Technique: Effects Basics
By Anderton |
Profiles of popular effects types - what they are, their main controls, annoyng habits, and hot tips
by Craig Anderton
Effects are to recording or live performance as spices are to cooking—they can really enhance whatever’s already there, although a little goes a long way. Yet a lot of people aren’t really that familiar with their effects; they just dial up a preset and hope for the best.
If you understand how these boxes or plug-ins tick, you can use them much more effectively. The following roundup of common effects clues you in not only to what they are, but to their crucial parameters, annoying quirks, and some of the most popular applications.
Profile. A compressor/limiter (C/L for short) evens out dynamic range variations by amplifying soft signals to make them louder, and/or attenuating loud signals to make them softer. The result is less level difference between soft and loud signals.
How it works. Once a signal exceeds a user-settable threshold, compression occurs where increasing the input signal does not increase the output level by an equivalent amount. For example, with a compression ratio of 2:1, every additional 2 dB of input level results in only 1 dB of additional output level.
Crucial parameters. Threshold sets the level above which signals will be compressed or limited. Signal below the threshold are not processed.
Ratio selects how the output level changes in relation to the input once the input exceeds the threshold. The higher the ratio, the greater the amount of compression, and the more “squeezed” the sound. Extremely high ratios put an absolute “ceiling” on the signal, which is called limiting.
Output adds gain to offset the lower level caused by restricting the dynamic range.
Attack sets the reaction time to input level changes. A longer attack time “lets through” more of a signal’s original dynamics before the compression kicks in. For example, adding a bit of attack time retains the initial “thwack” of a kick drum.
Release or Decay determines how long it takes for the C/L to return to its normal state after the input goes under the threshold. With short release times, the C/L tracks very slight level changes, which can produce a “choppy” sound.
An Auto or Program Dependent switch, if enabled, sets the Attack and Decay times automatically and re-adjusts these settings as needed for different program material.
Annoying habits. Over-compressing results in a thin, unnatural sound, and brings up noise. Don’t add any more compression than needed. Also, controls interact—for example, changing the ratio can change the threshold.
Hot tips. When used with other effects, if possible place the compressor early in the chain so that it doesn’t bring up the noise from previous stages. • If it seems like there’s been a sudden increase in compression but you didn’t increase the compression amount, then the input signal going to the compressor may have increased. • Some music from the 60s featured a drum sound that sounded like it was “sucking” and inhaling. To create this effect, apply lots of compression with an extremely short release time.
For more information:
Profile. Distortion mimics the way an amplifier behaves when overloaded, so it’s a popular effect for guitar. However, distortion can also spice up drums, synthesizers, and even vocals.
How it works. Not all types of distortion (tube, transistor, digital, etc.) sound the same. Some devices include a tube stage or other analog distortion circuit that can be modified under computer control. Others use DSP to emulate particular types of distortion.
Most musicians prefer “soft” clipping, where the output signal becomes progressively more distorted as the input signal level increases. With hard clipping, the output signal remains undistorted up to a certain point, then becomes extremely distorted as the input increases past that point. This sounds harsher.
An undistorted signal compared to soft- and hard-clipped versions.
Crucial parameters. Sensitivity, Drive, or Input determines the amount of signal level needed for the onset of distortion. Maximum sensitivity gives the most distortion.
Output. Since distortion often adds a great deal of amplification, the output parameter trims the effect’s output level to something reasonable.
Tone controls. Some distortion effects include tone controls. Distortion adds harmonics to the signal, which increases the high frequency content; pulling back on the highs reduces shrillness, while boosting the bass gives more depth.
IK Multimedia's AmpliTube is one of many popular amp sim plug-ins.
Annoying habits. Because of their high gain, distortion boxes can generate a lot of hiss. Also, because many distortion devices are designed for guitar, it’s hard to find stereo models for mixing applications.
Hot tips. Patch a distortion unit into a mixer’s aux bus, and bring the returns back to the mixer. To add some “bite” to a channel, turn up its aux bus send to taste. • A little distortion can really increase the punch of drum and bass sounds. • Distortion can make a synthesizer sound a lot more “rock and roll.” Add some crunch to organ patches that use rotating speaker effects, or to that classic Yamaha DX7 that’s sitting around feeling neglected.
For more information:
Profile. An equalizer emphasizes (boosts) and/or de-emphasizes (cuts) certain frequencies to change a signal’s timbre. The amount of boosting or cutting is expressed in decibels (dB).
How it works. Equalizers use filter circuits that pass certain frequencies and reject others. The four most common filter types are lowpass (passes all frequencies below a certain cutoff frequency), highpass (passes frequencies above the cutoff frequency), bandpass (boosts only those frequencies around its resonant frequency, while rejecting higher and lower frequencies), and notch (all frequencies around the notch frequency are rejected, while frequencies higher and lower than the notch frequency pass through to the filter output). The range of frequencies affected by the boost or notch is called the bandwidth.
