Transcript
The Tech Production Network Exam Review
Production techniques
Imagine with the live mixing console the signal starts at the top and flows down through the mixer channel. A studio mixing console is doing the same thing. A virtual mixing console is the one that is inside our software that we have been using inside programs like Logic and ProTools. Signals start on the tracks being played back as part of a multitrack recording. But all software programs have a virtual mixer page and the signal starts at the top of the channel and flows down in a similar manner. Imagine the track at the top of the mixer channel. There is one very important difference to our live mixing console. The live mixing console has all the controls and sections provided in the channel strip but all these things are there wether you need them or not. In the virtual mixer inside or software, we almost design the channel strip as we need it on the go by creating the switches and sends as we need them. EQ for example becomes a plug‐in which is a little piece of software that runs inside the main program. We insert an equaliser plug‐in when we need it instead it being on every channel. The signal is still flowing down a channel strip now but hits a plug‐in and goes in at the top and out the bottom of the plug‐in. We can have several plug‐ ins in a row and one can mix different types of plug‐ins. There is no limit to what type of plug‐in configurations are possible. Each channel strip can be different. The main types of plug‐ins are signal processing plug‐ins and effects. Some parts of a virtual software channel strip are already there like the channel fader and the PAN controls. You done need to create and insert those. Auxiliary Sends need to be created as we need them and processing plug‐ins for EQ, Dynamics and Time based effects. The reason we need to create certain aspects of a channel strip is that not all channels may need EQ for example, so by not inserting any plug‐ins on those channels one can see that certain channels have a minimum number or no processing at all. It simplifies the appearance of the mixer surface. This theory paper is mainly about the different types of processor plug‐ins you are likely to encounter and what they are called and what they do. There are three main types of signal processors. These processors take plug‐in form and you have all been using plug‐ins as part of the work you have done so far in the VET Music course. The three types are Dynamic Processors, Frequency Processors and Time Based Processors. We will start with Dynamics processors. These affect the volume or level of the signal in some way.
Version 10.1 modified 1/11/09
Page 1 of 9
© PKY Media Pty Ltd 2009
Let’s start with the Noise Gate. Every signal has an associated LEVEL with it. If a signal is soft then the level of that signal is LOW. If a signal is loud the level of the signal is HIGH. A gate can be considered as such: A gate has a threshold setting. Any signal then is either below that threshold or above the threshold. If a signal is below the threshold, the gate does not open and the signal does not go through. If the signal rises above a threshold then the requirements are met and the gate opens and the signal goes through. What are gates used for? On any track you may have loud signals intermixed with soft ones in between. A good example is if we record a bass drum or kick drum with a microphone inside the kick. The kick will come through loud and clear of course but other drums such as snares and toms will still get picked up by that bass drum mike and they will be on the track too. (Those unwanted sounds are called spill) Granted the spill is much quieter than the kick drum itself. But a gate can clean up all the messy quiet stuff in between the wanted loud kick drum hits. Gates are good for cleaning up drums sounds both live and in the studio. Another example is recording a very loud guitar amp. When the guitarist stops playing there may be a lot of hum or noise left and a gate can clean all that up very easily. When the loud guitars come back in again another effect called masking takes over where the very loud guitars hide the noise and hum of the amp. So you don’t notice any noise when the guitars are cranking but as soon as they stop all the noise and hum becomes very loud and clear. Gates can remove that. Compressors In music we have what we call dynamic range. This is the difference between the softest sounds and the loudest sounds that a musician can make. An example of high or wide dynamic range is when a jazz drummer goes from very soft notes on a snare up to really loud cracking snares. Or a guitarist very gently playing the strings up to really loud power chords. Low dynamic range is a drummer hitting a snare the same level all the way through a song. Or a guitarist playing a very consistent rhythm level. Or a vocalist going from a whisper to a scream at full roar!!! Most good musicians play at very consistent levels. A compressor is a device that can lower the dynamic range. It can convert a very high dynamic range ie very soft to very loud into a more useable dynamic range ie not so soft to not so loud ie more consistent range. We need this in studio work a bit more. Live it’s OK to be a bit more dynamic but in recordings we need to be bit more consistent with things. Very soft sounds can get lost in a recording and very loud ones can stick out too far. So by making very soft sounds a little louder and very loud sounds a little less loud we have got a more friendly dynamic range.
