Title: Desktop Digital Studio
Very comprehensive explanation about computers and music, MIDI all the important topics.
Author: Paul White
Publishers: Sanctuary Publishing Limited
ISBN 1-86074-324-2
Extremely helpful for computer musicians.
-
Nestor
- Posts: 6676
- Joined: Tue Mar 27, 2001 4:00 pm
- Location: Fourth Dimension Paradise, Cloud Nine!
INTRODUCTION
The Digital Desktop Studio is written for anyone who wants to record and create music with a computer rather than with traditional studio hardware. A decade ago computer music meant MIDI sequencing and perhaps a little audio editing, but now it's possible to set up a complete multitrack studio within a Macintosh or PC computer, complete with software-driven effects and synthesisers. This makes for a compact and powerful solution, but there are also complexities associated with the setting up and operation of such a system. The beginner may feel overwhelmed by the imformation presented in software manuals, but The Digital Desktop Studio helps the newcomer focus in on what's necessary to get the job done.
The Digital Desktop Studio is not a substitute for a software manual but is rather a book designed to provide a thorough overview of the process of using computers for MIDI sequencing, audio recording and audio editing, as well as for sound synthesis and effects processing. In some respects, it's like a MASTER manual, designed to help you make sense of your various hardware and software manuals. It doesn't matter whether you use a Mac or a PC, or what software you use for that matter; this book deals with concepts common to all platforms.
Various types of desktop system are explored, including those that rely on a little external hardware as well as the ones that don't, and the benefits and compromises of each type are clearly explained. There's also a simple but practical introduction to MIDI and sequencing, as well as a whole chapter exdplaining those elusive terms that crop up so often when working with computer audio, such as sample rate, bit depth, S/PDIF, AES/EBU, ASIO, VST and so on. There's also a comprehensive glossary included at the back of the book Jargon is avoided whtrever possible, but it's used where it's essential, and it's explained simply and clearly.
Desktop studios provide a wonderful opportunity to make music and have fun at the same time, and I'd like to hope you make the most of that oportunity.
The Digital Desktop Studio is written for anyone who wants to record and create music with a computer rather than with traditional studio hardware. A decade ago computer music meant MIDI sequencing and perhaps a little audio editing, but now it's possible to set up a complete multitrack studio within a Macintosh or PC computer, complete with software-driven effects and synthesisers. This makes for a compact and powerful solution, but there are also complexities associated with the setting up and operation of such a system. The beginner may feel overwhelmed by the imformation presented in software manuals, but The Digital Desktop Studio helps the newcomer focus in on what's necessary to get the job done.
The Digital Desktop Studio is not a substitute for a software manual but is rather a book designed to provide a thorough overview of the process of using computers for MIDI sequencing, audio recording and audio editing, as well as for sound synthesis and effects processing. In some respects, it's like a MASTER manual, designed to help you make sense of your various hardware and software manuals. It doesn't matter whether you use a Mac or a PC, or what software you use for that matter; this book deals with concepts common to all platforms.
Various types of desktop system are explored, including those that rely on a little external hardware as well as the ones that don't, and the benefits and compromises of each type are clearly explained. There's also a simple but practical introduction to MIDI and sequencing, as well as a whole chapter exdplaining those elusive terms that crop up so often when working with computer audio, such as sample rate, bit depth, S/PDIF, AES/EBU, ASIO, VST and so on. There's also a comprehensive glossary included at the back of the book Jargon is avoided whtrever possible, but it's used where it's essential, and it's explained simply and clearly.
Desktop studios provide a wonderful opportunity to make music and have fun at the same time, and I'd like to hope you make the most of that oportunity.
*MUSIC* The most Powerful Language in the world! *INDEED*
-
Nestor
- Posts: 6676
- Joined: Tue Mar 27, 2001 4:00 pm
- Location: Fourth Dimension Paradise, Cloud Nine!
THE UPGRADE SPIRAL
The world of computer music is anything but static, so it makes sense to buy the most powerful computer that your finances will allow. Even so, you should still be prepared to upgrade to a newer and faster model in a ywar of two in order to maintain a state-of-the-art system. Some money may be lost in the process, but probably much less than would be lost because of the depreciation on a comparable hardware-based studio system that has no upgrade path. Of course, if the new system does all that's desired of it now, why change it at all?
One reason for the existence of teh upgrade spiral is that music software undergoes frequent revisions as more features are added. Naturally these features are very attractive, but more features means that more computing power is needed to run them. Another reason is that more software is becoming available that will run a the same time as a main sequencer package. (I'm thinking now of software-based plug-ins, such as audio-processing effects and software-based instruments like synthesisers, samplers and drum machines). It's true that it's not vital to subscribe to all of these modern wonders - it's possible to buy a system that only does what's necessary and then stick with it - but the temptation to keep up with technological developmento is almost overwhelming.
Alternatively, if upgrading to the latest machine proves to be prohibitively expensive, you may prefer to content yourself with being a year ot two behind the state of the art and instead upgrade by buying bargain-priced, second-hand computers as they become availabe. My own system is more than two years behind the current technology availabe, but it's still more than powerful enough for my own requirements.
The world of computer music is anything but static, so it makes sense to buy the most powerful computer that your finances will allow. Even so, you should still be prepared to upgrade to a newer and faster model in a ywar of two in order to maintain a state-of-the-art system. Some money may be lost in the process, but probably much less than would be lost because of the depreciation on a comparable hardware-based studio system that has no upgrade path. Of course, if the new system does all that's desired of it now, why change it at all?
One reason for the existence of teh upgrade spiral is that music software undergoes frequent revisions as more features are added. Naturally these features are very attractive, but more features means that more computing power is needed to run them. Another reason is that more software is becoming available that will run a the same time as a main sequencer package. (I'm thinking now of software-based plug-ins, such as audio-processing effects and software-based instruments like synthesisers, samplers and drum machines). It's true that it's not vital to subscribe to all of these modern wonders - it's possible to buy a system that only does what's necessary and then stick with it - but the temptation to keep up with technological developmento is almost overwhelming.
Alternatively, if upgrading to the latest machine proves to be prohibitively expensive, you may prefer to content yourself with being a year ot two behind the state of the art and instead upgrade by buying bargain-priced, second-hand computers as they become availabe. My own system is more than two years behind the current technology availabe, but it's still more than powerful enough for my own requirements.
*MUSIC* The most Powerful Language in the world! *INDEED*
-
Nestor
- Posts: 6676
- Joined: Tue Mar 27, 2001 4:00 pm
- Location: Fourth Dimension Paradise, Cloud Nine!
THE REST OF THE STUDIO
The fact that computers and recording software are such good value for money can lead the user into believeing that he or she can make do with equally cheap components in the rest of the studio, but this simply isn't true. Premium items such as capacitor mics and mic pre-amps are now much more affordable than they used to be, but they still seem expensive when compared with the cost of computers, considering that they only do one job and computers are capable of doing so much. There is a school of thought that says that most of the money available to the user should go into the computer, but this is not the way to plan a good studio. I would advise to budget for a good microphone, decent cables and, if possible, a mic pre-amp-voice channel. It's possible to still use an old dynamic microphone previously used for gigs, but in most instances a capacitor mic will provide noticeably better results on vocals and acoustic instruments. A good hardware reverb is also a real benefit, if it can be patched into the system.
