Filters: The Wow Factor

Filters: The Wow Factor

28 November, 2007 | 12.59PM

• Written by: francispreve
• Filed under: Music News: filters, guide to synthesis

Once your oscillators are configured to generate the basis for your sound, it’s time to shape the frequency content via filters.

At first glance, filters appear to behave like an equalizer, but there are subtle differences that give them a very different sonic character.

The most significant distinction is the fact that filters subtract frequency content, whereas EQs can raise or lower the volume of entire frequency ranges.

That doesn’t mean filters are less powerful than EQs. Far from it! Different tools for different jobs, that’s all.

In this tutorial, we’ll cover the four main types of filter and how each mode affects the sound generated by the oscillators.

Filter Types

The four basic types of filter are: low-pass, high-pass, band-pass and band-reject (also referred to as “notch”).

Each has a specific character and behavior that is useful for different types of sounds and effects.

Many modern synthesizers include a multi-mode filter, which can operate in any of these modes.

In addition, some filters also include options for toggling the rolloff slope of the filter between 12 dB per octave and 24 dB per octave.

In simple terms, a steeper slope (24 db/oct) creates a more pronounced effect.

Other implementations chain two filters in series.

Propellerhead Reason’s Subtractor, Malstrom and Thor all provide multiple filter paths, as does Ableton Live’s new Analog synth.

In these instances, the output of the first filter feeds the input of a second filter for more complex signal manipulation.

Low-pass

The image below, taken from Ableton Live’s Auto Filter device, shows a low-pass filter with a slight increase in resonance (sometimes called Q or emphasis).

The horizontal axis represents the frequency range from low to high.

The vertical axis represents the amplitude — or volume — of the curve.

A low-pass filter passes the frequencies below the cutoff frequency and attenuates (or reduces the volume of) the frequencies above it.

Lowering the filter frequency makes the resulting output duller or more muted.

Raising it opens up the sound, allowing the full range of upper frequencies to pass.

If you’ve ever heard a breakdown that drops down to a low rumble (like the sound of music through the wall of an apartment or underwater) then gradually gets brighter and clearer, then you’ve heard a low-pass filter.

Most older analog synthesizers relied exclusively on low-pass filters as their sole filter type, so if you’re wondering how classic 1980s era synthpop sounds were created, you should start your sonic explorations using low-pass mode.

High-pass

In this image, the curve shows the frequency response of a high-pass filter.

High-pass filters are essentially the opposite of the low-pass type.

They work by passing the frequencies above the cutoff frequency and attenuate the frequencies below.

This can make the sound thin and buzzy.

Some older analog synths — notably those from Roland, like the Juno 6/60/106 and Jupiter 8 series — included a simple high-pass filter in conjunction with the low-pass module.

High-pass filters are often used in remixes as a technique for introducing new patterns by starting them out trebly and tiny, then slowly making the sound larger and fuller by allowing more low frequency content through.

Band-pass

While the image below may look like an EQ, the band-pass filter mode works in a somewhat different fashion.

Band-pass filters allow a specified range of frequencies to pass unaffected, while attenuating all frequencies on either side of that range.

Sharing characteristics of both high- and low-pass filters, band-pass filters are useful for sculpting signal content in a manner not unlike that of the mid-band EQ found on a DJ mixer.

On high-pass and low-pass filters, the frequency knob determines the cutoff frequency at which the filter begins attenuation.

On band-pass and band-reject filters, this parameter determines the center frequency of the filter’s operation.

Band-pass filters often sound quite good when the frequency is swept (continuously raised and/or lowered) either by automation, envelope or LFO.

The result is quite similar to a guitarist’s wah-wah pedal — and with good reason, since band-pass filters are commonly used in this type of effect pedal.

Band-reject (Notch)

The image below shows how a band-reject filter’s response is a mirror image to that of the band-pass.

A notch filter functions in an inverse manner to band-pass filtering.

It rejects the frequencies within a specified range, while allowing all others to pass.

The sound of a notch filter is quite subtle in practice, bearing a slight resemblance to a phaser when the center frequency is swept via an envelope or LFO.

Another good use for notch filters lies in the removal of problematic frequencies within a sample, much like a surgical grade EQ.

Next up…

In the next installment, we’ll go deep on the specific parameters that govern the behavior of filters.

• (5) Comments
• (7091) Views