Version 10.1 modified 1/11/09
Page 2 of 9
© PKY Media Pty Ltd 2009
A compressor also has a threshold setting. Signals below the threshold don’t get affected but once a signal attempts to go above the threshold then the compressor kicks in and does its thing. It changes how much the louder signals are affected by diving any input changes by a factor called the ratio. If the ratio is 2 : 1 then for every 4 db the input attempts to change the output only changes by 2 db. Compressors tend to drop the overall level of the signal coming out so we add a bit of gain after a compressor and that is called makeup gain. This is all done within the compressor and it the reason why the softer sounds end up being louder and the very loud sounds being softer. A limiter is very similar except the output does not change level at all once the limiter threshold is met. It just keeps the output at a very constant level no matter what the input is doing. It is more severe in its action but there are several places where limiters are useful in live and studio work. In live work limiters are often used to protect speaker systems from damage. A band can get very loud but once a signal hits a limiter threshold then the sound cannot get any louder and that protects the speakers from damage. In a live PA the limiter is usually set fairly high so it does not really kick until things get very very loud. If a musician drops a microphone through a 50,000 watt PA it could destroy all the speakers out front in a second!!!! And hurt the audience at the same time! A limiter will arrest the signal to the point where the speakers will still be making the sound and it will be loud but they will cope with that and not blow up. In the studio we use the limiter during the mastering phase to keep overall song levels very consistent and now these there seems be a race to see how loud one can master a track and limiters are being used a lot nowadays for this purpose. It can harm the music in the end a bit and it’s unnecessary. (over limiting that is) But pop music needs limiting to make it nice and consistent and clear and loud. Sibilance and De Ess These are still dynamics processors but they work in a small part of the spectrum only as opposed to the whole spectrum like a compressor does. Sometimes when a singer sings a song the ‘s’ and ‘t’ is all get a bit loud and raspy. A de‐esser is a device that reduces these ‘s’ and ‘t’ sounds and makes them lower and smoother and more in balance with the rest of the vocal sounds. It is used on an individual tracks more. eg the vocal track. More a studio device but can be handy for keeping Kylie’s vocals under control in a live situation. FREQUENCY PROCESSORS The EQ as we have seen now in the mixing console is also in the studio in the form of a plug‐in. Firstly when we talk about frequency based processing we need to think of our graphical
Version 10.1 modified 1/11/09
Page 3 of 9
© PKY Media Pty Ltd 2009
representation as being in the frequency domain This is vertical or Y axis still representing level or amplitude but now the horizontal axis or X axis represents frequency. Frequency Domain diagrams. A flat response is when all frequencies are passed or not affected at all in any way. Hence there will be no change to the signal on a given track or stereo output. The human hear hears from 20 Hz to 20,000Hz The EQ Plug In Equaliser Think on the ends of the spectrum first. We have filters at both ends of the spectrum and the cutoff frequency is the frequency where frequencies are rolled off. We have a Low Filter (or High Pass as it can be called) The slope is the rate of change once the cutoff frequency is hit. Filters don’t boost they only cut things away. Handy for taming either end of the spectrum. On the ends of the spectrum are two more shelving EQ’s and these can boost and cut. Instead of a sloped response the helving EQ’s have a flat table like response curve hence the title shelving eq. We still have a frequency at which the shelf kicks in but the shelf affects all the frequencies the same amount. This group of frequencies can be boosted or cut (attenuated)
Version 10.1 modified 1/11/09
Page 4 of 9
© PKY Media Pty Ltd 2009
Peak/Dip Parametric EQ As seen in the mixing console we have another the of equaliser with a bell shaped curve that can move up and down and be flexible in terms of where it’s doing its work. The frequency or freq control sets the area where the equaliser is either going to boost or cut. Then we can boost or cut these frequencies with the boost/cut knob control . Another new control can be aged here and it is called Q factor and this determines the shape of the bell. The parametric equalizers on a live mixing control don’t often have the third control and that is Q. High Q settings mean very sharp bell curves which are often called steep peak and in the case of a steep cut it often called a steep notch. Very narrow notches are handy sometimes for removing an unwanted sound like a high squeak in a bass drum pedal. It can vary how many parametric equalisers can be in between the ends of the spectrum. It can be from 2 up to many. Obviously the more you have the more control over what you might be doing to different parts of the frequency spectrum. Graphic Equalisers These divide the frequency into 10 or 15 or often 31 fixed bands. They have sliders for every band and the sliders boost when pushed upward and attenuate or cut when slid down. In the middle there is no boost or cut ie flat. If you have more bands eg 31 then the bands are narrower and will
Version 10.1 modified 1/11/09
Page 5 of 9
© PKY Media Pty Ltd 2009
affect less frequencies at a time. You can get in and affect only a narrow range. They tend to be used more in Live PA use. They are good because a direct look at where all the sliders are sort of show out the frequency response of the graphic itself. Looking at a parametric eq wont give you any indication. You have to study where the frequencies are, how boost and cut there is and what Q there is and after all they info you can form a plot of the response. But with a graphic it is much more visible just from looking at it. In live situations time is of the essence so graphic equ’s are very good live. TIME BASED EFFECTS We are back in the Time Domain again now. Vertical axis or Y axis is still level or volume or amplitude. Horizontal axis or X axis is now time. The Time based effects include Delay, Echo, Reverb and Chorus, Flanging and Phasing. It is the time parameters that determine if we are in the longer time effects eg Delay, Echo and Reverb or the very short delays effects which include chorus, flanging and phasing. Let’s start with Delay. If we feed a signal into a delay the signal will be delayed and come out the other end a little later in time. This would make those tracks sound late compared to the others right? Yes they do but we can look at another model of this where the dry signal concept needs to learned.
