Articles Comments

KARAGYOZOV.COM » Sound theory » Lecture 6 – Reverberation

Lecture 6 – Reverberation

REVERBERATION

 

When we have a sound emitted in an enclosed space, upon arrival to the listener it comes with many discrete reflections from the surfaces of the room.

These multiple reflections cause auditory impression of “prolongation” of the sound until the reverberation process died down.

The amount of reverberation generated by the surfaces of the hall is directly proportional to the magnitude and and inversely proportional to the sound absorption, which is carried out from its surface.

 

 

The acoustic energy decreases as due to its absorption by the surfaces of the room, as well as during its transmission on the air.

reverbfaq_reverb.l

 

The discrete echoes, that accompany the main sound are not perceived as such due to the time proximity and to the fact that they are numerous. Therefore they could not be differentiated separately but as a whole queue of progressively decreasing signal.

This process can either to exist in real sound space, as well as it can be simulated.

As we can see from the graph above, there are certain parameters of the reverberation process.

  • In the first place it’s the reverberation time.

It would be incorrect to define this time as its decay to zero (complete disappearance of the reverb signal) due to the fact that this time will depend on the level of sound emitted into the room. Therefore, this time is defined as the time needed by the initial sound to be reduced by the value of 60 dB (or 1 million times).

This is called reverberation time.

  • Another parameter of the reverberation is called. pre-delay.

This is the time that elapses between the sounding of the main signal and the beginning of the reverberation process. It is measured in milliseconds. Auditory impression is such that at a higher amount of pre-delay we have the impression of greater dimension of the room and back.

  • The next parameter is the so-called early reflections.

These are the first reflections after reaching the main sound that can be perceived as discrete ones and are caused by the reflection in the closest surfaces – walls, floor and ceiling of the scene as an example. An important role in auditory impression of reverb have the values ??of the early reflections – their level and their time of arrival and the relationship between them and the reverberation tail like levels.

Early reflections and reverberation tail are the main components of the reverb. If the early reflections are more dense and arrive more quickly, they create the impression of a more compact room, if they are more widely placed – for larger.

 

As seen in the last graph, the early reflections may exist as a parallel process in the time, and not just precede the rest of the reverberation process.

To discuss the other parameters of the reverb, we must list  first the methods by which reverberation works in the sound engineering job.

First, it is important to say that the reverb is very important segment of the perception of sound in both physiologically and aesthetically meaning. The sound for the ear and the brain is closely related to the concept of space in which he resides and from this perspective it is important what the real or virtual space we will put it.

On the other hand, in terms of understandability of speech, for example, the reverb is a factor that rather is preventing it, so that in certain situations the reverb is more obstructive and we must try to reduce it while  recording through proper mic positioning in the recording mode (we are speaking only for those specific sound examples).

The hearing is so arranged that it manages with great accuracy to determine the type of surrounding sound space (size, shape, surrounding surfaces reflect sound similar objects, etc.) based only on the auditory impression of his reverb. Therefore, it is extremely difficult to create a good reverb algorithm that creates realistic auditory picture of our environment.

On the other hand, it is often necessary in sound processing because of the fact that in many cases we pickup a sound through a microphone in a room, which is very different from one in which we would like to put our virtual artist.

Often to avoid reverberation of the room in which we record and which would respectively  be recorded in parallel with the main sound (which would be undesirable in view of the subsequent processing), a record is made in a room (studio) with high sound absorption, preferably uniform in all frequency regions.

Let’s see how we can create an artificial reverb.

The first way is the echo chamber.

For this purpose we  are using a   room in which we have microphones and speakers.

 

The example here is the echo-chamber at Capitol Studios. In this way you can get a very natural sound of reverb, especially in the classical field.

The problem with this technique, first, is the space we need to do this, and, on the other hand – in the limited way to control the reverberation time and the another parameters.

Spring-Reverb

 

The second way is the creation of artificial reverb surfaces – plate and springspring reverb and plate reverb.

It uses input and output magnetic transducer. Excitation of the vibrations in one part of of the device results in the repeated reflections captured in the other part.

The problem in these reverbs is the unnatural speed and density of the reverberation curve, as well as the increased energy at high frequencies compared to the way the same process takes place in the air, where they naturally fade away with greater speed.

Furthermore, those reverbs gave specific “ringing” effect caused by the standing waves evolving in such environments. However, this should also be related to the specifics of this kind of reverbs.

plate_reverb

 

Nevertheless, these algorithms are still used because of fashion that they have been imposed in the popular music. For example, the plate reverb EMT 140 is widespread and many imitated in many software emulations due to the preserved nature of its sound and to the numerous recordings made  with it over the years.