The same applies to monitor speakers. If they aren't accurate, it's impossible to know what a recording really sounds like. Good-quality monitor loudspeakers should be used, placed on rigid stands and set up properly, as described in chapter eight. It's not necessary to monitor loudly,but there has to be enough volume to overcome the physical noise made by the fans and drivs of the computer. If possible, computers should be placed in a ventilated cupboard, which will reduce the noise, but be careful not to block off ventilation. Lining the cupboard with an inch-thik layer of furniture foam (fireproof, of course) will help to damp out the noise.
The fact that computers and recording software are such good value for money can lead the user into believeing that he or she can make do with equally cheap components in the rest of the studio, but this simply isn't true. Premium items such as capacitor mics and mic pre-amps are now much more affordable than they used to be, but they still seem expensive when compared with the cost of computers, considering that they only do one job and computers are capable of doing so much. There is a school of thought that says that most of the money available to the user should go into the computer, but this is not the way to plan a good studio. I would advise to budget for a good microphone, decent cables and, if possible, a mic pre-amp-voice channel. It's possible to still use an old dynamic microphone previously used for gigs, but in most instances a capacitor mic will provide noticeably better results on vocals and acoustic instruments. A good hardware reverb is also a real benefit, if it can be patched into the system.
The same applies to monitor speakers. If they aren't accurate, it's impossible to know what a recording really sounds like. Good-quality monitor loudspeakers should be used, placed on rigid stands and set up properly, as described in chapter eight. It's not necessary to monitor loudly,but there has to be enough volume to overcome the physical noise made by the fans and drivs of the computer. If possible, computers should be placed in a ventilated cupboard, which will reduce the noise, but be careful not to block off ventilation. Lining the cupboard with an inch-thik layer of furniture foam (fireproof, of course) will help to damp out the noise.
*MUSIC* The most Powerful Language in the world! *INDEED*
-
Nestor
- Posts: 6676
- Joined: Tue Mar 27, 2001 4:00 pm
- Location: Fourth Dimension Paradise, Cloud Nine!
INTRODUCING MIDI
For users who have already had some experience with MIDI, it may not be necessary to read all of this chapter; but it should help to bring those coming froma moretraditional recording background up to speed without burdening them with unnnecessary complexities.
First of all, what does MIDI stand for? Answer: Musical Instrument Digital Interface. However, before starting to explain how MIDI works and howto use it, it might be more useful to ask what kind of things it can do. It should be noted that MIDI isn't only applicable to keyboard players, but because the keyboard is the best-suited instrument on which to generate MIDI information the majority of MIDI music is made using keyboards. There are practical alternatives for musicians who prefer to pluck, bow, blow or hit things, but I'll be referring to keyboards during this explanation.
THE MIDI VIRTUAL ORCHESTRA
The combination of MIDI and a suitable sequencer (which is a larga part of the trust of this book) will enable the user to record all of the different musical parts of a score from the keyboard one at a time, and then hear them playing together in perfect synchronisation, each part payed back with a chosen synth sound. The sounds can be changed even after recording has been completed, and the tempo of the finished recording can also be changed. It's also possible to experiment with musical arrangements by copying verses and choruses to new locations withing a song.
The reason why MIDI allows the user to perform all of these amazing functions is that it isn't a system for transmitting sound at all; rather, it's actually a system for transmitting instructions to devices that produce sounds. Take the analogy of a traditional musician, for whom the musical score provides the structions and the player's instrument provides the sounds. MIDI works in much the same way, except in the context of MIDI the score is electronic and the sounds that are heard are produced by bancks of synthesisers.
INTRODUCING THE MIDI SEQUENCER
At its simplest, MIDI means that a keyboard player can play several electronic instruments from a single keyboard rather than having to dash around the stage whenever a change of instrument is required. However, shortly after the introduction of MIDI came the MIDI sequencer, a special type of multitrack recorder that is capable of recording not sound but MIDIinformation.
Whatever the type of MIDI system used, a MIDI keyboard (or other suitable MIDI instrument) with MIDI In, Out, Thru sockets on the back panel will be essential. The keyboard should also have velocity sensitivity (and virtually all serious keyboards do), which means that the instrument will respond like a real instrument in that, the harder the keys are hit, the louder the notes will be. If the keyhboard doesn't have velocity sensitivity, all of the notes will sound at the same level,like an organ. The master keyboard may be DUMB (ie with no built-in sounds), or it may be a conventional keyborad synthesiser with built-in sounds.
THE MIDI LINK
Lining MIDI instruments is accomplished by menas of standard MIDI cables, which are available frommost music shops. As with computers, MIDI data is stored in a digital form, which may be though of as a kind of ultr-fast Morse code for machines. The method of MIDI connection is quite straightforwared, as we shall see shortly, but what is more important at this early stage is to appreciate precisaly what musically-useful information can be passed from one MIDI instrument or device to another. This following description covers the most important basic aspects of MIDI, but in the interests of keeping this introductiong quick and simple it is by no menas comprehensive.
THE ANATOMY OF A NOTE
When a key is depressed on a MIDI keyboard, a signal know as a Note On message is sent from the MIDI Out socket, along with a note number identifying the key. When the key is released, a Note Off message is sent. This is how the receiving MIDI instrument knows which note to play, when to play it and when to sotp playing it. Up to 128 different notes can be handled by MIDI, where each key on teh keyboard has its own note number. The loudness of the note depends on how hard the key is hit, which is really the same thing as sayhing how fast the key is pushed down. This speed, or velocity, is read by circuitry within the keyboard and used to control the volume of the sound being played.
The pitch of the note is determined by the location of the key which is pressed, although it's quite possible to transpose MIDI data before it reaches its destination. However, in order to keep things simple, let's assume that, unless otherwise stated, pressing a key results in the corresponding musical note being played.
For users who have already had some experience with MIDI, it may not be necessary to read all of this chapter; but it should help to bring those coming froma moretraditional recording background up to speed without burdening them with unnnecessary complexities.
First of all, what does MIDI stand for? Answer: Musical Instrument Digital Interface. However, before starting to explain how MIDI works and howto use it, it might be more useful to ask what kind of things it can do. It should be noted that MIDI isn't only applicable to keyboard players, but because the keyboard is the best-suited instrument on which to generate MIDI information the majority of MIDI music is made using keyboards. There are practical alternatives for musicians who prefer to pluck, bow, blow or hit things, but I'll be referring to keyboards during this explanation.
THE MIDI VIRTUAL ORCHESTRA
The combination of MIDI and a suitable sequencer (which is a larga part of the trust of this book) will enable the user to record all of the different musical parts of a score from the keyboard one at a time, and then hear them playing together in perfect synchronisation, each part payed back with a chosen synth sound. The sounds can be changed even after recording has been completed, and the tempo of the finished recording can also be changed. It's also possible to experiment with musical arrangements by copying verses and choruses to new locations withing a song.
The reason why MIDI allows the user to perform all of these amazing functions is that it isn't a system for transmitting sound at all; rather, it's actually a system for transmitting instructions to devices that produce sounds. Take the analogy of a traditional musician, for whom the musical score provides the structions and the player's instrument provides the sounds. MIDI works in much the same way, except in the context of MIDI the score is electronic and the sounds that are heard are produced by bancks of synthesisers.
INTRODUCING THE MIDI SEQUENCER
At its simplest, MIDI means that a keyboard player can play several electronic instruments from a single keyboard rather than having to dash around the stage whenever a change of instrument is required. However, shortly after the introduction of MIDI came the MIDI sequencer, a special type of multitrack recorder that is capable of recording not sound but MIDIinformation.
