077 - Parametric vs. Graphic Eq
Both Parametric and Graphic EQ can dramatically alter a sound, but each tends to work best in specific situations.
Written by Scott Adamson
EQ, short for equalization, is a powerful tool that helps us shape the tone of a sound and attenuate or eliminate problematic frequencies. It exists all over the signal chain. We use EQ on inputs at the channel strip, on output busses, and later on in system processing once sound has left the console.
Here we’ll discuss two important kinds of EQ used all the time in live sound — parametric and graphic. Throughout this piece, I mention “boosting” and “cutting” the gain. You can do both, however, in live sound, it’s far more common to cut than it is to boost.
First, let’s review a fundamental term. When we discuss a particular portion of the frequency spectrum, we call it a band. For example, we associate low-mid frequencies with roughly 100Hz to 500Hz. That frequency range is a band! When we use EQ, we manipulate different frequency bands.
With parametric EQ, there are often three or four different bands. This might look like one knob for the high frequencies, one for the mids and one for the lows. This is called a 3-band EQ. Many pro consoles will have a 4-band EQ for lows, low-mids, high-mids and high frequencies. 7-band EQs are also fairly common — and very powerful.
But, of course, every console is slightly different and features will vary depending on the quality and whether it's analog or digital.
On each frequency band of a parametric EQ, there are three important parameters: gain, frequency and bandwidth/Q. Gain determines how much is being boosted or cut. Frequency, just as it sounds, determines which frequencies are being boosted or cut. Bandwidth, also called Q, adjusts how wide or narrow of a frequency band you affect.
Let’s break each of these down. With the gain knob, we’re generally able to boost or cut up to 12dB on each band. This is a big move that can drastically change a sound. There’s no right answer for how much gain to boost or cut, but starting out with 3-6dB and re-evaluating from there can set you on the right path. (You may realize it’s too much, in which case, pull back a bit!) When no frequencies are boosted or cut, this is called a flat EQ.
You can also choose which frequencies to boost or cut. Sometimes the frequency assigned to a particular band is fixed, meaning you can’t change it. For instance, on some consoles the low frequency knob may be fixed to 80Hz. EQ bands that let you adjust the frequency are called sweepable. On digital consoles, we work almost exclusively with sweepable bands.
As we’re boosting or cutting gain with EQ, we not only have control over which frequency we’re affecting, but how wide or narrow the band is. This is called bandwidth. It can also be called Q.
Get real-world live sound mixing tips straight to your inbox.
Like all settings on the EQ, bandwidth can have a big impact and can make the EQ serve different purposes. An especially wide bandwidth, meaning a range of frequencies will be affected, will be more useful for shaping the general tone of a signal. If an acoustic guitar needs a bit of overall brightness, we could try using a very wide bandwidth and boost 3dB somewhere around 5kHz.
Say that same guitar has some low end feedback through the main speakers. With feedback, we're usually dealing with a very specific frequency. You could use a high pass filter to cut out all the low frequencies, but that may alter the signal more than you intend. In this instance, we could use a very narrow EQ band. This makes it work more like a graphic EQ, which has narrow bands and is good for controlling feedback. Since we're looking for a specific feedback frequency, you could make the cut and sweep the frequencies until you find the right placement that kills the feedback.
As I mentioned, another commonly used type of EQ is graphic EQ, sometimes abbreviated GEQ. Pictured above is Ashly’s MQX-2310.
Compared to the parametric EQ’`s 3-, 4- or 7-bands, the graphic EQ divides the entire frequency spectrum in 31 fixed, narrow, ⅓ octave bands that you boost or cut as needed (some graphic EQs use 15 bands instead). This makes it a great tool for cutting problematic frequencies without changing the overall sound too much.
It’s common to see a graphic EQ inserted on a system’s main outputs, tweaking the overall main mix to better suit the particular room, and on monitor mixes to help quell feedback.
When we start to hear feedback in monitor wedges, we have two options. Assuming we know which microphone is feeding back, we could simply turn it down in the wedge, but that isn't always a good option. What if the artist still needs it to be just as loud? A graphic EQ would come in handy here.
The frequencies you need to notch out will change depending on the mic, the wedge and the room you’re in. Since wedges sometimes need to be really loud, you might have to do quite a bit of cutting along the graph.
As you mix more and more shows, you find that certain mics or certain wedges will be problematic at the same frequencies all the time. Engineers with a lot of experience know these frequencies well. Every graphic EQ uses the same 31 frequencies, no matter what the manufacturer, so it’s standard to ask for a cut at, say 630Hz, or 3.15k.
Knowing exactly what these frequencies sound like is an invaluable skill. If you have access to a graphic EQ, play music through it and start boosting and cutting various frequencies. Run a mic through it and hear what those frequencies sound like on your voice. This is a great way to train your ear.
Overall, there aren’t hard and fast rules for EQ. You can certainly use a parametric EQ on a wedge and you can also use a graphic EQ on any particular input. Regardless of which tools you choose, don’t get wrapped up in what your EQ curves look like. Let your ears guide the way.
