Hi, ive long neglected to learn the basics on the subject so ive got some questions.
Well my main interest is what to do per sound by analyzing the graphical spectrum.
The other day i was analyzing my mix per channel and was filtering out any clearly visible db peaks within the frequency spectrum, is this something that u should always do or only if a peak is above a certain amount of dbs? ive read something about not equalizing more then 3dbs per frequency spectrum on the final mix but does this imply i should keep the db difference per channel preferrably within the 3 db as well?
Are there any guidelines on the overall curve per sound or is a balanced curve what u always want no matter the sound? I understand bass sounds should have more levels in the low spectrum and hihats in the high spectrum, but is there some other guideline besides that? Apart from the cutoff on both sides (if any) what is recommended to do when u only look at the graphical curve and analyze db differences. is there a maximum db level change u want to maintain, regardless if its a peak or a wider range?
Any help on the matter would be very much appreciated.

 

Analyzers and meters are powerful tools, but you still have to use your ears and your own judgement. General guidelines are just that, "in general". I believe the less EQ you use the better. For recording live instruments, good mic placement will save trying to keep track of lots of knobs and faders. For any kind of recording, if you make every track have the same pristine EQ, things start to clash and cancel when you try to cram it all into two channel mix.
One of the best things about digital mixing is you have filtering on every channel. I like to use that before I start to use equalizers. There is no point in being in some frequency for some tracks, so you filter all the frequencies that you can't hear anyway.

 

Last edited: Jun 12, 2017

Love your avatar, man!

This reflects my experiences. I started my eq journey soloing each track, and making it sound the best I could. End result, everything sounds great, just not together, and depth wasn't ideal (everything was "in your face"). I've had to learn to compromise eqing in certain tracks to let other tracks punch through when needed, eg. kick drum and bass guitar, rhythm guitars and vocals, etc. as well as filter the high/low end to simulate the effect of distance (coupled with compression and reverbs).
I also read where it's a bad idea to solo a track and "sweep around" to find out what you want to cut and boost. Reasoning is that your ears adjust themselves as you are listening and compensate somewhat. Instead, learn what frequencies do what (eg. stick, snares, bang, boxy for snare drums) for the track you are working on, adjust, then listen to see if your blind adjustment did what expected. This trains your ears how to achieve what you are looking for and associates it with the specific frequency you are boosting/cutting.
Many have said only cut, not to boost. Watching CLA videos, he appears to do either. I think boosting is OK as long as the eq of choice that you're using is "musical" when doing it.

 

ok I use analysis tools, a lot and my ears too, by analysis you can see many peaks as you said. If you start applying subtractive eq, especially with narrow Q to all those peaks you ll end up ruining the sound for sure. I ve seen a tutorial about kick equing where the guy creates 4-5 narrow Q poles with a -9 dB each and tells to go on like that for the entire spectrum. Maybe he doesn t really like the... kick sound !
What I do is creating a pole with a Q of 15 , + 21 dB to start moving it to analyze the signal, from say 50 Hz upwards
when I hear strong resonaces, I know where to cut. You have to know what the main GOOD frequencies are, cause a kick would resonate at around 50 to 100 Hz but that s all good. If it resonates at 200 it s not good for me. That s the roundness it has to be cut. Same at say 300 to 600 Hz, cut. Never more than 3 parametric cuts (add low and hi cuts if needed)
Now in the Deep low end I can use narrow Q equalization, till a Q of 15, normally 8.
from 150 Hz to the rest of the spectrum my cuts are more broad, from Q of 1.6 to 4.
I don t use a single eq to cut. I use a pre comp eq, and a post comp eq. If I need a -8 dB cut, it will be -4 on the first eq and -4 dB on the second one.
I use very broad low and hi cuts (6 to 12 dB/oct).
I can boost more than 3 dB as you ve stated, rarely more than 5-7 dB for a parametric eq, never more than 2 dB for high end shelvings.
ok that s all comes on my mind by now, cheers

 

Last edited: Jun 12, 2017

I agree with the two other scenario descriptions:
micing is the key, especially on guitars and bass guitars. Recently I only use low cuts and highs shelvings.
I only apply 2-3 dB of compression to the room mic, from the recording mic pre
drums need equing for me, even drastic, on SSL simulations kicks presets have a + 7 to + 9 dB on 50 Hz that s cause you need it. I need it. I cannot live without it. lol
same for snares...
but I never boost more than 2 dB on high end shelvings and never pre comp

when I start a mix I start from vocals (if there are vocals)
I try to make the best natural sounding vocals, usually eq-comp-limiter-eq-Waves Vocal rider
then I add a short room reverb with a bpm wise pre delay (calculating it)
then kick, snare, hats, overheads then bass
then I check everything with everything, lows mids highs
then I go ahead with the rest of the mix and adjust all the instruments
if my main poles on vocals are at say as istance, 230 and 5300 Hz, I d pay attention to other conflicting sounds that can have resonances in that part of the spectrum. So I may cut one or two dB on violins, at 5.3 kHz to make space for the vocals, or cut around 230 Hz on bass, kick and toms... and I would stay away from 230 when boosting guitars lows... things like that, continuously checking and double checking.

