A Brief Overview of MIDI

[ LiB ]

A Brief Overview of MIDI

MIDI stands for "Musical Instrument Digital Interface." It represents two things: First, MIDI is a communication system used to transmit information from one MIDI-compatible device to another. These devices include musical instruments (samplers, synthesizers, sound modules, drum machines) and computers (or other hardware devices, such as synchronizers). Second, it represents the hardware. the ports and jacks found on all MIDI instruments and the MIDI cables connecting them to allow the transmission of musical data. Each time a key is pressed or a wheel is moved, one or more bytes are sent out from a device's MIDI out port. Other devices connected to that sending device are looking for those bytes to come over the wire, which are then translated back into commands for the device to obey.

MIDI sends information at a rate of 31,250 bps (or bits per second). This is called MIDI's baud rate . Since MIDI is transferred through a serial port, it sends information one bit at a time. Every MIDI message uses 10 bits of data (eight for the information and two for error correction), which means that MIDI sends about 3,906 bytes of data every second (31,250 bps divided by 8 bits to convert into bytes). If you compare this with the 176,400 bytes (or 172.3 kilobytes) transfer rate that digital audio requires when recording or playing back CD-quality sound without compression, MIDI may seem very slow. But, in reality, it's fast enough for what it needs to transfer. At this speed, you could play approximately 500 MIDI notes per second.

What Does MIDI Transmit?

MIDI sends or receives the following information:

• Events related to your performance, such as a note played or released.

• Parameters for these actions, such as the channel setting. Each MIDI cable or port can support up to sixteen channels of information, much like having up to sixteen separate instruments playing at once.

• Wheels and pedal controls (pitch bend wheels, modulation wheels, sustain pedals, and switch pedals).

• Key pressures of pressed keys, also known as Aftertouch information, sent by the controller keyboard or by the sequencer to a sound module. Note that not all keyboards support this function, but when they do, the information is sent as MIDI data.

• Program changes (or patch changes), as well as sound bank selections.

• Synchronization for MIDI devices that have built-in timing clocks. These timing clocks may determine the desired tempo of a drum machine, for example. Through synchronization, MIDI devices can also follow or trigger other devices or applications such as sequencers or drum machines, making sure each one stays in sync with the "master" MIDI clock.

• Special information, also called System Exclusive messages , used to alter synthesizer parameters and control the transport of System Exclusive-compatible multitrack recorders .

• MIDI Time Code or MTC, which is a way for MIDI-compatible devices to lock to a SMPTE device. a translation of SMPTE into something MIDI devices can understand.

MIDI transmits performance data , not sound. You can think of MIDI as an old player piano using a paper roll. The holes in the paper roll marked the moments at which the musician played the notes, but the holes themselves were not the sounds. MIDI information is transmitted in much the same way, capturing the performance of the musician but not the sound of the instrument on which he or she played. In order to hear the notes that MIDI data signifies, you will always need some kind of sound module that can reproduce the musical events recorded in MIDI. This sound module could be an external synthesizer module, a sampler, a virtual synthesizer inside your SX software, or even the synthesizer chip on your sound card. This is precisely one of the types of information Cubase allows you to work withrecording a musical performance through your computer, using a keyboard to trigger the events, and Cubase as the recording device and the sound generator, thus creating a virtual paper roll inside the application.

MIDI Connectors

MIDI devices typically have either two or three MIDI-connector plugs: In and Out, or In, Out, and Thru. Usually, two-port configurations are reserved for computer-related hardware (see left side of Figure 1.1), as well as software-based synthesizers. This is due to the fact that the output connector can be switched within a software application, usually to act as a MIDI output or as a MIDI thru connection. Hardware devices, on the other hand, will typically host the three-connector configuration, as seen on the right side of Figure 1.1.

Figure 1.1. On the left is an example of a two-connector configuration typical of computer MIDI interfaces. On the right is a three-connector configuration typical of keyboards and sound modules.

MIDI Out

MIDI does not transmit sound over wires the way audio components in a sound system do. Instead, MIDI sends a message capsule that contains an identifier portion and its associated parameters. For example, when you play a note, the identifier would be that this is a "note on" event, and the parameters would represent the note number corresponding to the key you pressed, along with a velocity parameter indicating how hard you hit that note.

As you play on a MIDI keyboard, the computer in the instrument examines your performance. The instrument's computer then converts the performance into a stream of MIDI code that translates your actions. That information is sent out over an instrument's MIDI output to other synthesizers that reproduce the performance using their own sounds.

A MIDI output will not echo (retransmit) any MIDI events your device receives from its MIDI input. If you wish to do so, you will need to use the MIDI Thru connector, which is described below.

MIDI In

MIDI keyboards can be viewed as two machines in one (see Figure 1.2):

• A MIDI interface: The computer processor that monitors the keyboard, program memory and front panel displays, and MIDI ports.

• A sound module: The part under the control of the onboard computer, the electronics that actually make the sounds.

Figure 1.2. Configuring your keyboard's MIDI input.

The MIDI input receives incoming MIDI information and sends it to the instrument's computer. The computer analyzes and acts upon the information in much the same way as a performance on the original instrument, such as pressing a key to play notes. It makes no difference to the sound-making parts of a synthesizer whether the command to play notes comes from a key press on the instrument itself or as a command from other MIDI devices.

When you are working with a sequencer such as Cubase, it is recommended that you set your keyboard's Local properties to off since both Cubase and the keyboard would be sending MIDI information to the sound module portion of your keyboard if your keyboard was connected to Cubase through MIDI. The local setting on a keyboard tells this keyboard to play the sounds directly when you press the keys when it is set to on and does not play the sounds when it is set to off. In other words, setting it to off will disconnect the bridge that exists inside your keyboard between the actual MIDI input (the keyboard) and the sound module part that allows you to hear the keyboard's sounds as you play the keys.

When using Cubase, you will use your keyboard to send MIDI to it through this keyboard's MIDI Out. Cubase will then record the information you play and send it back to your keyboard through its MIDI In connector. If your keyboard's MIDI setting is not set to local off, the sound module portion of your keyboard will play the sounds twice: once when you play the notes on your keyboard and once when Cubase sends the MIDI information back to it.

On the other hand, if you have a sound module without a keyboard, you will not need to take this precaution, since there is no MIDI being sent to the device's MIDI input besides what is connected to this input.

MIDI Thru

MIDI Thru retransmits the MIDI data that came through the MIDI input of a device so that it can be received by another device in a chain. An important concept to understand when putting together a MIDI-based music system is that anything played on a keyboard goes only to the MIDI Out and not to the MIDI Thru. This third port is very useful when you want to avoid MIDI loops when hooking your MIDI devices together.

A MIDI loop occurs when MIDI information is sent from one instrument to another and then back to the initial instrument. This will cause the instrument to play each note twice and, in some cases, will cause a feedback of MIDI data that could potentially cause your sequencer to crash.

If you have a MIDI patch bay or a multiport MIDI interfaceMIDI devices with multiple MIDI inputs and outputs called MIDI portsyou are better off using a separate MIDI output for each connected device, thus reducing the amount of information flowing in a single MIDI cable. Each MIDI port in a MIDI setup sends or receives up to sixteen MIDI channels. For example, if you are using a MIDI interface with four MIDI ports, you will have four MIDI inputs and four MIDI outputs and will have control over 64 MIDI channels. If you do not own a multiport or MIDI patch bay, daisy-chaining MIDI devices using the MIDI Thru socket is your best bet (see Figure 1.3).

Figure 1.3. Using the MIDI Thru connector to hook multiple MIDI devices together.

[ LiB ]