 

 

 

With the advent of digital transmission of the sound and with the improvement into performance of microprocessors logically occurs and the digital reverb.

Should be taken into account that the process of reverb, as we said in the beginning, is very complex and therefore difficult to emulate, be it in a digital environment.

The problem is not only in the calculations in real time, witch are necessary for this purpose, and to give a “random” nature of the generated reflections in order to avoid unnatural and mechanical sound of the reverb. Brands like Lexicon, Tc Electronic and Bricasti are able to do it.

clip_image002

Signal path in the software reverb of SONNOX.

 

Copyright – Sonnox Ltd.

 

 

 

The principle of the so-called. algorithmic reverb is the generation of a set of early reflections with adjustable parameters and multiple delays that are multiplied through feedback effects applied on them. This way we obtain an emulation of a reverberation tail.

 

While some brands rely on the generation of delays based on the early reflections, another brands as Lexicon and Tc Electronic rely on separate algorithms for the early reflections and the overall reverb. This creates a more realistic effect and allows more control of the independent parameters.

 

Algorithmic reverbs are among the most popular because they do not have the disadvantages of the spring / plate predecessors. On the other hand, the various possible algorithms of creation offer unlimited possibilities for experimentation not only in the field of emulation of halls and rooms, but also in the creation of artificial sound spaces of all kinds.  Often, additional effects such as filtering, modulation, chorus, etc. are added to them, which effects additionally can create new, non-existent until now sound spaces.

Along with this, with the modern computational capabilities of the processors an algorithmic presets can be manufactured that are commensurable to real halls but with the ability to edit all the possible parameters of the reverb.

 

The last type of reverbs has developed in end of the 20th and the 21st century and is called. convolution reverb.

PTY

This process uses convolution, in which is made a “snapshot” of an existing room. For this purpose, the so-called. impulse response is used.

 

If we produce in the room a very strong uncorrelated noise that is extremely short, the residual reflected signal is the impulse response of the room. Saving it, we can load it into the specially
developed device that is processing the input signal through the algorithm of this impulse response.

 

sweep-spectrum

Another method is the use of a frequency-variable sine wave – ie. sweep tone. This is a more perfect method, but with it is necessary to implement deconvolution – removing the impulse response of the overall sound, which is accomplished by the appropriate mathematical operation.

 

Whichever method is used, for most users it is essential the result (unless you do not intend to record your impulse response). And the result is more than good in terms of the realism of reverb, because here we do not have an artificial modeling  of the reverb process, but a “snapshot” of a particular space with all its details. We can create many impulse response of both halls and different kinds of spaces. With their help we can “put” the listener in a room of our choice.

These impulse responses, when made in halls are often recorded from different places in the hall, enabling us to subsequently select the virtual space of the hall-microphones and the hall- radius that we want to achieve. In recent times, often a libraries with impulse responses in surround-format are made, which can be loaded into surround – reverbs. It gives us great features in the field of post-production for films and concerts.

convolution reverb Altiverb

In making motion pictures the convolution technique allows us to “scan” the sound of spaces in which any shooting process happens and subsequently to prepare their own impulse responses. Thus, in post-production there can be achieved full compliance with the reverb of the space, not just imitation, as is the case with algorithmic reverbs.

And last but not least by the technique of impulse response we can “sample” as impulse response not only acoustic spaces, but also an electronic reverb devices. Thus this device becomes a perfect emulator of other reverbs – analog and digital.

This is used heavily today in terms of both vintage plate and spring reverbs, and on expensive and fashionable as an algorithm  devices, that may not be present in our rack with effects, but can be successfully mimicked by their impulse response loaded in convolution reverb.

Let us now, on this basis to consider the basic parameters that we can monitor and regulate in a digital reverb. In the beginning we just noted some of them, which are characteristic of the reverb at all. These are the decay time (or reverb time, or RT60), predelay, early reflections. We clarified how they affect the auditory sensation of reverb.

Practically, these are the basic parameters with which change is changing the feel for the size of the sound space, in which we find ourselves. The early reflections give a feeling of size, distance and shape of the closest spaced surfaces, while the overall reverberation tail – about the depth, height and volume of the hall. Predelay such as distance from the main signal also defines the hall as a room size.

Thereafter there are some other parameters in which the sense of the characteristics of the room changes when we modify them.

For example, the reverb is also frequency regulated – on one hand by adjusting the input signal in the reverb and from another – of the output signal.

