Problems with mixing/balancing levels

[findingthesound]

Hi everybody,
I'm currently recording/mixing my first album (metal/electro) and everything is about to be ok (=eqs & comp are ok, but I'll maybe re-record some Rhythm guitar tracks with a bit less comp and another gauge for the lower strings). I'm using SSD 3.5 for drums and love the sound.

However i have problems to make a good balance of the levels for the voice and the drums with others instruments.

I know that my mix (to be well prepared for mastering) should be peaking at -3dB and RMS at 14dB (mixing ITB) but the drums are very (too much?) dynamic since they have around 21/24 dB of Dynamic Range (D.R.= RMS-Peak), so if the snare peak at -3dB, RMS level is around -27dB RMS, so, really lower than the 14dB RMS level.
Do I have to put a limiter on the drums to reduce the D.R. to 12dB? (It'll sound bad but what to do?)

Others tracks are:

bass: peak= -9dB; RMS= -15
metal Rhythm guitars: peak= -7dB; RMS=16dB; D.R.= 9dB
Voice: peak= -8dB; RMS=-22db; D.R.= 14dB

there are some lead synth+lead guitar+percs+fx+...
But I'll make the levels by ears and move the faders more on those tracks.

What I really need is a strong Rhythm section (drums+bass+ Rhythm guits) and the voice sitting well just on top of this. I'm after an "american modern metal" kind of sound. (think of devil driver/underoath/trivium/lamb of god...)

I don't know how to achieve that since i'm going to master (hum...learn to master!) my tracks with some good eqs I have (sontec mep-250EX+ssl X-eq) + the FG-X mastering consol (as soon as it's available!) and/or some other limiters.
I don't know how to prepare my tracks with levels since i would put the voice "a hair" on the top of the snare, but from what I've understood, the limiter will cut the peaks of the drum, reducing the D.R. of the tracks from around 11dB to 6 or less dB to make it "commercially loud". I know it's not that simple (lots of different things to do in mastering like cleaning things by eq and more....)to do it in a musical way.

So If the voice before mastering is on top of the snare and kick, she will be distorted by the limiter...

So, how do you prepare your tracks for mastering?
How to set drums, vocals, bass and Rhythm guitars levels (peaks and RMS) to make the mix be ready for the cut drum's peaks part and have the voice at the right level?


I know that's not exact science and I will have some answers like "learn how to mix by ears and not by math", But I believe there's some "range" of levels that would be a good place to start for the kind of music I do (= modern deathmetal/electro). That's what I'm looking for.


A big thank you for those how will try to make my mixing knowledge evolved!!!
(and sorry for my non native english!)

 

[CraigTNelson]

I know that's not exact science and I will have some answers like "learn how to mix by ears and not by math", But I believe there's some "range" of levels that would be a good place to start for the kind of music I do (= modern deathmetal/electro). That's what I'm looking for.

You've answered the question right there. It is not the mix engineer's job to worry about overall loudness. This is why studio monitors come with volume knobs; mix at a conservative level, monitor at 85dB, and leave the rest to the mastering engineer.

I never A/B to other mixes when I mix, every mix is a unique puzzle and A/Bing usually takes me farther off track.

So, if you're working at 24-bit (most of us are), you don't have to worry about overall level since your noise floor is so staggeringly low. Make it sound as good as possible but always err on the side of caution to give the mastering engineer space to work since it's a pain in the ass to work on a limited track, but no problem at all to limit a dynamic track in mastering.

 

[-Noodles-]

I think you're over thinking things.

take the faders down 6dB and give yourself (or the ME) some room to work with later and turn your speakers up. I know you're already at (i presume minus) -14dBFS RMS, so aim for that with the whole mix - don't over think the dynamic range of one particular instrument unless it's sticking outside of the mix at times.
As Craig said, 24-bit resolution will give you a huge dynamic range - and that's a good thing. The noise floor is WAAAYYY south of where you are now.

If your ears are telling you that something (eg; drums) are too 'dynamic' for the track - compress them. Otherwise just leave 'em!