Whatever the type of MIDI system used, a MIDI keyboard (or other suitable MIDI instrument) with MIDI In, Out, Thru sockets on the back panel will be essential. The keyboard should also have velocity sensitivity (and virtually all serious keyboards do), which means that the instrument will respond like a real instrument in that, the harder the keys are hit, the louder the notes will be. If the keyhboard doesn't have velocity sensitivity, all of the notes will sound at the same level,like an organ. The master keyboard may be DUMB (ie with no built-in sounds), or it may be a conventional keyborad synthesiser with built-in sounds.
THE MIDI LINK
Lining MIDI instruments is accomplished by menas of standard MIDI cables, which are available frommost music shops. As with computers, MIDI data is stored in a digital form, which may be though of as a kind of ultr-fast Morse code for machines. The method of MIDI connection is quite straightforwared, as we shall see shortly, but what is more important at this early stage is to appreciate precisaly what musically-useful information can be passed from one MIDI instrument or device to another. This following description covers the most important basic aspects of MIDI, but in the interests of keeping this introductiong quick and simple it is by no menas comprehensive.
THE ANATOMY OF A NOTE
When a key is depressed on a MIDI keyboard, a signal know as a Note On message is sent from the MIDI Out socket, along with a note number identifying the key. When the key is released, a Note Off message is sent. This is how the receiving MIDI instrument knows which note to play, when to play it and when to sotp playing it. Up to 128 different notes can be handled by MIDI, where each key on teh keyboard has its own note number. The loudness of the note depends on how hard the key is hit, which is really the same thing as sayhing how fast the key is pushed down. This speed, or velocity, is read by circuitry within the keyboard and used to control the volume of the sound being played.
The pitch of the note is determined by the location of the key which is pressed, although it's quite possible to transpose MIDI data before it reaches its destination. However, in order to keep things simple, let's assume that, unless otherwise stated, pressing a key results in the corresponding musical note being played.
*MUSIC* The most Powerful Language in the world! *INDEED*
-
Nestor
- Posts: 6676
- Joined: Tue Mar 27, 2001 4:00 pm
- Location: Fourth Dimension Paradise, Cloud Nine!
INTRODUCING MIDI
For users who have already had some experience with MIDI, it may not be necessary to read all of this chapter; but it should help to bring those coming froma moretraditional recording background up to speed without burdening them with unnnecessary complexities.
First of all, what does MIDI stand for? Answer: Musical Instrument Digital Interface. However, before starting to explain how MIDI works and howto use it, it might be more useful to ask what kind of things it can do. It should be noted that MIDI isn't only applicable to keyboard players, but because the keyboard is the best-suited instrument on which to generate MIDI information the majority of MIDI music is made using keyboards. There are practical alternatives for musicians who prefer to pluck, bow, blow or hit things, but I'll be referring to keyboards during this explanation.
THE MIDI VIRTUAL ORCHESTRA
The combination of MIDI and a suitable sequencer (which is a larga part of the trust of this book) will enable the user to record all of the different musical parts of a score from the keyboard one at a time, and then hear them playing together in perfect synchronisation, each part payed back with a chosen synth sound. The sounds can be changed even after recording has been completed, and the tempo of the finished recording can also be changed. It's also possible to experiment with musical arrangements by copying verses and choruses to new locations withing a song.
The reason why MIDI allows the user to perform all of these amazing functions is that it isn't a system for transmitting sound at all; rather, it's actually a system for transmitting instructions to devices that produce sounds. Take the analogy of a traditional musician, for whom the musical score provides the structions and the player's instrument provides the sounds. MIDI works in much the same way, except in the context of MIDI the score is electronic and the sounds that are heard are produced by bancks of synthesisers.
INTRODUCING THE MIDI SEQUENCER
At its simplest, MIDI means that a keyboard player can play several electronic instruments from a single keyboard rather than having to dash around the stage whenever a change of instrument is required. However, shortly after the introduction of MIDI came the MIDI sequencer, a special type of multitrack recorder that is capable of recording not sound but MIDIinformation.
Whatever the type of MIDI system used, a MIDI keyboard (or other suitable MIDI instrument) with MIDI In, Out, Thru sockets on the back panel will be essential. The keyboard should also have velocity sensitivity (and virtually all serious keyboards do), which means that the instrument will respond like a real instrument in that, the harder the keys are hit, the louder the notes will be. If the keyhboard doesn't have velocity sensitivity, all of the notes will sound at the same level,like an organ. The master keyboard may be DUMB (ie with no built-in sounds), or it may be a conventional keyborad synthesiser with built-in sounds.
THE MIDI LINK
Lining MIDI instruments is accomplished by menas of standard MIDI cables, which are available frommost music shops. As with computers, MIDI data is stored in a digital form, which may be though of as a kind of ultr-fast Morse code for machines. The method of MIDI connection is quite straightforwared, as we shall see shortly, but what is more important at this early stage is to appreciate precisaly what musically-useful information can be passed from one MIDI instrument or device to another. This following description covers the most important basic aspects of MIDI, but in the interests of keeping this introductiong quick and simple it is by no menas comprehensive.
THE ANATOMY OF A NOTE
When a key is depressed on a MIDI keyboard, a signal know as a Note On message is sent from the MIDI Out socket, along with a note number identifying the key. When the key is released, a Note Off message is sent. This is how the receiving MIDI instrument knows which note to play, when to play it and when to sotp playing it. Up to 128 different notes can be handled by MIDI, where each key on teh keyboard has its own note number. The loudness of the note depends on how hard the key is hit, which is really the same thing as sayhing how fast the key is pushed down. This speed, or velocity, is read by circuitry within the keyboard and used to control the volume of the sound being played.
The pitch of the note is determined by the location of the key which is pressed, although it's quite possible to transpose MIDI data before it reaches its destination. However, in order to keep things simple, let's assume that, unless otherwise stated, pressing a key results in the corresponding musical note being played.
MIDI NOTE DATA
As the information determining Pitch, Note On, Note Off and Velocity exists in the form of electronic signals, it's possible to send those signals along cables to control a second MIDI instrument some distance away from the controlling keyboard. This works as follows: a small computer inside the keyboard monitors the physical motion of the keys and converts these actions to MIDI messages, which appear at the MIDI Out socket of the keyboard. If the MIDI Out of the keyboard currently playing (which is called the MASTER keyboard) is plugged into the MIDI In socket of a second MIDI instrument (which is called the SLAVE)then the slave is able to play the notes as performed on the master keyboard.
WHAT DO THOSE MIDI SOCKETS DO?
MIDI Out sends information froma controlling MIDI device (the master) to another MIDI devices that it is controlling (slaves).
MIDI In receives MIDI information, which is then passed on to the MIDI thru socket unchanged. However, if any of the incoming information is addressed to the instrument in question, it will act on that MIDI data exactly as if it were being controlled directly from a kayboard.
MIDI Thru sends a copy of the MIDI In signal to multiple destinations, thus allowing several MIDI instruments to be linked together.
THE SYNTHESISER MODULE
The ability to link a second instrument via MIDI means that the sounds of both instruments can be played by using just one keyboard. This may be convenient, but it's hardly likely to revolutionise music as we know it! However, a little further thought will reveal that the second instrument doesn't actually need a keyboard at all because all of the playing is done on the master keyborad.
