Equalisation. Many times I’ve found myself at war with the word. A mix can go from sounding natural to contorted very quickly. Two things I’ve learned about Eq if you’re trying to make things sound natural – balance, and less is more.
Different types of Eq
Parametric: This type of Eq is usually made up of different “bands” or frequency ranges, each band having 3 different controls: frequency, gain and Q. Frequency is self-explantatory (measured in Hz or kHz). Gain can either be a cut or boost, measured in dB. The Q parameter controls the sharpness of the frequency curve. Parametric can be used surgically to remove certain frequencies with a sharp Q setting, or more “naturally” with wider Q settings which is generally more favourable for use on whole mixes.
Graphic Eq: Graphic Eq is made up of different bands also, but in this case each band has a fixed Q setting and the frequencies of the bands are fixed. There are generally many more bands than in Parametric Eq, which can make it quicker to tweak a particular frequency range just by adjusting the gains on the relevant bands.
Linear phase Eq: Due to the nature of sound frequencies, when you boost or cut a certain frequency band there is some delay added to it. This can cause “phase shifts” between the bands which may or may not be audible. What linear phase Eq does is compensate for the micro-delays, causing all frequency bands to arrive at the output “linearly” or at the same time. This can be useful in some applications such as mastering where you might want to avoid artefacts caused by phase shifts, but it can introduce some other artefacts such as “pre-ringing”
The golden rule of Eq in mixing is to ensure each instrument has its own space. This means to try as much as possible to avoid clutter and enhance clarity. Many producers these days take this to the extreme, filtering out big chunks of frequencies.
My approach is to be less severe with Eq. As I like things to sound as natural as possible, I will tend to use a “high shelf” for instance, cutting by a few dBs as opposed to the violent use of a low or high pass filter, or extreme cutting/boosting. Don’t forget there are frequencies in there which support other frequencies harmonically, so it’s important to achieve a balance.
For some reason most people perceive the range of 400Hz to 800Hz to be not so pleasing to the ear, or “roomy” sounding. Therefore it’s very common to scoop out these frequencies in mixing and mastering. I’d say again, less is more. That range can contribute a lot to a sound, so it’s important to get the balance right.
A note on different types of frequency curves, there are two main types; bells and shelves. Bells are ideal for cutting/boosting a certain range, and shelves are good if you want to cut or boost everything from or before a certain point. One thing I’d say about shelves is be careful, they often bring up unwanted sub or ultra high frequencies which aren’t too pleasing to the ear.