miércoles, 4 de junio de 2014

Algorithmic and Convolution Reverb



Hello, I am Carlos Devizia from Argentina, and today I will write about reverb.

Reverb is a very useful effect, which belongs to the category of "Delay Effects".  We can use it to re-create a real space or to create an exotic sonic environment, not necessary inspired in the real world.

Usually, when we record an instrument, we tend to isolate it, to remove reflections in the room and remove as much noise as possible. Though, this is useful at that stage, to record the best signal possible; in the final mix, this kind of sound results unnatural. So, we can use reverb to create the correct space for our production. As we said earlier, we have different approaches to it. We can re-create a natural space, a completely fantastic atmosphere, we can use reverb to separate the instruments in different stages, like if some of them were closer than others, we can define a common space for all the instruments as if they were recorded in the same room and at the same time, etc. We can be completely creative here.

As all the effects in this category, reverb has some basic parameters to care about. The delay time is the amount of time from the input to the output of the signal, the dry/wet control balances the delayed and undelayed signal and feedback determines how much the wet signal is routed back to itself. 

Reverb units are divided in two categories: algorithmic reverbs and convolution reverbs. The first one uses a kind of formula to create the effect. An algorithm is a step by step procedure, a series of rules defined to produce a certain activity to achieve a concrete result. This is the way this kind of reverb works. On the contrary, convolution reverb works based on a recording of a real space that you apply to your sound file.

Both types have their strong and weak points. Convolution reverb is great to re-create real spaces, because you are applying to your sound a record of a reverb made in the real world, however it is limited in the way you can manipulate this. Algorithmic reverb, on the contrary, offers an incredible amount of manipulation possibilities but is does not sound as realistic as convolution reverb.

We´ll take a look at both, we use a couple of plug-ins for our examples, and it is important to say, that though from plug-in to plug-in the names and locations of the controls may vary, the concepts are essentially the same. You may find that some plug-ins have more controls than others, but the main features remain the same. So, no matter with VST you´ll be using, if you understand the main concepts you´ll be able to work with your favourite reverb VST.


Algorithmic reverb

They are usually divided in two separate components: the early reflections which are a series of short delays and the difuse reverb.

In the early reflections section you´ll find a pre-delay control, which works on the beginning on the entire reverb delaying it a little bit. We also can manipulate the space with a control named room space, room shape or a similar name according to your particular plug-in, and that will help us changing the location of the early reflections.

In the difuse reverb section you will control mainly the time parameter. This will affect the size of the room your sound is in. The high end (also high cut, damp, etc.) parameter controls the materials of the room (Some materials reflect the sound better than others, and some are softer than others and this affects the way the sound propagates).

According to the different plug-ins, you may encounter different options and possibilities, like pan controls to panoramize the reflections to the right or to the left, a freezing control to avoid the decay of the reflections, some EQ filters associated to the reverb, to enhance certain reflections on certain frequencies and attenuate some others, etc.

Let´s hear a couple of examples of algorithmic reverb applied to a short fragment of music, in this case played with an alto sax. The first sound shows the dry sound and then the sound processed with different reverb settings. All the examples in this writing do not intend to be musical, they aim to show differents effects achieved by the use of reverb.











Here are some very good freeware reverbs: 



Epic Verb


 FreeverbToo


 Mverb





Convolution reverb

In this type of reverb we choose an impulse response, which is a file containing a record of a real reverb that we will apply to our sound. Usually convolution plug-ins have several files associated to them and there are also sets of files on the web downloadable to use with convolution VSTs. The interesting thing about this is that you have an incredible amount of impulse responses available for your use, from very realistic ones to totally strange things. The trick here is to imagine in which space you would like to hear your music and based on that selecting your impulse file. Thus, your music can sound as if it were played in a small room, in a concert hall, in a big cavern, etc. In a convolution plug -in you´ll also find some controls to manipulate the reverb.

Let´s hear a couple of examples of the sound presented earlier in this tutorial with different convolution reverbs applied to it. Let´s listen to the same sound with a couple of convolution reverb applied to it.






Here is a very good Convolution reverb freeware,

SIR




In both algorithmic and convolution reverb you´ll also have the dry/wet control to mix the signal without reverb and the processed signal. The dryer the sound it will sound closer to the listener and the wetter , it will sound far away from the listener.




Conclusion

Reverb is in nearly every element of modern music production, so it has a dramatic impact on the final results we will achieve. Changing the way we apply reverb will change our final sound, so, the more we know reverb and the way it works the more our music will be sounding in the way we want. A good starting point for using reverb is being subtle, unless we are doing something crazy on purpose, subtle adjustments provide great results.

martes, 27 de mayo de 2014

Riding the Fader



Hello, I'm Carlos Devizia  from Argentina
Today I will explain the technique called "Riding the Fader".
Sometimes you have a performance that you like, but there are some parts in which it is too loud and some other parts in which it is too quiet. It often happens with the vocals. In this example, and for the safety of your ears I won´t sing. Instead, I'll demonstrate this technique over an instrument performance. I´ll use Zynewave Podium free, but you can replicate this technique with your favourite DAW. You may encounter some differences between DAW to DAW, but the main concepts are the same.

The concept behind "riding the fader" is easy, when the volume of the sound goes up, we turn the fader down, when the volume goes down, we turn the fader up. Just like that. What we are actually doing is a kind of manual compression. In fact, every compressor does this kind of job, first it analyzes the signal and then compensates the volumes in the desired way.