There are several types of equalizers. Shelving equalizers boost or cut a fixed amount over a range of high or low frequencies. The graphic equalizer uses multiple bandpass filters to split the audio spectrum up into a number of bands, with an individual boost/cut control for each band. A parametric equalizer is a sophisticated form of tone control. Unlike the graphic equalizer, which can boost/cut only at fixed frequencies, a parametric can boost or cut over a continuously variable range of frequencies. In addition, the bandwidth is variable, from broad to sharp. Note that there are also quasi-parametric (also called pseudo-parametric) equalizers that include frequency and boost/cut controls but no bandwidth control.
The three main parameters of a parametric equalizer, and how they relate to level and frequency.
Crucial parameters. Frequency sets the specific part of the audio spectrum where the boosting or cutting occurs.
Boost/cut determines the amount of equalization at the selected frequency.
Bandwidth, resonance, or Q. This control determines the sharpness of the boosting or cutting action. Narrow bandwidth settings affect a very small part of the audio spectrum, while broad settings process a broader range.
Annoying habits. Some equalizers don’t include bypass switches, making it difficult to compare equalized and unequalized versions of a sound. Also, some equalizers have a fixed bandwidth, which always seems too narrow or too broad for the intended application.
Hot tips. You’ll have more headroom if you cut rather than boost. For example, it’s often better to cut the midrange than boost the treble and bass. • Frequently compare the equalized and non-equalized sounds. You don’t want to get into a situation where you boost the treble a lot, which makes the bass seems thin so you boost that, which then makes the midrange seem weak so you boost that, and so on. • Always use the minimum amount of equalization necessary. Just a few dB of change can make a big difference. • Suppose you’re playing a rhythmic piano part behind a vocalist, but since the piano and voice occupy a similar frequency range, they conflict. The solution: pull back on the piano’s midrange somewhat to make room for the vocal frequencies.
For more information:
TIME DELAY: FLANGING, CHORUS, ECHO
Profile. Time delay produces effects including flanging, echo, chorusing, tapped delay, stereo simulation, and others. Some devices provide dedicated effects for each function; others simply include a general purpose time delay effect that is flexible enough to provide these different effects.
How it works. Time delay effects stuff the input signal into digital memory, then read it out a certain amount of time later. Feeding some of the output back to the input recirculates the delayed sound, thus creating a repeating echo effect. Modulation, which varies the delay time over a particular range, produces an animated kind of sound as the delay time sweeps back and forth between a maximum and minimum value.
Crucial parameters. Initial delay sets the amount of delay time. With echo, this is the time interval between the straight sound and the first echo. With flanging and chorusing, modulation occurs around this initial time delay. Some devices let you synchronize the delay time to MIDI song tempo. Another option is a tap function, where hitting a switch or button twice sets the delay time interval.
Balance, Mix, or Blend. This parameter adjusts the balance between the dry and delayed signals. Flanging typically uses an equal blend of dty and delayed signals, while chorusing uses more dry than delayed sound.
Feedback, Recirculation, or Regeneration. This parameter determines how much of the output feeds back into the input. With echo, minimum feedback gives a single echo; more feedback increases the number of echoes. With flanging, adding feedback increases the effect’s sharpness, much like increasing a filter’s resonance control.
Sweep Range, Modulation Amount, or Depth determines how much the modulation section (also called LFO, or sweep) varies the delay time. For example, a delay with a 2:1 sweep range can sweep over a 2:1 time interval (e.g., 5 ms to 10 ms, or 100 ms to 200 ms). A wide sweep range is most important for dramatic flanging effects; chorus and echo don’t need much sweep range to be effective. With longer delays, adding a little bit of modulation provides chorusing, but too much modulation will cause detuning effects.
Modulation type. The modulation usually comes from periodic waveforms such as triangle or square waves, but some devices include randomized waveforms and/or envelope followers (where the modulation tracks the incoming signal’s dynamics).
Modulation Rate sets the modulation frequency. Typical rates are 0.1 Hz (1 cycle every 10 seconds) to 20 Hz. With flanging and chorusing, modulation causes the original pitch to go slightly flat, return to the original pitch, go slightly sharp, then return to the original pitch and start the cycle all over again.
Annoying habits. The delay readouts on older hardware models are not always 100% accurate. Also, changing delay times via MIDI usually results in burping and belching as the device flushes its memory and refills.