This leads nicely onto the subject of MIDI modules, which are simple the sound-generating and MIDI-interfacing electronics of a keyboard instrument packaged in a rather more compact and generally less expensive box. There's also no reason why multiple mudulres shouldn't be controlled from a single master keyboard, but in order to appreciate the full implications of this it's important to understand the concept of MIDI channels. These are the menas by which certain messages are ADDRESSED so that they are recognised by certain instruments and ignored by others.
MIDI CHANNELS
In a typical master/slave MIDI system, the way in which the instruments are linked together like a daisy chain means that all of the slaves will receive the same MIDI information. The MIDI channel system was devised in order to allow the master instrument to communicate with just one specific slave without all of the others trying to play along with it. The basic idea is that MIDI note messages are tagged with an invisible address label which carries their MIDI channel number. The messages are therefore only acted upon when they are received by a MIDI instrument or device set to the same MIDI channel number. All other MIDI devices will politely ignore the message.
There are 16 MIDI channels, which are, logically enough, numbered one to 16, the idea being that MIDI information sent to channel one will only be acted upon by slave instruments also set to receive on channel one. For example, if the master keyboard were st to MIDI channel one, and then three different MIDI instruments set to receive on channels one, two and three were connected, only the instrument set to channel one would respond. The other modules would still receive the information, but the MIDI data would tell them that the information wasn't not on their channel, so they would ignore it.
By switching channels on the master keyboard, up to 16 different MIDI instruments set to 16 different channels can be addressed individually, even though they are all wired into the same system. The concept of MIDI channels will become vitally important when we move onto the subject of MIDI sequencers.
OMNI MODE WARNING
If a MIDI instrument is inadvertently set to Omni mode (an option usually found in the MIDI Set-up menu), the system won't behave as expected. Most MIDI instruments can be set to receive on any of the 16 MIDI channels, but if Omni mode is used then this will allow a MIDI instrument to respond to all incoming data, redardless of its channel. In other words, everyting that comes along the MIDI caable is played, which is roughly analogous to one member of an orchestra trying to play all parts of a score at the same time. For gerular, 16-channel operation, instruments should be set to Poly mode. There will be fore about modes later.
MORE ABOUT MODULES
So far I've described modules as being MIDI synthesisers in boxes without keyboards, and this definition is true enough, as far as it goes. However, a great many modern modules actually contain several independent sound-generating sections, each of which can be addressed on a different MIDI channel.
These sound-generating sections are often know as PARTS, because in a typical system each section can be made to play a separate musical part. For example, a 16-part multitimbral module can play back up to 16 different musical sounds at the same tiem, each part controlled via a different MIDI channel. For most purposes, a multi-part modle can be visualised as being analogous to several synthesisers sharing the same box. The same is also true of most computer souncards that have MIDI synth sections.
MULTITIMBRALITY
Modules capabla of playing two or more different parts with different sounds are said to be multitimbral, although the individual synthesiser sections they contain are rarely entirely independent of each other. For example, they will all share the same set of front-panel controls, and some parameters may affect all of the voices globally. What's more, on low-cost modules (and also budget souncards) the outputs from the various parts are usually mixed to stereo, and they then emerge from a single stereo pair of sockets. Modules also usually include effects sections, which have to be shared between the parts in some way. However, it's invariably the case that independent control may be eserted over the choice of availabe sounds (or patches, as they call themin synthspeak) that are selected, the relative levels of the different voices, the left/right pan positions and the amount of effects (such as reverberation) added to each part.
Drummachines are a special type of MIDI module, and are equipped with their own built-in sequencers which allow them to store and replay thythm patterns and complex arrangements based on permulations of those pattersn. The main difference between a standard synth patch and the way in chiw a drum machine organises its sounds is that a synthesiser tends to interpret incoming MIDI note data as different pitches of the same basic sound, whereas a drum machine produces a different drum, cymbal or precussion sound for each MIDI note. Most multitimbral synthesiser modules and computer soundcards usually have one part dedicated to drum sounds, so it's no longer essential to buy a separate drum machine, although it may prove desireble to make use of the preset rhythm patterns that drum nmachies invariably provide.
(*There is a whole lot more about MIDI staff, far to large to put here. It goes deeper and deeper through many pages. If you want more, please buy the book, it's really good*)
For users who have already had some experience with MIDI, it may not be necessary to read all of this chapter; but it should help to bring those coming froma moretraditional recording background up to speed without burdening them with unnnecessary complexities.
First of all, what does MIDI stand for? Answer: Musical Instrument Digital Interface. However, before starting to explain how MIDI works and howto use it, it might be more useful to ask what kind of things it can do. It should be noted that MIDI isn't only applicable to keyboard players, but because the keyboard is the best-suited instrument on which to generate MIDI information the majority of MIDI music is made using keyboards. There are practical alternatives for musicians who prefer to pluck, bow, blow or hit things, but I'll be referring to keyboards during this explanation.
THE MIDI VIRTUAL ORCHESTRA
The combination of MIDI and a suitable sequencer (which is a larga part of the trust of this book) will enable the user to record all of the different musical parts of a score from the keyboard one at a time, and then hear them playing together in perfect synchronisation, each part payed back with a chosen synth sound. The sounds can be changed even after recording has been completed, and the tempo of the finished recording can also be changed. It's also possible to experiment with musical arrangements by copying verses and choruses to new locations withing a song.
The reason why MIDI allows the user to perform all of these amazing functions is that it isn't a system for transmitting sound at all; rather, it's actually a system for transmitting instructions to devices that produce sounds. Take the analogy of a traditional musician, for whom the musical score provides the structions and the player's instrument provides the sounds. MIDI works in much the same way, except in the context of MIDI the score is electronic and the sounds that are heard are produced by bancks of synthesisers.
INTRODUCING THE MIDI SEQUENCER
At its simplest, MIDI means that a keyboard player can play several electronic instruments from a single keyboard rather than having to dash around the stage whenever a change of instrument is required. However, shortly after the introduction of MIDI came the MIDI sequencer, a special type of multitrack recorder that is capable of recording not sound but MIDIinformation.
Whatever the type of MIDI system used, a MIDI keyboard (or other suitable MIDI instrument) with MIDI In, Out, Thru sockets on the back panel will be essential. The keyboard should also have velocity sensitivity (and virtually all serious keyboards do), which means that the instrument will respond like a real instrument in that, the harder the keys are hit, the louder the notes will be. If the keyhboard doesn't have velocity sensitivity, all of the notes will sound at the same level,like an organ. The master keyboard may be DUMB (ie with no built-in sounds), or it may be a conventional keyborad synthesiser with built-in sounds.
THE MIDI LINK
Lining MIDI instruments is accomplished by menas of standard MIDI cables, which are available frommost music shops. As with computers, MIDI data is stored in a digital form, which may be though of as a kind of ultr-fast Morse code for machines. The method of MIDI connection is quite straightforwared, as we shall see shortly, but what is more important at this early stage is to appreciate precisaly what musically-useful information can be passed from one MIDI instrument or device to another. This following description covers the most important basic aspects of MIDI, but in the interests of keeping this introductiong quick and simple it is by no menas comprehensive.
