When we hear sounds in the “real world,” they are in an acoustic space. For example, suppose you are playing acoustic guitar in your living room. You hear not only the guitar’s sound, but because the guitar generates sound waves, they bounce off walls, the ceiling, and the floor. Some of these sound waves return to your ears, which due to their travel through the air, will be somewhat delayed compared to the direct sound of the guitar.
When we hear sounds in the “real world, ” they are in an acoustic space. For example, suppose you are playing acoustic guitar in your living room. You hear not only the guitar’s sound, but because the guitar generates sound waves, they bounce off walls, the ceiling, and the floor. Some of these sound waves return to your ears, which due to their travel through the air, will be somewhat delayed compared to the direct sound of the guitar.
This resulting sound from all these reflections is extremely complex and called reverberation. As the sound waves bounce off objects, they lose energy and their level and tone changes. If a sound wave hits a pillow or curtain, it will be absorbed more than if it hits a hard surface. High frequencies tend to be absorbed more easily than lower frequencies, so the longer a sound wave travels around, the “duller” its sound. This is called damping. As another example, a concert hall filled with people will sound different than if the hall is empty, because the people (and their clothing) will absorb sound.
Reverberation is important because it gives a sense of space. For live recordings, there are often two or more mics set up to pick up the room sound, which can be mixed in with the instrument sounds. In recording studios, some have “live” rooms that allow lots of reflections, while others have “dead” rooms which have been acoustically treated to reduce reflections to a minimum – or “live/dead” rooms which may have sound absorbing materials at one end, and hard surfaces at the other. Drummers often prefer to record in large, live rooms so there are lots of natural reflections; vocalists frequently record in dead rooms, like vocal booths, then add artificial reverb during mixdown to create a sense of acoustic space.
Whether generated naturally or artificially, reverb has become an essential part of today’s recordings. This article covers artificial reverb – what it offers, and how it works. A companion article covers tips and tricks on how to make the best use of reverb.
Different Reverb Types
|
|
There are two main types of artificial reverb: Synthesized and convolution-based. Synthesized reverb “models” the sound of a room through the use of various algorithms (Fig. 1). For example, a “Hall” algorithm will take into account that waves travel further in a concert hall than in a small room, so the reverb will take longer to decay. A “Room” algorithm might model a small room, like a club or practice space. Other algorithms model artificial reverbs, such as “Spring” reverbs found in guitar amps, or “Plate” reverbs that were used extensively in the 60s. Each algorithm has a different sound quality, but they all work in the same basic way: A signal comes into the reverb, is analyzed, and the reverb algorithm generates echoes and reflections that mimic what happens in the chosen acoustic space.
Convolution reverb is a relatively new type of technology that “samples” the sound of a room. Typically, adevice like a sports starting pistol will create an impulse that creates reflections in a room. These reflections are recorded, analyzed, and converted into a very accurate model of that specific room. A good analogy is that a convolution reverb’s impulse is like a “mold” that you pour sound into, and the sound acquires the characteristics of being in that room.
|
|
You can think of the difference between synthesized and convolution reverb as the difference between a synthesizer and a sampler. The synthesizer will give more control over the sound but have a more “impressionistic” character, while a sampler provides an extremely accurate, but generally less editable, sound.
Another consideration is that convolution reverb is a very processor-intensive operation. Only recently have computers become powerful enough to allow for real-time operation, and even then, you might experience some audible delays due to processing. Fortunately, as reverbs are based on delays anyway, with fast computers you might not notice anything objectionable.
Reverb Elements
|
A sophisticated reverb will have many parameters, but few people know how to optimize these parameters for specific recording situations. So, let’s discuss how the various parameters affect your sound.
Reverb has two main elements (Fig. 3):
The early reflections (also called initial reflections) consist of the first group of echoes that occur when sound waves hit walls, ceilings, etc. These tend to be more defined and sound more like “echo” than “reverb.” You can often adjust the level of early reflections.
Decay, which is the sound created by these waves as they continue to bounce around a space. This “wash” of sound is what most people associate with reverb, and is often called the reverb tail.
Another parameter, Pre-Delay, sets the time for the first sounds to travel from the source to the first set of reflections. The larger the space, the larger the pre-delay because it takes more time for the signal to arrive at a wall or ceiling and start bouncing around.
Advanced Parameters I
Following are some of the parameters found in higher-end synthesis-based reverbs; less expensive reverbs will have a subset of these parameters. Convolution reverbs generally have fewer parameters, but in the past few years, engineers have figured out how to make convolution reverbs more editable.