Hot tips. For vibrato, set a short initial delay (5 ms or so), monitor delayed sound only, and modulate the delay with a triangle or sine wave at a 5 to 14 Hz rate. • To create a “comb filter,” mix a straight signal with the same signal passing through a short, fixed (unmodulated) delay. Try an initial delay of 1 to 10 ms, minimum feedback, no modulation, and an equal blend of processed and straight sound. • For mono to pseudo-stereo conversion, set a stereo chorus depth parameter to maximum and rate to minimum (or off). This creates a stereo spread without the motion that would result from having a higher modulation rate. • To calibrate the echo repeat time to a particular rhythmic value, such as an eighth or quarter note, the following formula translates beats per minute (tempo) into milliseconds per beat (echo time): 60,000/tempo = time (in ms).
For more information:
Profile. The pitch transposer synthesizes a harmony line from an input signal. Simple pitch transposers are limited to parallel harmonies, while more sophisticated models produce “intelligent” harmonies if you specify a key and mode (major, minor, etc.).
How it works. A pitch transposer essentially cuts a signal into a little pieces, then glues them all back together—in real time, except for a few milliseconds of processing time—so that they take up less time (shifts pitch up) or more time (shifts pitch down).
Crucial parameters. Transposition sets the harmony line interval, typically in semitones but with an additional fine tuning control.
Blend or Mix sets the balance of dry and transposed signals.
Feedback, Regeneration, or Recirculation feeds some of the output back to the input to create stepped harmonies and other special effects.
Intelligent harmony settings consist of key and scale data so the pitch transposer generates harmonies based on the rules of harmony for the specified scale.
Waves’ UltraPitch generating a harmony from a vocal track.
Annoying habits. It takes a lot of processing power to do pitch transposition, and the sound sometimes suffers. For example, there might be a fluctuating tremolo effect, or occasional glitches. The greater the degree of transposition, the more objectionable the sonic problems.
Hot tips. Even if your transposer doesn’t offer “intelligent” harmonization, you can often change the transposition amount via MIDI by using continuous controllers as you play. • For glissando effects, set the transposed pitch very slightly higher than normal (a few cents), then advance the regeneration control. This recirculates and pitch shifts each note, thereby initiating a stepped, upward glissando effect (the harmony pitch control controls the step interval). • Pitch transposers can give excellent flanging/chorusing effects. Set the pitch control for a very slight amount of transposition (1 to 20 cents or so) and add regeneration to taste.
For more information:
Profile. The noise gate helps remove noise and hiss by shutting off the audio whenever the input signal drops below a certain threshold. As a bonus, some noise gates can also provide special effects.
How it works. The presence of a loud musical signal masks hiss, which becomes audible only during quiet parts when the music is not playing. Setting the threshold just above the hiss level will allow the signal to pass if its level exceeds the threshold, but will block the output if the signal level drops below the threshold and consists solely of hiss.
Crucial parameters. Threshold or Sensitivity determines the reference level above which the gate opens. High threshold levels are useful for special effects, such as removing substantial amounts of an instrument’s decay to make a more percussive or gated sound.
Attenuation. Some noise gates feature adjustable attenuation for the gate-off state. With less attenuation, the gate doesn’t shut down all the way so that some of the signal can still pass through.
Decay time sets a fadeout time for the audio when the signal goes under the threshold.
Attack time works in reverse: when a signal exceeds the threshold, the noise gate fades in over a specified period of time.
Key Input or Sidechain Input allows an external audio signal to open and close the gate.
Focusrite Gate, part of their Scarlett suite of plug-ins.
Annoying habits. Somet mes the gate dr ps out some sig als that y u do want to hear. Also, noise gates work best on signals that don’t need to be cleaned up too much. Eliminating high noise levels also means nuking substantial portions of the signal.
Hot tips. If possible, avoid noise gates for noise reduction since they tend to destroy low-level dynamics. • The key input is very cool for special effects. For example, gate a sustained chord with a kick drum beat to “chop” the chord into rhythmic slices. • For a huge drum sound, mic the drums so they include a lot of room sound, compress the signal, then gate it with a high threshold. This lets through bursts of room sound, but eliminates the reverberant decay.
For more information:
Profile. Reverberation simulates the sound of audio reflections bouncing around inside an acoustic space (e.g., large hall or auditorium). Digital reverb can also create spaces that don’t exist in nature.
How it works. Digital reverb processes digital audio through an algorithm that creates a series of delays with filtering, similar to the reflections that would occur by sound waves bouncing off acoustical surfaces.
Crucial parameters. Type or Algorithm determines the kind of reverb to be emulated: room, hall, plate, spring (the classic “twangy” reverb sound used in guitar amps), etc.
Room Size determines the room’s volume. Changing this parameter often changes other parameters, such as low and/or high frequency decay.