THE ANATOMY OF A NOTE
When a key is depressed on a MIDI keyboard, a signal know as a Note On message is sent from the MIDI Out socket, along with a note number identifying the key. When the key is released, a Note Off message is sent. This is how the receiving MIDI instrument knows which note to play, when to play it and when to sotp playing it. Up to 128 different notes can be handled by MIDI, where each key on teh keyboard has its own note number. The loudness of the note depends on how hard the key is hit, which is really the same thing as sayhing how fast the key is pushed down. This speed, or velocity, is read by circuitry within the keyboard and used to control the volume of the sound being played.
The pitch of the note is determined by the location of the key which is pressed, although it's quite possible to transpose MIDI data before it reaches its destination. However, in order to keep things simple, let's assume that, unless otherwise stated, pressing a key results in the corresponding musical note being played.
MIDI NOTE DATA
As the information determining Pitch, Note On, Note Off and Velocity exists in the form of electronic signals, it's possible to send those signals along cables to control a second MIDI instrument some distance away from the controlling keyboard. This works as follows: a small computer inside the keyboard monitors the physical motion of the keys and converts these actions to MIDI messages, which appear at the MIDI Out socket of the keyboard. If the MIDI Out of the keyboard currently playing (which is called the MASTER keyboard) is plugged into the MIDI In socket of a second MIDI instrument (which is called the SLAVE)then the slave is able to play the notes as performed on the master keyboard.
WHAT DO THOSE MIDI SOCKETS DO?
MIDI Out sends information froma controlling MIDI device (the master) to another MIDI devices that it is controlling (slaves).
MIDI In receives MIDI information, which is then passed on to the MIDI thru socket unchanged. However, if any of the incoming information is addressed to the instrument in question, it will act on that MIDI data exactly as if it were being controlled directly from a kayboard.
MIDI Thru sends a copy of the MIDI In signal to multiple destinations, thus allowing several MIDI instruments to be linked together.
THE SYNTHESISER MODULE
The ability to link a second instrument via MIDI means that the sounds of both instruments can be played by using just one keyboard. This may be convenient, but it's hardly likely to revolutionise music as we know it! However, a little further thought will reveal that the second instrument doesn't actually need a keyboard at all because all of the playing is done on the master keyborad.
This leads nicely onto the subject of MIDI modules, which are simple the sound-generating and MIDI-interfacing electronics of a keyboard instrument packaged in a rather more compact and generally less expensive box. There's also no reason why multiple mudulres shouldn't be controlled from a single master keyboard, but in order to appreciate the full implications of this it's important to understand the concept of MIDI channels. These are the menas by which certain messages are ADDRESSED so that they are recognised by certain instruments and ignored by others.
MIDI CHANNELS
In a typical master/slave MIDI system, the way in which the instruments are linked together like a daisy chain means that all of the slaves will receive the same MIDI information. The MIDI channel system was devised in order to allow the master instrument to communicate with just one specific slave without all of the others trying to play along with it. The basic idea is that MIDI note messages are tagged with an invisible address label which carries their MIDI channel number. The messages are therefore only acted upon when they are received by a MIDI instrument or device set to the same MIDI channel number. All other MIDI devices will politely ignore the message.
There are 16 MIDI channels, which are, logically enough, numbered one to 16, the idea being that MIDI information sent to channel one will only be acted upon by slave instruments also set to receive on channel one. For example, if the master keyboard were st to MIDI channel one, and then three different MIDI instruments set to receive on channels one, two and three were connected, only the instrument set to channel one would respond. The other modules would still receive the information, but the MIDI data would tell them that the information wasn't not on their channel, so they would ignore it.
By switching channels on the master keyboard, up to 16 different MIDI instruments set to 16 different channels can be addressed individually, even though they are all wired into the same system. The concept of MIDI channels will become vitally important when we move onto the subject of MIDI sequencers.
OMNI MODE WARNING
If a MIDI instrument is inadvertently set to Omni mode (an option usually found in the MIDI Set-up menu), the system won't behave as expected. Most MIDI instruments can be set to receive on any of the 16 MIDI channels, but if Omni mode is used then this will allow a MIDI instrument to respond to all incoming data, redardless of its channel. In other words, everyting that comes along the MIDI caable is played, which is roughly analogous to one member of an orchestra trying to play all parts of a score at the same time. For gerular, 16-channel operation, instruments should be set to Poly mode. There will be fore about modes later.
MORE ABOUT MODULES
So far I've described modules as being MIDI synthesisers in boxes without keyboards, and this definition is true enough, as far as it goes. However, a great many modern modules actually contain several independent sound-generating sections, each of which can be addressed on a different MIDI channel.
These sound-generating sections are often know as PARTS, because in a typical system each section can be made to play a separate musical part. For example, a 16-part multitimbral module can play back up to 16 different musical sounds at the same tiem, each part controlled via a different MIDI channel. For most purposes, a multi-part modle can be visualised as being analogous to several synthesisers sharing the same box. The same is also true of most computer souncards that have MIDI synth sections.
MULTITIMBRALITY
Modules capabla of playing two or more different parts with different sounds are said to be multitimbral, although the individual synthesiser sections they contain are rarely entirely independent of each other. For example, they will all share the same set of front-panel controls, and some parameters may affect all of the voices globally. What's more, on low-cost modules (and also budget souncards) the outputs from the various parts are usually mixed to stereo, and they then emerge from a single stereo pair of sockets. Modules also usually include effects sections, which have to be shared between the parts in some way. However, it's invariably the case that independent control may be eserted over the choice of availabe sounds (or patches, as they call themin synthspeak) that are selected, the relative levels of the different voices, the left/right pan positions and the amount of effects (such as reverberation) added to each part.
Drummachines are a special type of MIDI module, and are equipped with their own built-in sequencers which allow them to store and replay thythm patterns and complex arrangements based on permulations of those pattersn. The main difference between a standard synth patch and the way in chiw a drum machine organises its sounds is that a synthesiser tends to interpret incoming MIDI note data as different pitches of the same basic sound, whereas a drum machine produces a different drum, cymbal or precussion sound for each MIDI note. Most multitimbral synthesiser modules and computer soundcards usually have one part dedicated to drum sounds, so it's no longer essential to buy a separate drum machine, although it may prove desireble to make use of the preset rhythm patterns that drum nmachies invariably provide.
(*There is a whole lot more about MIDI staff, far to large to put here. It goes deeper and deeper through many pages. If you want more, please buy the book, it's really good*)
*MUSIC* The most Powerful Language in the world! *INDEED*
-
Nestor
- Posts: 6676
- Joined: Tue Mar 27, 2001 4:00 pm
- Location: Fourth Dimension Paradise, Cloud Nine!
THE TECHNICAL STUFF
It would be nice to be able to tell you ahta making music on a computer is as simple as turning it on, hitting Record and recording it, but before you can get to that stage there are things that you'll need to know. While you can certainly drive a car without knowing anyting about mechanics, the same isn't true of computers. You don't have to know everyting, but you won't get far without some basic knowledge of thwat the hardware and software is up to. Bying a ready-configured system flattens the learning curve considerable, and unless you-re a computer expert I strongly recommend that you follow this route. Even so, you'll still need a certain amont of background information so that you can make sense of what's going on. This chapter explains some of the more common concepts and standards that you'll encounter along the way,but don't feel you have to read it all in one go. Treat it as a source of referente if that suits you better.
