August 29, 2012, by Matthew Weiss
All recorded sound is equalized…
Before it runs through any EQ – if you record a voice, with a mic, into a preamp, and play it out of your loudspeakers – you have a sound that’s already been EQ’d six, arguably seven times.
—The way the vocalist shapes their mouth, throat, and diaphragm will effect the frequency curve of the voice.
—The way the room responds will effect the way the tones are perceived.
—The microphone will naturally dip and accentuate certain frequencies – as will the preamp.
—The loudspeakers will not be perfectly flat (though some are pretty freakin’ close), and the room in turn where playback occurs will also effect the way the sound is heard.
—Also, you, as the engineer have a unique brain that interprets all of that information – some are more sensitive to high end, others more sensitive to low end.
—Compression in the tracking phase? Yep, it’s going to effect the tone.
The point of this is that all sound has a tone shape. The only question when reaching for an EQ is: are the tones working?
And that’s what EQ is: Tone Shaping.
But it seems that even though the sound source (voice in this example) has been EQ’d several times over – people automatically reach for another EQ. The temptation might sometimes be for the engineer to make the sound hers/his.
There’s a pride in knowing that one has crafted the sound. I have to stress that this mindset is ultimately detrimental – the quest and pride should be in making the necessary moves, and not EQ’ing for the sake of EQ’ing.
Why EQ Something?
Masking – So this is every engineers’ favorite idea. What is masking? When a tone from one instrument hides the audibility of the tone of another instrument.
As engineers we commonly mistake masking for being a “bad” thing. I’d like to offer the idea that masking is just a thing – and that whether it’s good or bad is completely context dependent.
Masking to some is layering to another. Clutter can just as easily be wall-of-sound. In an orchestra, instruments walk all over each other and this is done purposefully in order to create a sea of sound (when done harmoniously), or chaos and tension.
Sometimes, a masking tone can provide added energy in the spectral area of the element it would mask, if the level is set properly. For example, cymbals will assuredly mask the high end of a vocal if too loud. But you wouldn’t turn them way up and then EQ out the treble (well, you might, but probably not) – because then you get loud, dark cymbals.
Instead, you mix them low enough so that they don’t step on the vocal’s presence – they just provide more energy behind that domain. Masking is only an issue when you want two things to be of equal role importance that share significant content in the same frequency area.
And then, removing as little as possible from one of the elements will provide a fuller sound which may be preferable to a bigger cut, which will lead to a more open sound.
Resonance and build up removal – The vocalist may have been too close to the mic. The room may have a strange ring. The snare may be un-damped and the player might play with too little tip.
All these things can lead to unwanted resonances or frequency build up. Now again, these are just things – not necessarily bad. A sine wave synth is basically just a moving resonance, and that sound dominated West Coast Hip Hop in the early 90s.
Proximity effect can also be a useful way to fake a little weight to an otherwise thin sound source. So pick and choose what really needs to go.
Generally, if there’s unwanted proximity effect or resonance, something sounds clouded or peaky – and with decent monitoring, these things are not too hard to find. Ideas such as “openness” and “thickness” should come into play when adding or subtracting these characteristics.
Highlighting or suppressing tones. This is the grey area for engineers. There’s a million different ways to highlight or suppress. And which ones to pick and choose become very much negotiable.
What makes a tone “good” vs. “bad” is truly in the ear of the beholder.
Tips For Using EQ
Cut, don’t boost – OK, well this is sort of like the old saying “feed a cold, starve a fever.” It harkens back to the yester years of exclusively working in the analog realm. Boosting was a fundamentally different practice than cutting. In the software world, this is not really as true.
However, there is an important psychological practice here. Everything sounds better when it’s louder. So boosting isn’t always the most honest way of enhancing the sound.
If,on the other hand you cut something – and that improves the quality of the sound – you know you’re doing it right.
Do what needs to be done – Nothing more and nothing less. Doing something for the sake of doing something is death to your mix. At the same time – if 30 dB of high end sounds good, DO IT.
Acoustic drums in particular often take a load of EQ in modern music – adding 10 dB of low end, cutting 6-10 dB of mid-range, and adding 15 dB of treble sounds like overkill to a kick drum – but if that’s what it takes so be it.
Different EQs sound different – So what’s up with all the EQs all over the place in big time mixer’s productions? Well, to quote Dave Pensado from his Sound-On-Sound interview, “I think you get more by not trying to get everything out of one EQ. Every EQ is good for certain things.”
This is quite true – some EQs are better for precise resonance removal, some are better for broad brush tone moves. In the analog world particularly, some EQs excel at high end sculpting – some are stronger in the lows – and some just have nice gain stages.
Don’t seek and destroy – Another common practice is to sweep a heavily boosted narrow notch and find “problem frequencies.” In my experience, every frequency will sound problematic this way. For some people, this technique works, but I’ve never had much luck.
My feeling is to take an experienced guess as to what frequency you are hearing, and check where you think something is going on with a moderate boost and a tight Q. Maybe move that back and forth around the area you think is troubling to get a precise area to effect.
Q is IMPORTANT – Q is just as important as the amount cut or boost being applied. A wider Q often yields a more transparent sound, but less focused as well. The idea behind both the amount of eq and the width, is to do just enough so that you barely notice it – if at all.
If bringing out a tone, go as wide as possible before you start effecting things you don’t want to effect. If removing an ugly sound, go only as wide as necessary to get the sound, and leave the rest as in tact as possible.
Remember – every sound already has an EQ curve – and has been EQ’d many times over. So there’s never any reason to EQ for the sake of EQ’ing. The key is to figure out exactly what is needed for what you are trying to accomplish, and do it.
Don’t be EQ happy, but don’t be shy of cranking the knobs to extremes either.
Matthew Weiss is the head engineer for Studio E, located in Philadelphia. Recent credits include Ronnie Spector, Uri Caine, Royce Da 5’9” and Philadelphia Slick.
Be sure to visit the Pro Audio Files for more great recording content. To comment or ask questions about this article go here.