Version 10.1 modified 1/11/09
Page 6 of 9
© PKY Media Pty Ltd 2009
A dry signal is just the original signal without any change or effect, it is just the main signal itself. A wet signal is used to describe the dry signal after it has been effected by the effect! So the little circle with the + inside is just a two channel mixer. The dry signal goes straight through to the output as well as going into the time based effect and coming out later ‘wet’ We feed the wet signal so our output is now a total of two things. The original dry signal and mixed with that is a wet or effected signal and in this case it’s a delay on the dry signal. This can be called slapback echo sometimes. You can adjust the time of the delay though often from milliseconds (1/1000th of a second up to several seconds even 8 or 10 seconds for extreme delays) Echo With the delay we would only hear one delay so if some said Hello you would just hear Hello, Hello. If we take some of the delayed signal and feed it back and now mix it with the dry signal that is going back into the delay then the delayed signal goes around and around more than once and it gets weaker each time it goes around so ‘Hello’ becomes ‘Hello, Hello, Hello, Hello, Hello, Hello, Hello Remember the first ‘hello’ is the dry signal and all the others are the wet signal. Block diagram of the Echo chamber. We now have a few more controls. The Time control is the length of the delay, the Feedback control is the number of repeats or how many times the echoes go around and around. And yes
Version 10.1 modified 1/11/09
Page 7 of 9
© PKY Media Pty Ltd 2009
you can set it that the echoes to get louder and louder and then the echo chamber goes mad and completely out of control. (Used very well by Jim Hendrix in ‘Electric Ladyland’) A third control called Mix is also often the case with all time based effects. The Mix control allows you to adjust the ratio of the dry signal and the wet signal. Often the dry is the loudest and the wet is a bit softer but it can be the other way around. No rules here! Processors that have this Wet Dry balance or Mix control are called side chain processors. This means the input signal is split up into two paths. The Dry goes straight through to the Mix control and out again and the dry also goes into the effect and the wet signal comes out of the effects and needs to be mixed back with the dry. Dynamics and Equalisation processors are NOT side chain effects but what we call Insert Effects processors This means there is no wet/dry mix because the dry signal just goes straight into and out of the effect. Simpler. Reverb So far we have heard delay or slap back and echo. These things are less natural but you will get echo if you stand in the right place outside between mountains and things and scream out you will hear your echoes a bit later. Reverb is a more natural thing that occurs indoors really because the walls of a room tend to bounce sound around so much that all the echoes come out a smooth reverb effect. Eg Go into a large church and clap your hands. You hear the clap plus this smooth long reverb hanging in the air. This is reverb. A very important parameter in reverb is the Reverb Time or RT60 as it sometimes called. This is the time it takes for the reverb to completely die down to nothing once the initial sound excites the room. It is measured in seconds. In a big church it could be 4 seconds but in your bathroom it might only 0.8 seconds. Clap your hands in the bathroom at home hear the clap but listen for the reverb afterwards. It will die quickly but you will still hear it though. In the Taj Mahal in India it is 28 seconds!!!! Reverb is also a Wet Dry Mix effect as well. There are other parameters as well in reverb such as Pre delay a which is the time before any reverb is heard and other parameters that can colour the reverb but the main thing to know is what it sounds like. We can have very large reverb spaces or smaller ones. Reverb is good for adding a bit of life and ambience to something that was recorded in a very dry or dead room. It is a big studio use effect mainly but reverb is also used live to add spice to vocals and other instruments.
Version 10.1 modified 1/11/09
Page 8 of 9
© PKY Media Pty Ltd 2009
Chorus, Phase and Flanging When the delays are very short we get into another group of effects and these are them Also what is different about these effects is that the delay time is being modulated or changes all the time. It sits around a nominal delay time but goes shorter and longer and back. When we do that we get these effects such as chorus. Chorus can be used to thicken up sounds eg guitars and vocals and all sorts of things. Best way to hear them is to insert them into your projects and really listen to what they are doing. Basic diagram of a Chorus/Flanger/Phasor. These are also Wet/Dry Mix side chain effects.
Version 10.1 modified 1/11/09
Page 9 of 9
© PKY Media Pty Ltd 2009