 

[findingthesound]

First thank you!
I've posted this thread in some other forums and got not an answer!

You've answered the question right there. It is not the mix engineer's job to worry about overall loudness. This is why studio monitors come with volume knobs; mix at a conservative level, monitor at 85dB, and leave the rest to the mastering engineer.

OK, but there are two things I don't know/understand:

-How I know that my speakers are set to 85dB? Is it related to the volume that get out of the speaker or a standard level like +4dBvu, -18dBFS ect...?
I've read this number on other post but never get it right.

-What are for you conservative levels?

So, if you're working at 24-bit (most of us are), you don't have to worry about overall level since your noise floor is so staggeringly low. Make it sound as good as possible but always err on the side of caution to give the mastering engineer space to work since it's a pain in the ass to work on a limited track, but no problem at all to limit a dynamic track in mastering.

I think you're over thinking things.

take the faders down 6dB and give yourself (or the ME) some room to work with later and turn your speakers up. I know you're already at (i presume minus) -14dBFS RMS, so aim for that with the whole mix - don't over think the dynamic range of one particular instrument unless it's sticking outside of the mix at times.
As Craig said, 24-bit resolution will give you a huge dynamic range - and that's a good thing. The noise floor is WAAAYYY south of where you are now.

If your ears are telling you that something (eg; drums) are too 'dynamic' for the track - compress them. Otherwise just leave 'em!

Ok so I keep my mix level at around -14dBFS RMS and peaks at -3 dBFS, Cut by half the dynamics of the drums with a limiter to fit with other instrument?

I thought that part of the mix is to compress things in a way instruments have the same kind of dynamic range (with bass a bit more compressed) to make them glue more together, avoiding the "demo" aspect on the mix...Don't know how to tell this better... generally around 12/14 of DR (sometimes 6 of DR for bass type of sound) i find things seems to work together well.
But this drum is really dynamic and I should have it almost limited and cut half of is dynamic to go to 14 of DR... And It'll sound terrible...
Maybe will try some aggressive parallel comp +limiting...
Any tips for a well done heavy comp? or just send Kick+Snare into a parallel limiter?


There is another problem too: when the double kick come, the dynamic range of the drums goes from the 20 dB to 11db of DR... What should I do??
cut drums track in two part and compress the low part and not the second?

About the vocals, how do you set the level and how much are they compressed in metal generally?

I really can't find how to make things properly with leveling and drums;
I'm going to be crazy since I've search the whole net for days to find an answer on my questions

you're alright I'm overthinking things, but that's because i'm alone with this f***** computer and don't know how to finish my first mix after a lot of work, time and money invested to build a kind of "pro" home studio...I'm the only thing not pro here!!!!!

In fact I have no confidence on my mixing skills and need some references to begin with.

And I can't find a finish mix of modern metal song with separate tracks to see how are things right before mastering... I'm dying to see this!...

Really need your help!

 

[CraigTNelson]

The only way to know if you're mixing at 85dB is to get an SPL meter.

When I'm mixing I don't measure anything inside the DAW. Just go for no clip lights on any individual tracks, plugins, or on the 2-buss. If it sounds good at 85dB on my monitors I couldn't really care less how I'm doing for dBFS.

The problem with mixing using the meters is that your DAW's meters are sensitive only to peak voltage, whereas our ears are more sensitive to RMS voltage. SO, what that means is that just because you're peaking your meters doesn't mean it sounds LOUD. Make your mix sound big and full, monitor at 85dB, set your gain stages properly (no red lights!) and don't worry about anything else.

As a great mix engineer once said: mixing is all about achieving maximum illusion using minimum voltage.

 

[TheDriller]

I never A/B to other mixes when I mix, every mix is a unique puzzle and A/Bing usually takes me farther off track.

TrueFact

 

[findingthesound]

The only way to know if you're mixing at 85dB is to get an SPL meter.