Algorithm. We’ve already mentioned hall and room algorithms, as well as algorithms that emulate “vintage” synthetic reverbs. But you may also find algorithms like cathedral, gymnasium, small room, closet – anything is possible! There are even “reverse” algorithms where the decay builds up from nothing to full volume rather than decay from full volume to nothing, and “gated” algorithms that abruptly cut off the reverb tail below a certain level (this effect was very popular in the 80s, particularly with Phil Collins’ albums).
With convolution reverbs, the equivalent concept is called an impulse. Impulses may capture the sound of specific rooms (like particular concert halls), or even the sound of spaces like guitar cabinets. It’s even possible to create impulses of older reverbs, so there could be an impulse that sounds like an old Lexicon PCM-70.
Here are some examples of different room types. These examples all use a single percussive hit to excite” the reverb, as this provides an easy way to compare the effects of different reverb types and settings.
Small Room – Plate – Cathedral – Bright Chamber
Room size. This affects whether the paths the waves take while bouncing around in the “virtual room” are long or short. Just like real rooms, artificial rooms can have “standing waves” and resonances. If the reverb sound has flutter (a periodic warbling effect), vary this parameter in conjunction with decay time (described next) for the smoothest sound.
Decay time. This determines how long it takes for the reflections to run out of energy. Remember that long reverb times may sound impressive on instruments when soloed, but rarely work in an ensemble context (unless the arrangement is very sparse). The spec for decay time is called RT60, which means the time it takes for a signal to decay to –60dB of its original amplitude. For example, if RT60=1.5, then it takes 1.5 seconds for the signal to decay to –60dB or its original level.
Damping. If sounds bounce around in a hall with hard surfaces, the reverb’s decay tails will be bright and “hard.” With softer surfaces (e.g., wood instead of concrete), the reverb tails will lose high frequencies as they bounce around, producing a warmer sound. If your reverb can’t create a smooth-sounding high end, introduce some damping to place the focus more on the midrange and lower frequencies. Listen to these two audio examples to hear the difference.
Much Damping (Dull) – No Damping (Bright)
Advanced Parameters II
High and low frequency attenuation. These parameters restrict the frequencies going into the reverb. If your reverb sounds metallic, try reducing the highs starting at 4 – 8kHz. Note that many of the great-sounding plate reverbs didn’t have much response above 5 kHz, so don’t worry if your reverb doesn’t provide a high frequency brilliance – it’s not crucial.
Reducing low frequencies going into reverb reduces muddiness; try attenuating from 100 – 200Hz on down.
Early reflections diffusion (sometimes just called diffusion). Increasing diffusion pushes the early reflections closer together, which thickens the sound. Reducing diffusion produces a sound that tends more toward individual echoes than a wash of sound. For vocals or sustained keyboard sounds (organ, synth), reduced diffusion can give a beautiful reverberant effect that doesn’t overpower the source sound. On the other hand, percussive instruments like drums work better with more diffusion, so there’s a smooth, even decay instead of what can sound like marbles bouncing on a steel plate (at least with inexpensive reverbs). You’ll hear the difference in the following two audio examples.
Maximum Diffusion – No Diffusion
The reverb tail itself may have a separate diffusion control (the same general guidelines apply about setting this), or both diffusion parameters may be combined into a single control.
Early reflections predelay. It takes a few milliseconds before sounds hit the room surfaces and start to produce reflections. This parameter, usually variable from 0 to around 100ms, simulates this effect. Increase the parameter’s duration to give the feeling of a bigger space; for example, if you’ve dialed in a large room size, you’ll probably want to add a reasonable amount of pre-delay as well.
Reverb density. Lower densities give more space between the reverb’s first reflection and subsequent reflections. Higher densities place these closer together. Generally, I prefer higher densities on percussive content, and lower densities for vocals and sustained sounds.
Early reflections level. This sets the early reflections level compared to the overall reverb decay; balance them so that the early reflections are neither obvious, discrete echoes, nor masked by the decay. Lowering the early reflections level also places the listener further back in the hall, and more toward the middle.
High frequency decay and low frequency decay. Some reverbs have separate decay times for high and low frequencies. These frequencies may be fixed, or there may be an additional crossover parameter that sets the dividing line between low and high frequencies.
These controls have a huge effect on the overall reverb character. Increasing the low frequency decay creates a bigger, more “massive” sound. Increasing high frequency decay gives a more “ethereal” type of effect. With few exceptions this is not the way sound works in nature, but it can sound very good on vocals as it adds more reverb to sibilants and fricatives, while minimizing reverb on plosives and lower vocal ranges. This avoids a “muddy” reverberation effect that doesn’t compete with the vocals.
THE NEXT STEP: APPLYING REVERB
Now that we know how reverb works, we can think about how to apply it to our music – but that requires its own article! So, see the article “Applying Reverb” for more information.