It would be nice to be able to tell you ahta making music on a computer is as simple as turning it on, hitting Record and recording it, but before you can get to that stage there are things that you'll need to know. While you can certainly drive a car without knowing anyting about mechanics, the same isn't true of computers. You don't have to know everyting, but you won't get far without some basic knowledge of thwat the hardware and software is up to. Bying a ready-configured system flattens the learning curve considerable, and unless you-re a computer expert I strongly recommend that you follow this route. Even so, you'll still need a certain amont of background information so that you can make sense of what's going on. This chapter explains some of the more common concepts and standards that you'll encounter along the way,but don't feel you have to read it all in one go. Treat it as a source of referente if that suits you better.
*MUSIC* The most Powerful Language in the world! *INDEED*
-
Nestor
- Posts: 6676
- Joined: Tue Mar 27, 2001 4:00 pm
- Location: Fourth Dimension Paradise, Cloud Nine!
DIGITAL BASICS
All digital audio equipment designed to accept analogue signals - such as the outputs from microphones - must convert the analogue input into a digital format before it can process it or record it. In the case of sound, the analogue signal voltage varies in proportion to changes in air pressure. A rapidly vibrating string, for example, will create equally rapid fluctuations in air pressure which a microphone can convert to variations in voltage.
Digital systems use binary numbers to measure amounts, and in the case of the signal froma microphone the digital data measures the changing analogue voltage by using a successin of ones and zeros. These digits are represented in the circuit by the presence or absence of a nominally-fixed voltage. In essence, converting an analogue signal into digital information involves measuring the analogue voltage at regular intervals and then turning these measurements into a series of binary numbers.
Sound needs to be sampled and measured several tens of thousands of times per second if the end result is going to be of CD quality. If you have enough instantaneous measurements per second, the original sound can be accurately recreated up to the highest frequency limit of human hearing.
All digital audio equipment designed to accept analogue signals - such as the outputs from microphones - must convert the analogue input into a digital format before it can process it or record it. In the case of sound, the analogue signal voltage varies in proportion to changes in air pressure. A rapidly vibrating string, for example, will create equally rapid fluctuations in air pressure which a microphone can convert to variations in voltage.
Digital systems use binary numbers to measure amounts, and in the case of the signal froma microphone the digital data measures the changing analogue voltage by using a successin of ones and zeros. These digits are represented in the circuit by the presence or absence of a nominally-fixed voltage. In essence, converting an analogue signal into digital information involves measuring the analogue voltage at regular intervals and then turning these measurements into a series of binary numbers.
Sound needs to be sampled and measured several tens of thousands of times per second if the end result is going to be of CD quality. If you have enough instantaneous measurements per second, the original sound can be accurately recreated up to the highest frequency limit of human hearing.
*MUSIC* The most Powerful Language in the world! *INDEED*
-
Nestor
- Posts: 6676
- Joined: Tue Mar 27, 2001 4:00 pm
- Location: Fourth Dimension Paradise, Cloud Nine!
This process of measuring and digitising minute sections of the input signal is known as sampling. Each sample is a discrete measurement taken at one point in time, and the more often these measurements are made, the more accurately the curves of the original analogue signal are followed.
Sampling theory states that you must take a sample at a minimum of twice the frequency of the highest frequency that you are likely to encounter if the output is to be reconstructed accurately. If the sampling frequency is less than twice the highest frequency, additional frequencies will be introduced based on the summand difference between the sampling frequency and the audio frequency. These frequencies won’t have been present in the original signal and so will sound musically unpleasant.
The creation of unwanted frequencies by sampling at a rate that is too low is known as aliasing, and to prevent this it is necessary to filter out any frequencies in the original signal that are above half the sampling frequency. Because no filters are perfect, the sampling rate must be a little higher than twice the maximum audio frequency, which is why for an audio bandwidth of 20kHz a minimum sampling frequency of 44.1kHz is used. If you’re thinking that 44.1kHz is an odd kind of number (why not 45kHz or 50kHz?) then you’re right; but, as is so often the case, the reason are historic.
The other major factor affecting audio quality is the accuracy with which individual samples are measured. Basically, the more digital bits that are used to represent each sample, the more accurate the measurement. CDs and DAT tapes use 16-bit sampling, although many digital multitrack recorders and signal processors now use 20- or 24-bit conversion.
Digital sampling proceeds in steps – there are no halves or thirds of a bit and the number of steps depends on the resolution of the analogue-to-digital converter used. Eight bits will provide only two to the power of eight steps, which works out as 256. That means that your loudest signal could have 256 steps but that your quieter sounds will have considerably fewer. This allows for a rather poor level of resolution, and causes what’s known as Quantisation distortion, an effect that gets more noticeable at lower signal levels.
Quantisation distortion actually sounds like nose. The main difference is that it disappears in the absence of a signal, unlike most other sources of noise. Nowadays, eight-bit sound is rarely used, other than in some undemanding computer applications. Using more bits allows for a vast improvement in resolution, and most current digital processors use 16-, 20- and 24-bit resolution converters. Each bit in a linear sampling system equates a dynamic range of 48dB at best – about as noisy as a cheap cassette recorder. 16-bit resolution gives a maximum dynamic range of 96dB, while 20- and 24-bit systems can still give practical dynamic ranges in excess of 120dB.
So far, then, we know that an audio signal can be represented by a series of numbers, but these numbers also have to be recorded and replayed with exactly the same timing relationship if the result is to be accurate. This is where sample rates come in.
Sampling theory states that you must take a sample at a minimum of twice the frequency of the highest frequency that you are likely to encounter if the output is to be reconstructed accurately. If the sampling frequency is less than twice the highest frequency, additional frequencies will be introduced based on the summand difference between the sampling frequency and the audio frequency. These frequencies won’t have been present in the original signal and so will sound musically unpleasant.
The creation of unwanted frequencies by sampling at a rate that is too low is known as aliasing, and to prevent this it is necessary to filter out any frequencies in the original signal that are above half the sampling frequency. Because no filters are perfect, the sampling rate must be a little higher than twice the maximum audio frequency, which is why for an audio bandwidth of 20kHz a minimum sampling frequency of 44.1kHz is used. If you’re thinking that 44.1kHz is an odd kind of number (why not 45kHz or 50kHz?) then you’re right; but, as is so often the case, the reason are historic.
The other major factor affecting audio quality is the accuracy with which individual samples are measured. Basically, the more digital bits that are used to represent each sample, the more accurate the measurement. CDs and DAT tapes use 16-bit sampling, although many digital multitrack recorders and signal processors now use 20- or 24-bit conversion.
Digital sampling proceeds in steps – there are no halves or thirds of a bit and the number of steps depends on the resolution of the analogue-to-digital converter used. Eight bits will provide only two to the power of eight steps, which works out as 256. That means that your loudest signal could have 256 steps but that your quieter sounds will have considerably fewer. This allows for a rather poor level of resolution, and causes what’s known as Quantisation distortion, an effect that gets more noticeable at lower signal levels.
Quantisation distortion actually sounds like nose. The main difference is that it disappears in the absence of a signal, unlike most other sources of noise. Nowadays, eight-bit sound is rarely used, other than in some undemanding computer applications. Using more bits allows for a vast improvement in resolution, and most current digital processors use 16-, 20- and 24-bit resolution converters. Each bit in a linear sampling system equates a dynamic range of 48dB at best – about as noisy as a cheap cassette recorder. 16-bit resolution gives a maximum dynamic range of 96dB, while 20- and 24-bit systems can still give practical dynamic ranges in excess of 120dB.