When I'm mixing I don't measure anything in terms of loudness. No clip lights on any individual tracks, plugins, or on the 2-buss. If it sounds good at 85dB on my monitors I couldn't really care less how I'm doing for dBFS.

Thanks for the info! Which SPL meter do you use? What brand is good?

The problem with mixing using the meters is that your DAW's meters are sensitive only to peak voltage, whereas our ears are more sensitive to RMS voltage. SO, what that means is that just because you're peaking your meters doesn't mean it sounds LOUD. Make your mix sound big and full, monitor at 85dB, set your gain stages properly (no red lights!) and don't worry about anything else.

That's why I prefer to talk using Dynamic Range. I use a meter in my DAW that show the DR, and find it really usefull when compressing things and look at peaks, RMS. http://www.pleasurizemusic.com/
And it's free.

In fact, I don't know how loud 85dB is but I generally prefer to work at low volume, to work more and without being tired by the volume of the sound.
But I will give a try to your method.

As a great mix engineer once said: mixing is all about achieving maximum illusion using minimum voltage.

I've already read that smoewhere...mixing with your mind? Maybe I should re-read it now!

 

[findingthesound]

Nobody knows where I can find an exemple of a good metal mix (unmastered) with seperated instruments?

 

[ahjteam]

Nobody knows where I can find an exemple of a good metal mix (unmastered) with seperated instruments?

yes, google. keyword is "multitracks"

 

[CraigTNelson]

In fact, I don't know how loud 85dB is but I generally prefer to work at low volume, to work more and without being tired by the volume of the sound.
But I will give a try to your method.



I've already read that smoewhere...mixing with your mind? Maybe I should re-read it now!

Just go to Radio Shack and buy one, if it's off by a dB or so it's not a huge deal so just buy one that looks like it won't break if it falls off your desk (it will).

Yeah, it was a Stav quote.

i said 85 dB for a reason, there's a little thing called the Fletcher-Munson curve. You'll get a "feel" for 85 dB after a few months and you can just use the SPL meter every once in a while to keep tired ears in check (tired ears love volume and compression - mine also love chorusing).

So, you work mostly at 85 db so you're actually hearing what you're mixing, occasionally doing a low-level check to make sure the bass doesn't totally diappear.

And TAKE BREAKS every 45 minutes or so when you mix. At least 5-minute breaks. It's the only way to keep your ears fresh over an 8-hour workday (although I can't remember the time I had one that short.. hahaha)

 

[findingthesound]

yes, google. keyword is "multitracks"

Just go to Radio Shack and buy one, if it's off by a dB or so it's not a huge deal so just buy one that looks like it won't break if it falls off your desk (it will).

Yeah, it was a Stav quote.

i said 85 dB for a reason, there's a little thing called the Fletcher-Munson curve. You'll get a "feel" for 85 dB after a few months and you can just use the SPL meter every once in a while to keep tired ears in check (tired ears love volume and compression - mine also love chorusing).

So, you work mostly at 85 db so you're actually hearing what you're mixing, occasionally doing a low-level check to make sure the bass doesn't totally diappear.

And TAKE BREAKS every 45 minutes or so when you mix. At least 5-minute breaks. It's the only way to keep your ears fresh over an 8-hour workday (although I can't remember the time I had one that short.. hahaha)

Thank you a lot for your help!

I'll post a part of a song as soon as it's mix, let's say at the end of this month if I have the time.

 

[Force666]

There's supposed to be a good APP for Iphone that is a spl meter. It got pretty good reviews in tapeop a couple months back. Just in case you "I" heads were looking for something.

 

[Trevoire520]

Best piece of advice I can give you, for this, and for everything you will ever do in audio:

STOP THINKING ABOUT THE NUMBERS AND START USING YOUR EARS AND LISTENING!

There are no "industry standard" levels and dynamic ranges for instruments. It's quite simple. If you can't hear the drums properly, turn them up. If they're too loud and overpowering everything else, turn them down. Simples!

If something has too big a dynamic range, to the point of being too loud at some points and too quiet at others then either:
A: Compress it
B: Automate it's volume.

