MIDI Basics (Part 1)
Confused by MIDI - or even been hiding from it in the hope it might go away? Your prayers are answered and your fears dispelled by Bob O'Donnell in the first of a series that takes MIDI from the top.
Four years since its arrival, MIDI is being used by more musicians in more situations than ever before - yet many users remain mystified by it. In the first part of a major new series, we explain what MIDI is, why it came about, and the sort of information it can convey.
MIDI, OR MUSICAL Instrument Digital Interface, has brought about a revolution in the performance, composition and even understanding of modern music. Really it has. If you don't believe me, just ask any of the other 3.141 million writers who have used minor variations on that sentence in one of their articles...
I'm not just talking about pop or avant-garde electronic music. Everybody wants to jump onto the MIDI bandwagon - and rightfully so, because we all know that in so doing, we can use the opportunities presented by what is an industry standard specification for the serial transmission and reception of musical data - which is what MIDI basically is - to our own benefit.
Well, actually, what we all want to do is write, record and perform music. But as a very powerful tool, MIDI can certainly help.
To make the most of a tool you obviously have to understand it, though, and if you're relatively new to the wonders of modern musical instruments, MIDI may be a bit confusing.
If you have spent any time at all with modern musical instruments, you'll know that most of them have a couple of sockets on the back of them that say "MIDI". You may also know that if two keyboards are connected together via this connection, one will be able to control the other remotely. In other words, if you play a C chord and move the pitch wheel on one, a similar effect will occur on the other. The reason for this is that the first keyboard is sending out messages over MIDI - that is, through the cable - which tell the other keyboard exactly what to do.
This is the basic premise of MIDI: sending out messages, or data, which allow two or more musical instruments to communicate with each other. Most of the confusion surrounding MIDI stems from a misunderstanding of how instruments send this information, and exactly what this information comprises.
Part of the problem is that along with the enormous growth in the popularity of MIDI-equipped instruments, has come a corresponding increase in the amount of information and misinformation available on what MIDI is, what it does, what it's for and how it works.
That's where this series of articles comes in. Its goal is to provide you with useful information that will allow you to make the most of the equipment you currently have and, just as importantly, help you make buying decisions in the future. The articles will address some of the important details of what MIDI can do (particularly this first segment), but the primary emphasis will be on practical applications.
The world of MIDI is a strange and exciting place for many musicians, and it's well worth further exploration. If you haven't made the plunge yet and you're one of those people who thinks our magazine occasionally reads like a foreign mystery novel - believe me, you're not alone - or if you're a little unsure about why that System Exclusive information is included in the back of your synth's owner's manual, then sit back, pop open a can of your favourite beverage, and read. You'll be amazed at what a simple, informative pleasure that can be.
SOME OF YOU probably expect a fast, easy answer to the question "What is MIDI?", so I won't disappoint you. MIDI is a hardware and software communications standard that allows various pieces of equipment to share information with one another (regardless of their manufacturer) and operate as a system. The information is divided into a number of different types of messages, which are carried over special cables that end in five-pin DIN plugs at a rate of 31.25Kbaud, or 31,250 bits per second. These cables plug into the MIDI In, Out and Thru jacks on the back of the various pieces of equipment, or properly equipped personal computers.
Like most simple answers, however, the above paragraph is nowhere near the whole story.
First of all, to understand how a computer-type interface standard can exist between musical instruments, you need to know that most MIDI-equipped devices are basically computers with the dedicated function (s) of producing or somehow manipulating sounds and musical data. They all contain a microprocessor of some sort to generate and/or respond to MIDI data.
Second, the most important thing to remember about MIDI is not the technical data itself, but the fact that it's a communication standard. Communication is inherently a process that is accomplished by more than one person - or, in this case, by more than one box. Which means that if you only have one MIDI-equipped machine, the MIDI part of it isn't going to do you any good. In other words, as in life, it takes two (or more) to tango in the land of MIDI.
Now, this doesn't mean you have to connect two musical instruments together; it could mean that you use a single MIDI keyboard and a computer equipped with a MIDI connection, or perhaps a keyboard connected to a signal processor of some description. But whatever the combination, you must have at least two devices to take advantage of MIDI.
Finally, another point to be aware of is that even though MIDI is a standard for the entire music industry, that doesn't mean every unit implements it the same way - that is, they don't necessarily share all of the same MIDI features. As you will see in forthcoming articles, different types of MIDI-equipped machines (synths, drum boxes, and so on) generate and respond to a variety of MIDI messages according to their capabilities.
