on them. My current trail of thourght lies with the simple leveling/balancing of the mix. To help you help me im going to try and paint a picture of a scenario in which my question exists:

Lets take a simple alternate rock song, with a simple Verse, Chorus, Verse, Chourus, Mid 8, Chorus. With simple instrumentation: Drum kit, bass guitar, Electric guitar (rhythm, lead and the bits inbetween), Vocals and backing. Hypotheticly speaking: The first verse has the kit, bass, rhythm guitar and a single lead vocal. The mixing is fine the Kit is singing with the bass, the vocals nicely sitting on top with the guitar nicely blended in all peaking at lets say -3 on the master fader. But when we get into the chorus we have the addition of backing vocals, a lead guitar part. and relative doubleing.

This is where my question lies: Obviously with the addition of new parts the over all volume of the part is louder, it would begin peaking. And the difference between the verse and chorus volume wise becomes very noticable. So the next part is asking the question; How do I level out the volumes? If you begin messing around with the volumes of each part via automation, it ends up throwing the balencing out, creating different dynamics between the trasitions, when you were perfectly happy with what you had. with the same effect if you start turning down cirtain parts. You could then place a compressor on the master fader to squash everything down to level out the track, but then you begin killing the dynamics, again which is not what you want. Would the best idea be to automate the master fader? so you have the verse peaking around -3 and the chorus doing the same, without changing the balance or killing dynamics, but keeping the overall listening level around the same.

I hope you kinda know what I am trying to explain, and you opinions would be very welcome.
Click to expand...

I don't understand the problem. If you don't have the headroom for the chorus, then why not bring down all the faders?

I can't tell you based on your description how much of a dynamic range you're using, but I can tell you, anything over 4dB of range on a modern rock track is about the limit given today's listening habits. That may seem small, but when you start extending beyond that, you'll find the verses will get lost (or chorus' will rip your head off) in real-world listening environments like your car.

This is an important consideration. People don't tend to sit down and listen to an album anymore. They tend to play music in the background, and if your dynamic range is too broad your exciting studio mix in the isolation of a control room can be rendered impotent everywhere else.

I recommend a hardware stereo compressor for mixing for a number of reasons: It greatly speeds up the mixing process. It gets the music out of the box. And last but not least, tt protects the balances in your mix when the ME does what you failed to do, use a compressor. Unfortunately, the MEs compressor changes all of your balances, and that mix you spent all that time on will completely change. Compressors on a mix affect internal balances. You as the mixer want to strap the compressor on your mix.

I understand the desire to have an exceptionally exciting and dynamic mix, but there are things that we can do as mixers to create the illusion of a larger dynamic range, and that's with contrast. Based on your description of the arrangement, it sounds like you have some decent contrast between the verses and chorus, what with the addition of background vocals, a lead guitar and a rhythm guitar. So, you might not need as much range as you're using, particularly if you want your mix to be effective anywhere other than a home theater. I can't say with absolute certainty that you're using too much dynamic range without hearing the mix. It could be you have nothing more than a gain staging issue, but at the very least you should be aware of this potential problem when you're mixing.

I'm guessing you don't have a hardware stereo compressor worthy of the 2-bus, so until you do, you might want to experiment with plugin compressors. I'm running out of time for this post, so in the interest of brevity, here's what I have to say in regards to this in Zen and the Art of Mixing:

"Your stereo bus compressor is the one compressor that you would be ill-advised to use in the plug-in form. This might be reasonable on a temporary basis as you learn (and earn!), but analog compressors are far better than their digital counterparts for the job. This is especially so when you directly compare the plug-in to the analog compressor it emulates. You can perform about 80 percent of the heavy lifting where compression is concerned with a hardware 2-bus compressor, but it’d better be pretty damn stellar at the job. If you get the impression that I freak out about everything and anything that goes on the 2-bus, you’re right; and there’s a good reason for this. The 2-bus affects the entire mix!"

Enjoy,

Mixerman
 
Last edited:
Mixerman said:
I can't tell you based on your description how much of a dynamic range you're using, but I can tell you, anything over 4dB of range on a modern rock track is about the limit given today's listening habits. That may seem small, but when you start extending beyond that, you'll find the verses will get lost (or chorus' will rip your head off) in real-world listening environments like your car.

This is an important consideration. People don't tend to sit down and listen to an album anymore. They tend to play music in the background, and if your dynamic range is too broad your exciting studio mix in the isolation of a control room can be rendered impotent everywhere else.
Click to expand...
Mixerman, you give some great advise and I have learned much from your writing, but I can't agree with this one.

*You should want to rip somebody's head off with a rock chorus. It's insane that aggressive music bends over backwards to not shock people.
*The bit about losing the music in the car is false aside from maybe a classical recording (I never struggled listening to a rock CD in my car in '93).