Early Reflections level. Early reflections are closely spaced discrete echoes, as opposed to the later “wash” of sound that constitutes the reverb’s tail. This parameter determines the level of these initial, discrete echoes.
Predelay sets the amount of time before the first group of reflections or room reverb sound begins, and is usually 100 ms or less. A longer predelay setting gives the feeling of a larger acoustical space.
Decay time adjusts how long it takes for the reverb tail to decay to a particular level (usually -60 dB). Note that there may be separate decay times for different frequency bands so you can more precisely tailor the room’s characteristics.
Crossover Frequency applies only to units with separate decay times for high and low frequencies. This parameter determines the “dividing line” between the highs and lows. For example, with a crossover frequency of 1 kHz, frequencies below 1 kHz will be subject to the low frequency decay time, while frequencies above 1 kHz will be subject to the high frequency decay time.
High Frequency Rolloff or Damping. In a natural reverberant space, high frequencies tend to dissipate more rapidly than lows. High frequency rolloff helps simulate this effect.
Mix, Balance, or Blend. Sets the mix between the reverberated and straight signals.
Diffusion is a “smoothness/thickness” parameter. Increasing diffusion packs the early reflections closer together, giving a thicker sound. Decreasing diffusion spreads the early reflections further apart. Some reverb units call this Density, and some diffusion controls affect all reflections, not just the early ones.
Annoying habits. Even the best digital reverbs don’t really sound like clapping your hands in a cathedral. An acoustic space remains the best way to do reverb.
Hot tips. Different instruments can sound better with different reverb settings. For example, low density settings can be problematic with percussive sounds, since the first reflection could sound more like a discrete echo than part of the reverb. Increasing the density solves this. However, low density settings can work very well with voice to add more fullness to the overall sound. • To create a “bigger” sound, set the low frequency decay longer than the high frequency decay. For a more ethereal sound, do the reverse.
For more information:
Profile. This provides a periodic amplitude change so that the sound seems to “pulsate.”
How it works. A modulation source, such as a triangle or sine wave, controls amplitude.
Crucial parameters. Modulation Amount, or Depth determines how much the modulation section varies the amplitude.
Modulation Rate sets the modulation frequency.
Modulation Type. Some tremolos include different modulation waveforms.
Annoying habits. The tremolo in old guitar amps can’t sync the tremolo modulation frequency to incoming tempo data.
Hot tips. Tremolo is the driving sound behind surf music, but it was also used on vocals back in the 60s, when people were so stoned they thought it actually sounded good.
Profile. The tradtional exciter increases brightness at higher frequencies without necessarily adding equalization; however, the term also applies to adding brightness to lower frequency ranges as well. The result is a brighter, “airier” sound without the stridency that can sometimes occur by simply boosting treble.
How it works. Different processes vary, but one popular model adds subtle amounts of high-frequency distortion. Sometimes phase changes will also factor into the sound.
Crucial parameters. Exciter Frequency sets the frequency at which the “excitation” starts to kick in.
Exciter Mix or Amount varies how much “excited” sound gets added to the dry sound.
Annoying habits. People usually turn these up too much, and ruin otherwise perfectly good-sounding songs.
Hot tips. Processing an entire mix through one of these boxes can be overkill. Instead, consider feeding the exciter with an aux bus and add in subtle amounts for various channels, as needed.
Profile. A vocoder primarily creates “talking instrument” effects, but can also be used to modulate one signal with another (e.g., modulate a sustained keyboard pad with drums).
How it works. A vocoder has two inputs, the carrier input for an instrument, and the modulator input for a microphone or other signal source. Talking into the microphone superimposes vocal effects on whatever is plugged into the instrument input by opening and closing filters that process the instrument sound according to the frequencies present in the human voice.
Crucial parameters. Carrier Input level sets the level of the carrier signal (duh).
Modulator Input level adjusts the modulator signal level.
Balance sets the blend of mic with vocoded sound.
Highpass Filter adds in some high frequencies from the mic channel directly into the output to increase intelligibility.
Annoying habits. The filters are so sharp, it’s easy to overload them and get distortion.
Hot tips. Vocoders are good for much more than talking instrument effects. For example, play drums into the microphone input instead of voice, and use this to control a keyboard playing sustained chords. • For best results, the instrument being processed should have plenty of harmonics. This is why distorted guitar works well with vocoding.
For more information:
Craig Anderton is Editorial Director of Harmony Central. He has played on, mixed, or produced over 20 major label releases (as well as mastered over a hundred tracks for various musicians), and written over a thousand articles for magazines like Guitar Player, Keyboard, Sound on Sound (UK), and Sound + Recording (Germany). He has also lectured on technology and the arts in 38 states, 10 countries, and three languages.