"Part of the confusion about MIDI actually has to do with grammar, because MIDI is an acronym that's used as a noun, an adjective and a verb."
Consequently, certain machines will have one type of MIDI implementation, while others may have a completely different one - though generally, a bit of overlap does occur. Even units within the same category can have different implementations of MIDI, because manufacturers and designers have different ideas about what they feel is important, so they'll include some features and not others.
This can be a source of confusion and frustration even among experienced MIDI users, and it's particularly important for MIDI newcomers to be aware of what the capabilities are for each of the devices being connected, or MIDI'd together. (Part of the confusion about MIDI, believe it or not, actually has to do with grammar, because MIDI is an acronym that's used as a noun, an adjective and a verb, sometimes all within the same sentence. Of course, you won't catch me doing that.) Manufacturers usually provide this information in the form of a MIDI implementation chart at the end of a machine's user manual.
THE PERENNIAL PHILOSOPHICAL question acts as a double-edged sword in reference to MIDI, as its meaning changes depending on which word you choose to emphasise. If you choose "why", then the answer is to take full advantage of the facilities of the equipment you use and own. Until we see the day of inexpensive music workstations that include everything we could possibly want for the performance and composition of music - and I'm not holding my breath - different pieces of equipment will continue to perform the specific, individual functions that they were designed to do. MIDI, in turn, will allow these various components to work together as a coherent whole, a system.
If, on the other hand, you choose to emphasise "MIDI", the question takes on an entirely different character, and the answer has more to do with history than philosophy.
The history runs something like this. Prior to the adoption of MIDI as an industry standard, a number of manufacturers had started to develop their own communication interfaces, but many of these were rather clumsy and each one worked only with that particular manufacturers' products. In other words, trying to connect a Yamaha synth to a Roland sequencer was a no-no.
Most of these early communication systems were based on the same kind of analogue control voltages that synthesisers of the time (pre-1983) used to create and manipulate their sounds. Others, like Roland's Digital Communication Buss (DCB) and the Oberheim System, used digital signals. But they all shared in the fact that their scope was rather limited.
MIDI, on the other hand, has grown into a wide-ranging means of communicating a great deal and a great variety of information. Though it started as a relatively simple proposition from Sequential's Dave Smith for a Universal Synthesiser Interface (USI), the collective knowledge of many American and Japanese engineers has transformed MIDI into a flexible, powerful tool.
SO NOW THE question is, what exactly is this "wide-ranging" information that MIDI data consists of? Well, as mentioned above, MIDI data is divided into different types of messages - all of which are sent from the instrument's microprocessor through the MIDI cables as a "stream" of information. The two basic types are Channel messages and System messages, and these are further broken down into five categories: Channel messages can be either "Voice" messages or "Mode" messages, while System messages are referred to as "System Common", "System Real Time", or "System Exclusive". Each of these serves a different purpose, and each allows various components of a connected MIDI system to communicate with one another.
But before I start going into the dirty details, I need to explain what this Channel and System rubbish means. MIDI data serves two purposes: one is to provide specific information to specific machines within a system, the other is to provide general information for the entire system. Channel and System messages, respectively - surprise, surprise - perform these functions.
Channels are necessary, in case you're wondering, because most systems have several things going on at once, and if you've got different units performing these different functions, you need to send them different messages. When I start explaining specific MIDI applications, the usefulness of these channels will become obvious. Trust me.
So now, you ask, how does MIDI separate the two? Magic? Well, not really. What actually happens is that Channel messages are assigned to one of the 16 specific channels that are included in MIDI, while System messages are not assigned to any channel at all. As a result, up to 16 independent "conversations" involving Channel messages but only one using System messages can occur at the same time. Amazingly enough, all of this information can be sent over a single MIDI cable.
To give you a better idea of how this works, I'll use the traditional analogy of a television that everyone else uses. Just as a TV receives all the channels that are broadcast but only displays the one you've selected, so all the MIDI devices connected into a system receive every Channel message that is sent over a MIDI cable, but only respond to the ones on the MIDI channel they've been set to (as long as the instrument is operating in the proper mode, which I'll explain shortly).
If you've got one of those fancy new TVs that can display more than one channel at a time, MIDI has an equivalent too; some synths can respond to more than one channel at a time, though we'll worry about them later. System messages can be thought of as Party Political Broadcasts; they're always on every channel, no matter what you try to do about them.