So far, then, we know that an audio signal can be represented by a series of numbers, but these numbers also have to be recorded and replayed with exactly the same timing relationship if the result is to be accurate. This is where sample rates come in.
-
Nestor
- Posts: 6676
- Joined: Tue Mar 27, 2001 4:00 pm
- Location: Fourth Dimension Paradise, Cloud Nine!
SAMPLING THEORY
This process of measuring and digitising minute sections of the input signal is known as sampling. Each sample is a discrete measurement taken at one point in time, and the more often these measurements are made, the more accurately the curves of the original analogue signal are followed.
Sampling theory states that you must take a sample at a minimum of twice the frequency of the highest frequency that you are likely to encounter if the output is to be reconstructed accurately. If the sampling frequency is less than twice the highest frequency, additional frequencies will be introduced based on the summand difference between the sampling frequency and the audio frequency. These frequencies won’t have been present in the original signal and so will sound musically unpleasant.
The creation of unwanted frequencies by sampling at a rate that is too low is known as aliasing, and to prevent this it is necessary to filter out any frequencies in the original signal that are above half the sampling frequency. Because no filters are perfect, the sampling rate must be a little higher than twice the maximum audio frequency, which is why for an audio bandwidth of 20kHz a minimum sampling frequency of 44.1kHz is used. If you’re thinking that 44.1kHz is an odd kind of number (why not 45kHz or 50kHz?) then you’re right; but, as is so often the case, the reason are historic.
The other major factor affecting audio quality is the accuracy with which individual samples are measured. Basically, the more digital bits that are used to represent each sample, the more accurate the measurement. CDs and DAT tapes use 16-bit sampling, although many digital multitrack recorders and signal processors now use 20- or 24-bit conversion.
Digital sampling proceeds in steps – there are no halves or thirds of a bit and the number of steps depends on the resolution of the analogue-to-digital converter used. Eight bits will provide only two to the power of eight steps, which works out as 256. That means that your loudest signal could have 256 steps but that your quieter sounds will have considerably fewer. This allows for a rather poor level of resolution, and causes what’s known as Quantisation distortion, an effect that gets more noticeable at lower signal levels.
Quantisation distortion actually sounds like nose. The main difference is that it disappears in the absence of a signal, unlike most other sources of noise. Nowadays, eight-bit sound is rarely used, other than in some undemanding computer applications. Using more bits allows for a vast improvement in resolution, and most current digital processors use 16-, 20- and 24-bit resolution converters. Each bit in a linear sampling system equates a dynamic range of 48dB at best – about as noisy as a cheap cassette recorder. 16-bit resolution gives a maximum dynamic range of 96dB, while 20- and 24-bit systems can still give practical dynamic ranges in excess of 120dB.
So far, then, we know that an audio signal can be represented by a series of numbers, but these numbers also have to be recorded and replayed with exactly the same timing relationship if the result is to be accurate. This is where sample rates come in.
This process of measuring and digitising minute sections of the input signal is known as sampling. Each sample is a discrete measurement taken at one point in time, and the more often these measurements are made, the more accurately the curves of the original analogue signal are followed.
Sampling theory states that you must take a sample at a minimum of twice the frequency of the highest frequency that you are likely to encounter if the output is to be reconstructed accurately. If the sampling frequency is less than twice the highest frequency, additional frequencies will be introduced based on the summand difference between the sampling frequency and the audio frequency. These frequencies won’t have been present in the original signal and so will sound musically unpleasant.
The creation of unwanted frequencies by sampling at a rate that is too low is known as aliasing, and to prevent this it is necessary to filter out any frequencies in the original signal that are above half the sampling frequency. Because no filters are perfect, the sampling rate must be a little higher than twice the maximum audio frequency, which is why for an audio bandwidth of 20kHz a minimum sampling frequency of 44.1kHz is used. If you’re thinking that 44.1kHz is an odd kind of number (why not 45kHz or 50kHz?) then you’re right; but, as is so often the case, the reason are historic.
The other major factor affecting audio quality is the accuracy with which individual samples are measured. Basically, the more digital bits that are used to represent each sample, the more accurate the measurement. CDs and DAT tapes use 16-bit sampling, although many digital multitrack recorders and signal processors now use 20- or 24-bit conversion.
Digital sampling proceeds in steps – there are no halves or thirds of a bit and the number of steps depends on the resolution of the analogue-to-digital converter used. Eight bits will provide only two to the power of eight steps, which works out as 256. That means that your loudest signal could have 256 steps but that your quieter sounds will have considerably fewer. This allows for a rather poor level of resolution, and causes what’s known as Quantisation distortion, an effect that gets more noticeable at lower signal levels.
Quantisation distortion actually sounds like nose. The main difference is that it disappears in the absence of a signal, unlike most other sources of noise. Nowadays, eight-bit sound is rarely used, other than in some undemanding computer applications. Using more bits allows for a vast improvement in resolution, and most current digital processors use 16-, 20- and 24-bit resolution converters. Each bit in a linear sampling system equates a dynamic range of 48dB at best – about as noisy as a cheap cassette recorder. 16-bit resolution gives a maximum dynamic range of 96dB, while 20- and 24-bit systems can still give practical dynamic ranges in excess of 120dB.
So far, then, we know that an audio signal can be represented by a series of numbers, but these numbers also have to be recorded and replayed with exactly the same timing relationship if the result is to be accurate. This is where sample rates come in.
awesome, nestor. hope you had a soft version of this to copy paste, and didn't type all this in by hand for our benefit!!
on the subject of useful learning resources, I found "Computer Music Tutorial" by Curtis Roads really helpful. (Or should say, am still finding - at 12 cm thick, its one of those "read for a few evenings, set aside, pick it up again week or two later" books. I'm still at it.)
Hardly a practical tutorial in any sense, instead this book is a readable theoretical overview of a very wide range of subjects I wanted to know more about. Read the table of contents and some excerpts on Amazon. Perhaps not for everyone, but I am finding it immensely helpful in understanding what I really dealing with. not cheap, but it's shed a lot of light onto a lot of formerly baffling concepts for me.
http://www.amazon.com/exec/obidos/ASIN/ ... 24-9462256
cheers
<font size=-1>[ This Message was edited by: dblbass on 2002-04-10 09:01 ]</font>
on the subject of useful learning resources, I found "Computer Music Tutorial" by Curtis Roads really helpful. (Or should say, am still finding - at 12 cm thick, its one of those "read for a few evenings, set aside, pick it up again week or two later" books. I'm still at it.)
Hardly a practical tutorial in any sense, instead this book is a readable theoretical overview of a very wide range of subjects I wanted to know more about. Read the table of contents and some excerpts on Amazon. Perhaps not for everyone, but I am finding it immensely helpful in understanding what I really dealing with. not cheap, but it's shed a lot of light onto a lot of formerly baffling concepts for me.
http://www.amazon.com/exec/obidos/ASIN/ ... 24-9462256
cheers
<font size=-1>[ This Message was edited by: dblbass on 2002-04-10 09:01 ]</font>
dblbass, I agree, Computer Music Tutorial is an excellent book, although not going into too much detail for each section, it covers a lot of ground including the various synthesis methods, etc and makes for excellent 'pleasure' reading... Also highly recommended from me as well.
BTW - thanks Nestor for all the info! Thats a fun read as well.
BTW - thanks Nestor for all the info! Thats a fun read as well.