OK, let´s work! We have our audio in an audio track. It is a short clarinet performance, and there are some noticeable differences in volume in it. In part due to the difference between the registers of the instrument but also because of the performance in which I played with more strength in some parts than in others.

We can listen to the original performance







We right-click on the track and select "automate parameter" and then we choose "level". 
 


This will create an automation child track related to our clarinet track




By double-clicking on the level track, we'll see the following:









This is our clarinet part and we can begin automation here. Then, you must select the "pencil" tool" and just draw the automation envelope following the wave.






After doing that you may notice that your actual level is too low, so you can correct it by selecting all the points (Ctrl + A), and with the arrow tool taking them to your desired level. You can even move individual points if needed.









One of the good things of Zynewave Podium is that this automation changes does not affect the main fader of the track. So you can still increase or decrease the volume level of the whole track by adjusting the gain fader.



In other DAWs this is not possible and you´ll see that as the music plays your fader volume moves up and down, so if you want to increase or decrease the whole volume of the track you'll be in trouble. To solve this, you can add a Gain VST filter (there are lots of them, and some of them freeware as the one that I show below), and then ride the gain fader of the filter leaving the track fader to adjust the global volume of the track.




After finishing our work we can listen to the performance that now has a lower dynamic range and the loud parts and quiet parts are closer.





Then we can normalize our audio.



The conclusion of this small experience is that we can manipulate our volumes in an easy simple way by using this technique, and though it will be really helpful using it as the way it was designed for (the manual compression we talked about) it let the doors of experimentation opened and would be interesting to ride the faders in unusual ways too.
Finally, it is important to point that every time you alter audio you are making a decision. In this case, we should ask ourselves how much we want to ride the fader to equal the volume levels as, the more we equal the levels, the less dynamic range we will have in the end. These last years have seen a rise in the "loudness war" losing dynamic range in music. I think that it is a metter of balance, but this is up to every musician and every producer.




miércoles, 21 de mayo de 2014

Categories of Effects



Hello, I am Carlos Devizia a musician from Argentina. Today I will write about Categories of effects, giving small definitions of each one and providing some audio clips as examples of how they work. Please note that these examples are not intended to be musical, but they aim to show the effect itself clearly. Also be aware that, depending on how you tweak the parameters of the effects processors, you may obtain results drastically different.  And that´s a great advantage in the creative field, if we know how to use it.
We can associate different effects to different principles of sound.

Dynamic Effects

These effects are related to amplitude and they can control volume in different ways.
 In this category we find: compressors, limiters, gates and expanders.

Compressor
Essentially, this effect reduces the dynamic range of a piece by reducing the volume when it gets too loud. In this case "too loud" is related to a point called threshold or ceiling. Everything that goes above that will be reduced in a desired amount.

Limiter
This is basically a compressor with a fast attack and a high ratio.

Gate
Also known as noise gate, this effect attenuates any signal below a determined point. We also can say that basically a gate allows a signal to flow only when it is above certain level, stoping everything that is below that level.

Expander
This effect increases the difference between the loud parts and the quiet parts in an audio signal.

In the following example you´ll hear the clean sound and then processed with a compressor, a limiter, a gate and expander (in that order)




Delay Effects

These ones add slight delays to the signal and they are used to represent a space the listener is inside. In this category we find: delays, flangers, phasers, choruses and reverbs.

Delay
This effect takes an audio signal and reproduces it a certain time later, mixing it with the original signal. The amount and strength of repetitions can be adjusted by the musician, producer or sound engineer.

Flanger
This effect is achieved by taking to identical signals of audio, leaving one of them intact while delaying the second one by a gradually changing period.

Phaser
It is very similar to flanger, in the fact that this effect takes two signals leaving one untouched while processing the other one with a small delay and altering the phase of it. When both signals are mixed the frequencies that are out of phase cancel each other.

Chorus
Once again we have two similar signals. One of them is slightly delayed and usually modulated with an LFO. This gives the impression of a thicker sound and we can perceive some kind of movement in the resulting sound.

Reverb
This effect is related to the reflections of sound in various surfaces before reaching the ear´s listener. With our reverb units (hardware or software) we can re-create an existing space, enhance certain sounds (like a vocal performance, for example) or create a totally weird ambience.

Now you´ll hear the audio processed  with delay, flanger, phaser, chorus and reverb (in taht order)




Filter Effects

These effects control the timbre of sound. Among them we find : high pass, low pass, band pass and EQs.

High pass filter
This filter allows, as it is suggested by its name, to pass frequencies that are over a determined one.
Low pass filter
It is exactly the opposite of the High pass filter. It allows to pass the frequencies that are below a certain point, attenuating the ones that are over that point.


Band pass filter
It allows to pass the frequencies that are in a determined band of frequencies, attenuating those that are below or above that range.

EQ
The EQ affects the timbre of the sound by enhancing or attenuating certain frequencies (boosting or lowing frequencies). It is a vital part of the music production. It can be used both as a corrective tool and as a creative tool. You can find different types of equalizers: graphic equalizers, parametric equalizers and paragraphic equalizers.

In the following example you´ll hear the audio processed with a high pass filter, a low pass filter, a band pass filter and an EQ filter (in that order).




Conclusion


As musicians, producers or sound engineers, it is vital for us to understand each category of effects, how they are produced and which impact they have on an audio source. This will give us the opportunity to use them in a creative way, knowing exactly what we are doing and achieving the results we want to achieve instead of random results.