Typically screaming vocals in metal have lots of compression going on to give them an aggressive character. Your best bet is to mess with a compressor on a vocal track and see what it does to the sound. Learn how the attack and release times effect the sound. Listen to the difference between one compressor doing -12db gain reduction and 3 compressors doing -4db gain reduction.

As far as monitoring goes, you don't have to necessarily monitor at 85db. Although our ears are most flat at this level I think most people on the forum would agree that it is more important to simply check your mix at high and low volumes. I tend to find that if the mix sounds good at low volumes then it'll sound great at high volumes. Where as it doesn't always work the other way around.

Also acoustic treatment is very important, mixing is very hard when what's coming out of your monitors isn't accurate, so if you haven't already then do some research and treat your room. This is the best money you will ever spend on gear of any kind. After all, what use is an API compressor if you can't properly hear what it's doing to your sound?

Dont be too afraid of heavy compression and limiting by the way, it's pretty common for most elements of a mix (other than guitars/synths) to be squashed somewhere along the line. This goes for vocals, drums and bass.

So If the voice before mastering is on top of the snare and kick, she will be distorted by the limiter.

Not necessarily. The human ear hears loudness via average levels, not peak. So even though your vocals seem louder than your drums if you record your mix onto an audio track you will see peaks above the average level of the track. These are the drum transients and can normally be soft clipped or limited with little impact on the drum sound. And it's these that will tend to effect your master compressor/limiter first.

 

[CraigTNelson]

As far as monitoring goes, you don't have to necessarily monitor at 85db. Although our ears are most flat at this level I think most people on the forum would agree that it is more important to simply check your mix at high and low volumes. I tend to find that if the mix sounds good at low volumes then it'll sound great at high volumes. Where as it doesn't always work the other way around.

Also acoustic treatment is very important, mixing is very hard when what's coming out of your monitors isn't accurate.

You absolutely DO need to mix at 85dB. If you want to mix while knowing full-well what you're hearing out of the speakers isn't what's coming out of your DAW, feel free to monitor lower or higher. If you can mix at lower levels, all the more power to you, but I wouldn't ever suggest it to someone as it's a dangerous thing to do.

That said, of course you should check your mix every once in a while at a low volume, or through another set of speakers. I always reference on these 10$ sony portable speakers I got at a drugstore at some point in the mix (for the worst-case scenario test).

 

[mephisto_loh]

Also acoustic treatment is very important, mixing is very hard when what's coming out of your monitors isn't accurate, so if you haven't already then do some research and treat your room. This is the best money you will ever spend on gear of any kind. After all, what use is an API compressor if you can't properly hear what it's doing to your sound?

+1 to that Trevoire


80% of the mixing problems are due to the room. Search about room modes and acoustic treatment. BUT you have to be wise because sometimes the companies sells you stuff telling you it's going to solve your problems. the true is that they minimize the problem but it won't dissapear at all.
We as AE will have to deal with the modes, resonances and be aware that they are going to be there no matter how. The best advice i give tell you is that most of the mixing problems are going "dissapear" by a good position of the monitors in the room. There is a program that is the Shit! it's called "Room Optimizer" and it calculates the better position of your speakers and the listener in a certain room (you input the size of the room). I hope is useful for you guys. i have more information if you want to deep in this topic.

 

[John_C]

You absolutely DO need to mix at 85dB. If you want to mix while knowing full-well what you're hearing out of the speakers isn't what's coming out of your DAW, feel free to monitor lower or higher. If you can mix at lower levels, all the more power to you, but I wouldn't ever suggest it to someone as it's a dangerous thing to do.

Ok so your ears perform best at 85db, or at least they have the flattest freq. response there (no mention of distortion levels etc)
What about the speakers themselves, they operate differently at different volumes, what about resonances in the room and so on....

It's not a route anyone wants to go down, so how about everyone just mixes at a level they feel comfortable with and regularly checks how the mix performs at different levels?