-
Nestor
- Posts: 6676
- Joined: Tue Mar 27, 2001 4:00 pm
- Location: Fourth Dimension Paradise, Cloud Nine!
EQUALISERS
The word equiliser was originally employed to describe a filter used to compensate for imperfections in the microphone signal chain, but in another term for tone control. While early studio equipment, and to a greater extent, the analogue telephone lines for which EQ was first developed, need a lot of corrective EQ to make them sound natural, modern recording equipment is capable of storing and reproducing sound that is virtually identical to the original. Nevertheless, the original sound isn't always what we want to hear, so EQ has evolved to take more of a creative role.
Equalisers are based around electronic circuits known as filters. Strictly speaking, a filter is a device that removes something, but in the context of active equaliser circuits, filters can be arranged within special circuit configurations so that they boost as well as cut.
Early equalisers were very simple affairs, usually no more than a single tone control offering only varying degrees of treble or bass cut. The first SERIOUS active equaliser to find popular acceptance was designed by British electronic engineer Peter J Baxandall and comprised separate bass and treble controls - both of which could provide either cut or boost. This meant that the controls had to be set to their centre positions if the signal was to pass through unafected.
The word equiliser was originally employed to describe a filter used to compensate for imperfections in the microphone signal chain, but in another term for tone control. While early studio equipment, and to a greater extent, the analogue telephone lines for which EQ was first developed, need a lot of corrective EQ to make them sound natural, modern recording equipment is capable of storing and reproducing sound that is virtually identical to the original. Nevertheless, the original sound isn't always what we want to hear, so EQ has evolved to take more of a creative role.
Equalisers are based around electronic circuits known as filters. Strictly speaking, a filter is a device that removes something, but in the context of active equaliser circuits, filters can be arranged within special circuit configurations so that they boost as well as cut.
Early equalisers were very simple affairs, usually no more than a single tone control offering only varying degrees of treble or bass cut. The first SERIOUS active equaliser to find popular acceptance was designed by British electronic engineer Peter J Baxandall and comprised separate bass and treble controls - both of which could provide either cut or boost. This meant that the controls had to be set to their centre positions if the signal was to pass through unafected.
*MUSIC* The most Powerful Language in the world! *INDEED*
-
Nestor
- Posts: 6676
- Joined: Tue Mar 27, 2001 4:00 pm
- Location: Fourth Dimension Paradise, Cloud Nine!
WHAT EQ DOES
How does an equaliser work? If you think of how an ordinary gain or volume control operates, it turns the whole signal up or down in level without affecting the tonal balance. All the frequencies present in the input signal are increased or decreased by the same amount, so that other than in level, what comes out is exactly the same as the signal that went in. A circuit that changes nothing but the level of a signal is said to have a flat frequency response. It's technically impossible to make a circuit that has an infinitely wide frequency response, but in the context of audio, a circuit that is flat between the upper and lower limits of human hearing (generally 20Hz to 20kHz, is said to have a flat response.
Unlike the level or gain control,the equaliser is designed specifically to affect the level of some frequencies more than others. A typical treble control, for example, raises or low4ers the level of the high frequencies while leaving the low frequency level virtually unchanged. The reason I say VIRTUALLY is that if you were to plot out the gain versus frequency characteristics of an equaliser on a graph, you'd find the grph followed a slope or curve - you don't see a sudden step at the frequendcies above 5kHz by so many dBs doesn't simply leave frequencies below that point unaffected - the effect is progressive - but the further below 5kHz you go, the less those frequencies are affected.
Notice that the cut and boost curves eventyally flatten out, rather than continuing to rise indefinitely. This particular equaliser characteristic is described as a SHELVING response, because the cut or boost part of the curve eventually forms a flat shelf. It is possible to make an equaliser whose degree of boost continues to increase with frequency, but this is usually undesirable.
In the case of a treble control for example, it would mean that the higher the frequency the more boost you'd have, and so far more boost than necessary would be applied at very high and even supersonic frequencies. This would lead to an increase in high frequency noise and the possibility of the circuitry baing overloaded to the point of distortion by frequencies too high even to hear. Conversely, a bass equaliser with similar characteristics would give more boost, the lower the frequency, and so give rise to rumble and hum problems, which can overload the electronics very easily. This is why most equalisers used in studio work tend to be either shelving or BANDPASS types.
The filter slope of an equaliser is usually specified in dBs per octave, and for music applications, slopes of 6dB/octave and 12dB per octave are common. The greater the number of dBs per octave, the sharper the filter slope and the less frequencies outside the band are affected. If it is required to filter out subsonic signals using a cut filter, say below 50Hz, or very high frequencies above 20kHz or so, then even sharper slopes of 24dB per octave are used and the filters are non-shelving. Such filters are called high-pass and low-pass, as they allow high and low frequencies, respectively, to be passed and are often found on more comoprehensive mixing consoles, mic preamps and sometimes on stand-alone equalisers.
How does an equaliser work? If you think of how an ordinary gain or volume control operates, it turns the whole signal up or down in level without affecting the tonal balance. All the frequencies present in the input signal are increased or decreased by the same amount, so that other than in level, what comes out is exactly the same as the signal that went in. A circuit that changes nothing but the level of a signal is said to have a flat frequency response. It's technically impossible to make a circuit that has an infinitely wide frequency response, but in the context of audio, a circuit that is flat between the upper and lower limits of human hearing (generally 20Hz to 20kHz, is said to have a flat response.
Unlike the level or gain control,the equaliser is designed specifically to affect the level of some frequencies more than others. A typical treble control, for example, raises or low4ers the level of the high frequencies while leaving the low frequency level virtually unchanged. The reason I say VIRTUALLY is that if you were to plot out the gain versus frequency characteristics of an equaliser on a graph, you'd find the grph followed a slope or curve - you don't see a sudden step at the frequendcies above 5kHz by so many dBs doesn't simply leave frequencies below that point unaffected - the effect is progressive - but the further below 5kHz you go, the less those frequencies are affected.
Notice that the cut and boost curves eventyally flatten out, rather than continuing to rise indefinitely. This particular equaliser characteristic is described as a SHELVING response, because the cut or boost part of the curve eventually forms a flat shelf. It is possible to make an equaliser whose degree of boost continues to increase with frequency, but this is usually undesirable.
In the case of a treble control for example, it would mean that the higher the frequency the more boost you'd have, and so far more boost than necessary would be applied at very high and even supersonic frequencies. This would lead to an increase in high frequency noise and the possibility of the circuitry baing overloaded to the point of distortion by frequencies too high even to hear. Conversely, a bass equaliser with similar characteristics would give more boost, the lower the frequency, and so give rise to rumble and hum problems, which can overload the electronics very easily. This is why most equalisers used in studio work tend to be either shelving or BANDPASS types.
The filter slope of an equaliser is usually specified in dBs per octave, and for music applications, slopes of 6dB/octave and 12dB per octave are common. The greater the number of dBs per octave, the sharper the filter slope and the less frequencies outside the band are affected. If it is required to filter out subsonic signals using a cut filter, say below 50Hz, or very high frequencies above 20kHz or so, then even sharper slopes of 24dB per octave are used and the filters are non-shelving. Such filters are called high-pass and low-pass, as they allow high and low frequencies, respectively, to be passed and are often found on more comoprehensive mixing consoles, mic preamps and sometimes on stand-alone equalisers.
*MUSIC* The most Powerful Language in the world! *INDEED*