 

the best tip for equalizing is use your ears not your eyes. EQ how you like it and how it sounds the best for you. If you run into Problems use your eyes. And analyze the Problem.

There are no rules for EQing. It surly depends what you are equing. Is it a well recorded piano or a or a super distorted synth sound from your vsti. You dont need much eq for a real intrument that is well recorded.

Yes there is differnce if you eq a single track or a bus. You should only do small amounts on the bus i really rarely do more than 2 db on on a bus. Cause it adds up. Its like adding the db on every track.

You should mainly use EQ as a tool to form your vision. Not as a tool to look for flaws. Strange peeking frequencies that you dont hear in the context of your song or maybe fit into the song. Use your ears is the better way.

 

Last edited by a moderator: Jun 12, 2017

@Adamdog , Yep, guitars shouldn't need much if you've got them mic'd good. Hard to pre visualize what you need for the final mix sometimes, but you can usually get it close. Then of course there's the low end/high end band passes that you "should" do. My mixes really opened up after I learned to do this.
Same here with the kick EQ'ing, I do a boost at 45hz and a cut around 55hz. Really makes that thump solid to my ears. Guilty of too much thump before I got a set of Auralux pads to set the monitors on. I'm usually running a triggered track (which I usually don't do much to) and a mic'd track. The mic'd track normally gets this treatment, and maybe a little boost around the beater frequency if I don't hear enough "click" in the mix. Maybe a cut around 500hz if it's too boxy sounding. Normally it's a 22" DW kick or a 26" Ludwig, so pretty robust on the low end to begin with.
That's the other thing I wanted to add. EQ'ing works best if your monitoring environment is a.) good, or b.) you know your room. If you are trying to EQ something that's either resonating in your room (you'll cut too drastically) or phase cancelling (you'll boost too drastically) can cause problems when you compare on other systems.

 

your low end boost and cut looks like a Pultec-like style (good!)
I always take a look at around 200-250 Hz on kick and 180-200 Hz (also 320-350) on bass for eventual (100%) problems
I can do some mid range equing on guitars, sure, in complex mixes, always tonal, gentle, I like Neveish virtual simulations for that, like PSP McQ or 1081 like plugs
I manly mix on PMC TB2SA speakers, Event TR8, Yamaha MSP5, cheap Philips hi fi yes treated room

 

"I believe the less EQ you use the better"

Sorry, but this statement doesnt make any sense, lol, you use the EQ as much as you need.

 

Because the more EQ's you have competing with each other the more phase distortion you have to deal with. And the trick is to keep your needs minimal.

 

Sure, but it still makes no sense. You use it when needed to get from a to b, you werent talking about number of equalizers, but the equalization as a process itself, that is how you wrote it.
I keep my needs realistic, not minimal, no more no less.

 

I wholeheartedly agree.

This topic is rather a theoretical one than a musical one. Soon it's going to be 100 years since the EQ was invented. Our entire lives we were listening to equalized audio production hence we were exposed to phase distortion on a daily basis. Our ears simply don't know other way for any musical performace involving electronic equipment. Unless we work on correlated signals (eq-ing various mics on a live drum kit for instance) I see no reason to mind phase distortion. I will EQ that instrument (so basically placing a phase distortion at a certain frequency - and this distortion will either enhace my mix or will get cancelled out by something else) until it sounds good in the mix, I don't care if it the EQd version sounds different in isolation from the non EQd one.

 

I definitely agree with the "there is no solid rules", "use your ears" and all but, I think gain-staging is actually pretty close to be a complete solid rule to follow. Every boosts or cuts needs to be compensated with EQ gain, I don't see many case where it doesn't apply (strictly talking about mixing, not sound designing EQs ofc), if I would have to force myself to apply ONE rule in mixing EQs, it would be this one.

Phase distortion or phase shifting from too much EQ/HPF/LPF isn't only hurting the musical part of the mix, it's not helping the mix too I believe! Sure there is tons of huge releases that has tons of phase shifting, so? I wouldn't stop minding about this just for this reason. There is tons of well-known tracks with ISPs, phase shifting, hardclipped output, destroyed dynamic range, weird limiting at -5dbfs, and the list goes on. This doesn't mean it's okay to let your tracks have these! Sure it won't stop a great track, a great production and a great mix to sounds good, but minding about all these will make it sound even better on highest end listening gear.