Reverb_rverb

Waves Renaissance Reverb – can be seen the function damping, and the overall equalizer on the output of the reverb.

 

In most devices the function High-damping is present. This is a possibility to adjust the reverb algorithm  in such a way that the reverberation time to decrease in the high frequency area. Sometimes we can have and regulation of low-damping. Although this seems like a frequency correction, it is not producing only this effect, because in this case we talk also about a different reverberation time. In the case we are simulating not only different reflecting surfaces, but also the natural attenuation of the high frequency audible component when crossing the sound in air.

Apart from this we have frequency control of the reverb by which we similarly mimic the selective reflection of the different sound frequencies for different surfaces.

Here is the place for two other parameters, witch are frequently used in the digital reverbs – density and diffusion.

Density is generally related to the total number and temporal distribution of the early reflections. It simulates also the manner in which the sound is diffused with various reflective materials. For example, surfaces such as marble and glass have almost no dispersion, while the wood due to the microporous structure has a high degree of dispersion.

The high value of the density in the reverb gives us a feeling of a rich and saturated reverb, while low density “thins” the reflections and makes it “light”. However, with very low values this ??can lead to a ringing effect due to the small number of early reflections. For this reason, care should be taken with the reduction of density parameter when we process musical instruments with fast transients such as snare or percussion.

Diffusion is a parameter linked to the density, as well as other parameters such as the size (size of the room), the shape, etc. It extends not only to the early reflections, but on the rest of the reverberation process and formed the feeling of how much “random” nature have the reflective surfaces in terms of type, reflectivity and placement in the hall. If a room has a strict form and uniform surfaces, this parameter is low, if you simulate various obstacles of different nature and with different structure of the material, the diffusion is large.

Recommended for cleaner sound (eg solo piano) to seek lower diffusion as the high value of this parameter necessarily leads to frequency coloration of the reverb and makes the process more “dull”. On the other hand, the high diffusion still creates a realistic sense of a reverberation process, without it it is somewhat artificial.

In some investigations the diffusion as a parameter is coupled with the stereo distribution of the reflections, in some it is rather a factor dependent on the density, and some there is a separate parameter for it. The fact that some of these studies have been published by creators of reverb algoritms talked about that in many cases, different algorithm in one particular  manufacturer results in a different implementation of these parameters and in many cases they don’t represent in one manufacturer what they represent to another.

Parameters such as density, diffusion, size, spread, shape, etc., can vary in the way in which they affect on the reverb relative to the various methods of reverb used during the years. In most cases does not apply to reverb time, early reflections, mix, predelay and in general to the basic parameters of the reverberation process.

Along with that there are parameters that are not always present in all the reverbs. This is the Shape, in which we adjust the shape of the reverberation curve. It is very important because in nature we do not always have an exponential curve on the reverb, but it depends on many factors, including the shape of the hall, reflective surfaces, etc.

In some Reverb shape is used to define the curve of the density of the early reflections.

There are many other parameters presented in certain types of reverb. That for example is spread – usually the width of the stereo distribution of the reverb signal, modulation – when we introduce a modulation in the signal, feedback – when part of the signal after the reverb is going back into the reverb circuit in order to produce more “random” effects, etc . etc. Very often presets set in reverbs are a combination of many of these parameters, witch through precise selection of these parameters simulates a given acoustic space. We can meet parameters such as width, size, position, absorbtion, etc.

In all cases it is important to regulate the basic parameters of the reverb – mix (dry / wet), decay time, ER mix, predelay, eq. Other parameters can be selected in most part by selecting a specific reverb preset that define them, and to change them only in case of need.

 

 

Usefull links

 

 

http://www.soundonsound.com/sos/Oct01/articles/advancedreverb1.asp

http://www.soundonsound.com/sos/may00/articles/reverb.htm

http://corpuscul.net/shkola/sozdanie-bita/effekty/

http://valhalladsp.wordpress.com

http://wiki.audacityteam.org/wiki/Principles_of_Reverb

http://en.wikipedia.org/wiki/Reverberation

http://www.retrosynth.com/~analoguediehard/studio/effects/digitalreverb.html

http://www.sonnoxplugins.com/pub/plugins/products/Reverb-Tech_Detail.htm

http://www.softube.com/index.php?id=predelay_diffusion

http://plus.pointblanklondon.com/reverb-parameters-explained-in-logic-pre-delay-attack-decay-diffusion-density-spread-room-shape/

Filed under: Sound theory

Leave a Reply

*