And the big one: Yes, your music might not work as background sound if it gets mildly dynamic. Yes, the current trend is to listen to music as background sound. Why on earth would you want to release something that falls into that listening pattern? Make something that can be background and since that is how people currently listen, that is what it will be. Make something that demands even a little bit of volume above "comfortable and ignorable" so as to not lose the softer bits, and now the listener is actually forced to pay attention when it gets loud. It seems like suicide to willfully present your product as non-intrusive filler.
 
It's time to go back to basics a bit here.

On any digital recording system, 0dB is 0dB(FS) (Full Scale) and is the point where digital clipping starts to occur. This equates to something between +18dB and +22dB on an analogue scale. Your -3dB level is actually somewhere around +15dB and, in a real world, you don't record everything at +15.

Besides this, the way the dB scale works, if you add two signals at -3, you end up at 0. If you add another one, you're into clipping.

With any decent gear, you can afford to have all your tracks in the -6 to -12dB (FS) range. Your noise floor should be far enough down that it's still inaudible even if you throw away the top few bits--but this gives you headroom to do your mix easily. It also makes certain that you don't lose a good take due to clipping if somebody gets over-excited and goes a bit louder than the previous six takes.

I understand what you say about not wanting to start pulling everything down once you have the mix almost there--in theory you can back everything off six dB and it should be exactly the same (but quieter) but even subtle differences can change what you hear. That's why I suggest STARTING your mix with everything a bit lower.

(And, yeah, I'm showing my old-git-recording-on-an-8-track-Brennell origins here!)

Bob
 
Chibi Nappa said:
And the big one: Yes, your music might not work as background sound if it gets mildly dynamic. Yes, the current trend is to listen to music as background sound. Why on earth would you want to release something that falls into that listening pattern? Make something that can be background and since that is how people currently listen, that is what it will be. Make something that demands even a little bit of volume above "comfortable and ignorable" so as to not lose the softer bits, and now the listener is actually forced to pay attention when it gets loud. It seems like suicide to willfully present your product as non-intrusive filler.
Click to expand...

I'd say this is a horses for courses kind of thing

I by no means can speak for mixerman or anyone else but he is a commercial mixer and in his books is generally coming from a standpoint of mixing with a view to selling in the current market and in that case, to appeal to a large market, it would be smart to mix with current trends in mind, since that is what the biggest segment of the market wants to consume. So if mass, generic appeal is what you are after generally you need to homogenize your music

If you are mixing for a small, very dedicated, knowlegable, audiofile fan base who want to sit down with your record and really listen to it in all it's dynamic glory and have come to expect big, head exploding choruses from your live shows then have at it. In this case creating what might have a more mass appeal could be a mistake and could hurt your dedicated fans, as they might see your attempt to be homogenous as selling out (just like what happpened to Green Day when they got popular)

If your are mixing your songs for your own enjoyment then do what makes you happy and really what anyone else wants or thinks is not relevant since this is for your own consumption

The right answer for you is the one that fits your circumstance

YMMV
 
If you're shooting for flash in the pan popularity and making an instant paycheck, mix for the market. If you're shooting for still being on everybody's playlist 20 years from now, mix for the music.

G.
 
Bristol Posse said:
If you are mixing for a small, very dedicated, knowlegable, audiofile fan base who want to sit down with your record and really listen to it in all it's dynamic glory and have come to expect big, head exploding choruses from your live shows then have at it.
Click to expand...
I don't know where the idea that dynamics are only for audiophiles came from, but it is a cancer and it makes no sense. So scales, melodies, and chord structures can be understood by pop fans and can be used in pop music. Syncopation, compound beats, harmony, and arrangements featuring 15 instruments are low-brow enough for knuckle-draggers to enjoy. But the difference between soft/loud? Arrgh! Caveman brain hurt! Not understand! We were trying to rock out, but the song got loud! Get that tea-sipping-extended-pinky-monocle-wearing-white-glove sissy stuff out of here! Maybe in Victorian England loud guitars took your head off in the chorus, but we're trying to be energetic kids here!

SouthSIDE Glen said:
If you're shooting for flash in the pan popularity and making an instant paycheck, mix for the market. If you're shooting for still being on everybody's playlist 20 years from now, mix for the music.

G.
Click to expand...
I think you could still be insta-popular if you included dynamics. I really don't think the fans shun it.
 
Chibi Nappa said:
I think you could still be insta-popular if you included dynamics. I really don't think the fans shun it.
Click to expand...
I agree. I'm just saying, that fad and fashion works in music much the same way it does in clothing. The songs and productions that tend to have the best staying power across generations are not the ones that try so hard to follow the current fads or fashion. The songs from the 60s, 70s and 80s that tend to still be on the most playlists these days are not the protest and phychedellia of the 60s, the rock opera over-productions of the 70s or the synth rock of the 80s, but rather the somewhat more timeless and less current-trend-dependent productions from those eras. This applies to both music styles and production styles.

G.
 
thats the point of mastering. after u have all the tracks condensed to the final 2, just maually adjust the volume on those parts while u bounce record them to yet another 2 tracks. then the dynamics between parts cant change, but the volume can go down when u need it to
 
You must log in or register to reply here.
Back
Top