How do devices for sound generation work

MIDI basics workshop

All information about the MIDI standard

Being able to transmit audio signals is good and of course very important, no question about it. But sometimes you also want to transport something else from one device to another.

The need to send information about pressed keys to a synthesizer, which then generates a sound, contributed to the development of MIDI. What we can (s) do today of course with various computer interfaces or wirelessly, was done differently in the field of music a long time ago. After almost exclusively analog voltages were used for transmission until the late 1970s, various manufacturers have come together to develop an interface that is inexpensive, simple, understandable and, above all, universal. A gentleman whose name will at least be familiar to the keyboard-savvy readers was significantly involved: Dave Smith.


The result of the development mentioned is that "M.usical I.nstruments D.igital I.nterface ", a system that met the requirements with flying colors. The use of simple standard components kept the price so low that no manufacturer had to think about whether or not to install the hardware. Since the beginning of the eighties, thousands upon thousands of devices equipped with MIDI have come onto the market. At that time it was the introduced MIDI specification 1.0, which (attention, scoop!) Is still valid today. The advantages are obvious: All MIDI devices can communicate with each other, there are no version or generation problems. Sure, with a few devices a connection makes no sense, but you want to pair a synthesizer from 1985, a multi-effects device from 1999 and a keyboard you just bought? Then go ahead, you namely speak the same MIDI language!

This connection is quite simple, because there are three types of sockets, the plugs and cables are always the same: MIDI Out and MIDI In hardly need to be explained, MIDI Thru simply forwards the signal from the IN, so that several devices can be used one after the other can switch, which first receive the same information.

OUT sends data created in the device, THRU is a loop-through socket for the MIDI messages pending at the IN.

MIDI is often held up for its age and thus its basically low speed. That’s not a big deal, however, because it takes less than a millisecond to transmit a key press, for example. This is roughly the time it takes for the sound of an acoustic guitar to reach the player's ears.

So what can this MIDI do? The most prominent example has already been mentioned: If the essential units keyboard and sound generation are separated from each other, MIDI takes over this connection. A "Note On" command is sent at the press of a button, which contains, among other things, information about the button number and the "Note-On Velocity", that is to say the velocity of the touch. This is necessary information for many sounds. But that's not all: Because MIDI is a data interface, pitch changes can also be sent through one of these wheels on the keyboard or pedal activities. With various options, for example using potentiometers on the keyboard, data can also be sent that regulate the parameters for sound generation itself in real time - for example the pulse width of a square wave or a filter frequency. The command used for this is almost always the "Control Change".

It couldn't be easier: The sending device sends MIDI data to the IN of a receiving device via OUT.

What works for keyboard and synthesizer also works for other connections. A drum module, i.e. the heart of an e-drum kit, can, for example, pass on information about hits on one of the playing surfaces to a sampler - the principle is identical to that of the keyboard and synthesizer. Mixing console surfaces do not have a keyboard or playing surfaces, but if values ​​can be changed in a synthesizer, mixing software will certainly do the same. Lo and behold - there are many "controller" devices that do almost nothing more than control their virtual counterparts in DAWs such as Logic, Cubase or Live with faders, buttons and potentiometers. Most computers today don't have MIDI ports directly, but if the audio interface doesn't offer something like that, a small MIDI USB box is used for translation.

& quot; Multi-Port & quot; MIDI USB interface with several inputs and outputs

And when you already have all this MIDI data in your computer, you can also record it directly, right? And this is exactly where one of the main advantages of the whole system lies: Since only control data is initially recorded instead of actual sound, a lot of things can be done with it afterwards. For example, it is possible to edit out errors by moving keystrokes to the appropriate pitch or even deleting them completely. The recording can be speeded up or slowed down later - if it turns out that the singer sings more comfortably in a different position, an entire song can be transposed with a few clicks. Grooves can be changed by shifting notes on the timeline (if this happens automatically, one speaks of "quantization", which means something like "grid"). This list could go on almost indefinitely.

A simple but very helpful option when dealing with effects devices - but also with all others - is to use the "Program Change" command, which causes a switch in the same way as when it is generated on the device itself. Guitarists can control this from a foot pedal, for example, and some live shows are even automated: A MIDI sequencer always runs along, provides the drummer with the click, creates surfaces "independently" and automatically switches preamp channels and effects programs. If it is also to be dynamically automated, for example, that a reverb flag appears on a certain word of the singer and then disappears again, that's no problem either.

For a delay device, for example, it can be useful to know the current song tempo. If this is the case, you can easily specify repetitions in musical time. After all, "an eighth triplet" makes more sense than "329.3 milliseconds". This information sent via "MIDI Clock" can, in cooperation with the "Song Position Pointer" and the "Commands", connect two sequencers with one another. This is often used when a MIDI sequencer / DAW software is to be connected to a hardware sequencer or drum computer. The fact that both are running in sync is often only a matter of seconds and requires a single MIDI cable from the "master" (which specifies the speed and starting point) to the "slave". By the way: MIDI can also be sent from A to B internally in the computer.

What you should also know about the start-up is that there are 16 "channels" when transmitting via the MIDI cable. These are of course not individual wires in the cable, but are simply expressed as a number in most commands. A sound generator, for example, can differentiate between the sound for which it should reproduce a MIDI message or where it has to reset a parameter. In addition, when using the thru socket, different devices can be addressed if necessary.

So, you now have a rough overview of what MIDI can do. From this vantage point we will now descend together and explore the landscapes that we have just viewed from above.