"Each of the five MIDI messages serves a different purpose, and allows various components of a system to communicate with one another.
Now back to the details. Most Channel messages actually consist of Voice data, which describe the notes or voices being played. These descriptions are not of the particular patches used, but instead what notes are being played, how hard they're played (velocity), whether or not they've been affected by pitch-bend or the mod wheel, and so on. Channel Mode messages, on the other hand, instruct the connected instruments to operate in one of four possible MIDI Modes, and these modes determine how the instrument will respond to the Voice messages. Remember that last statement, because this mode business can get confusing.
Not all instruments can operate in all Modes, however; again, it depends on the MIDI implementation of the particular instrument. If an instrument receives a Mode message that it can't respond to - or, for that matter, any MIDI message that relates to a function it does not have - it will simply ignore it.
The four modes are determined by the various combinations possible with two important variables, "Omni" and "Poly/Mono". Omni refers to the ability to receive on all MIDI channels at once (which isn't really as great as it sounds) and Poly/Mono refers to whether or not the notes on each channel will be responded to polyphonically or monophonically (this is independent of whether or not the synth is monophonic or polyphonic). If Omni is turned off, a state of affairs that is termed "Omni Off', the instrument will only receive on the one basic channel it is set to, and will ignore any other Channel messages (though it'll still receive all System messages). If the instrument is set to Omni On, then it will receive any Voice messages sent on any channel.
Now, as you might be able to gather, Omni On does not always serve a very useful purpose nowadays - particularly if you have a lot of pieces connected in your MIDI system - but when MIDI was created it was thought that some instruments wouldn't be able to change MIDI channels and thus, they needed a mode that worked with any other instrument regardless of which channel the other instrument was set to transmit on.
The four possible modes are:
- Omni On/Poly, which is also called Omni Mode or Mode 1;
- Omni On/Mono, or Mode 2;
- Omni Off/Poly, which is referred to as Poly Mode or Mode 3;
- and Omni Off/Mono, which is Mono Mode or Mode 4.
Each of these interprets the incoming Voice Messages in a slightly different way, and affects how the connected instrument will respond to those messages.
System messages, as mentioned above, provide general information for all of the connected devices in a MIDI system, regardless of their MIDI channel. System Common messages consist of basic, overall status information such as tune request, song select and song position. System Real Time messages contain timing data for drum machines, sequencers and other devices which depend on a common clock source to remain in sync with one another. Finally, System Exclusive messages cover a wide variety of information that is specific to each instrument. Specific patch parameters, sample dumps, and other kinds of data that are relevant only to a similar unit (or a computer with the appropriate software) are sent via System Exclusive, or SysEx, messages.
WELL, NOW THAT we've got the theory out of the way - I know it's not practical on its own and perhaps a bit confusing, but you have to understand a bit of it to make sense of real-world applications - we can move on to the fun stuff.
Let's start with a fairly obvious fact. The reason MIDI has made such a tremendous impact is that you can do so many incredible things with it. The benefits of having a MIDI system range from the ability to play many synths from a single keyboard and thus easily produce huge walls of sound, to automating the control of a number of devices in a sophisticated studio setup. In between are things like using personal computers to simplify the process of programming synths, playing synthesisers from controllers that aren't keyboards, and perhaps most importantly, the ability to edit minute sections of your music via a MIDI sequencer. Best of all, each of these applications can be achieved with a single type of interface and a simple MIDI connection.
The rest of the articles in this series will go into specific applications, and relate how the various types of MIDI messages are actually used in the real world. (I know I said I was going to talk about real world applications in this part, but I'm not going to. I lied.)
To understand them, though, you need to have a point of reference, so the accompanying diagram shows the basic MIDI system upon which all of our "real world" applications will be based. Each of the main components of a typical electronic music studio is included: a keyboard synthesiser which acts as the master controller, a "keyboard-less" synth expander, a MIDI sequencer, a drum machine, a MIDI-controlled effects device (like a reverb), and a MIDI Thru box to make sure everything is hooked together. You don't have to own one of each of these to work with MIDI (you may own more of some and less of others), but with a system like this, you have the tools to produce some incredible music - though a bit of talent always helps, too.
Next month we'll start putting the system together by explaining how a keyboard and an expander work together via MIDI. Until then, read your owner's manuals, and make some music.
Feature by Bob O'Donnell
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!