 

Last edited: Jun 13, 2017

If you're starting out I'd suggest trying to make a full mix without eq'ing much at first. Once the pieces of the track are together you'll have an idea of what the main focal points of each section are, which will be the instruments you want to boost more to make the center of attention. A lot of times you can get away with just using the volume knob to mix things, then if you notice mud or too much buildup in certain areas notch out frequencies from the background sounds so that the main instruments are cutting through. Keep in mind its really easy to get into the bad habit of over eq'ing things initially. Phase is an issue, especially with lower frequencies that are spread out so it helps to keep checking your mix in mono on the master channel.

 

Last edited by a moderator: Jun 13, 2017

Some years ago i was thinking just like you I mean i was following the so called expert opinion and shit like that But seriously at the end its you and your ears only no one on earth can teach you better EQ than your ears.
So Follow this tip and you will good to go
1. set your fav.EQ in which you love to work (Do not beat around the bush with saveral different EQ's) So set your EQ Mine is Parametric EQ2 for all mixer channels and a fabfilter PRO 2 EQ in master channel

2.Always give priority to your ears first and then to the visuals because every mix i repeat every mix required different EQ so you have to always apply different EQ every time based upon your mix and your ears.

 

I'm glad you mentioned this VERY important step. I'm constantly setting my gain staging anytime any adjustment is made that would affect level. After compression tweaking as well. I check from the plugin I'm adjusting back out to the end of the chain. Tedious work, but worth the effort!

 

Definitely worth yeah! Like you said it's not only on EQ,but also on comps, saturators, imagers... pretty much every plugin actually

 

wow im sorry i forgot all about posting here lol (i was high). So a lot of replys to read, lets start doing that first

 

Forget about EQ man their FFT is sh*t, just get Flux Pure Analyzer (http://www.fluxhome.com/products/analyzer_modules/pas_spectrum) and let the trip go

 

Well my main problem with EQ is actually that i do not really trust my ears. Ive seem to have deafen some on the high frequencies (thats why usually my mixes are was too bright. Also im not really at liberty to listen music at a proper volume at home to feel comfortable with my mix and I sure as hell dont trust my headphones anymore after some of my mixes i did on them, posted on soundcloud, played the next day over monitors and scared me straight how different and crappy it sounded... (ofcourse pulled it inmediatly off soundcloud again)
While they are actually quite decent headphones (krk 8400) they seem to work like mix anti depressants because it makes crappy mixes sound like they are half decent. Or i just have terrible monitors ofcourse heh. No but really headphones suck for mastering/mixing.
Now there has passed quite some time between when i started this topic and today, so in the meantime I have been trying to learn more about EQ mainly by just trying things.

Well ive discovered in fact that i used way too much EQ and way too extreme settings. If i open an old project now im like wtf was i thinking lol. Basicly i removed most of those EQ settings and the mix would be much better. Also im trying to rather take something off then to add eq which seem to be working better as well. Im only really adding any EQ if im going for a specific mid sweep EQ effect but apart from that i try not to do it anymore

Now for the substractive equalising and removing the peaks, yes i did indeed notice it making my sounds stale and boring so i really only do it when certain frequencies start screaming in your ear and are getting painfull even on normal volumes.

What also is a big thing why i dont always trust my ears is when u have heared your track like a 1000 times during the process already, its hard to still be objective about it. There comes a time u heard it so many times it becomes normal the way it is and u would choose it over a much better mix because any change just sounds weird. Well i know i get this problem at some point when i render a project unfinished dont work on it a while, but listen to the render on a regular base and then continue the project again. Its like a mixtape with a lot of errors and out of sync records that you listen over and over again untill the point they need to be there and it would be disruptive if suddenly it would be a near flawless mixtape and every part would be in good sync.

Now my main goal for asking about EQ was actually the fact my mixes were not loud enough and while EQ played some part in there it was mainly a dynamics and volume problem in my case actually. Well the low cuts on sounds that dont really need it helps quite a bit, and for the bass and kickdrum sound i filter below the 40hz.
But ive really been able to boost the volume of some old mixes with help from the glue compressor and the limitless by DMG. Its better to avoid a limiter on the master ive heard but when theres no other option and all you have is a wav file that limitless sure does one hell of a job. Also ive been using izotope RX advanced and love it so much i no longer feel the need to use soundforge anymore. It has a nice EQ match feature as well like the good old freefilter by steinberg which can be a helpfull tool when u have a teribble sounding mix.

Oh yeah ive been using eq's from DMG audio like equality and equilibrium and like the way they sound.

What still is difficukt for me is maintaining a clean mix while using quite a lot of distortion on individual sounds. But i probably know the solution for that, use less distoirtion hehe. Well ive heard artists that keep their distortion nice and crisp sounding and in my case it often clouds my mix a great deal. But when i try to clean the mix a little bit i often loose a great deal of rawness.

Anyway thanks all for replying, i will read it some more in case i missed some things earlier.
Very appreciated and my appologies for the late reply
yo