What Is MIDI and Why Should I Care?

Spoiler Alert: When you get to the end of this article, you will realize that MIDI is your new BFF.


MIDI is everywhere. It's in the tiniest home studio you've ever seen. It's in the fanciest commercial studio you've ever seen. It's on concert stages around the world.


Why?


For the same reason language is everywhere: it's incredibly useful.


I was intimidated by MIDI for a long time. I avoided learning about it the same way I avoided learning about what was under that pile of clothes in my closet. Easier to pretend it didn't exist.

Years later I was simply uninterested in MIDI. I associated MIDI with cartoonish, 8-bit representations of real instruments.

When I finally came around, MIDI became one of my absolute favorite studio tools.


Before we explain exactly what MIDI is and how MIDI works, here are just a few of the popular uses of MIDI:

  • Programming realistic performances for instruments we don't own (or know how to play)

  • Quickly coming up with and testing song ideas while composing new music

  • Triggering loops, sounds or samples on stage (or during rehearsal!)

  • Capturing a performance that can then be manipulated note by note without introducing weird artifacts (like we can sometimes get when editing audio)

Here's how we'll attack the subject:

- Part 1: How to make sense of MIDI using a real world example...what IS it?

- Part 2: What information does a MIDI file contain?

- Part 3: Some ideas for composing with MIDI

- Part 4: Mixing with MIDI files in your session


Part 1: What is MIDI? Try This Real World Analogy

Have you ever seen one of these?


It's a music box and we can break its design and operation down into three basic components.



First, there is the hair comb-like piece held down with two screws. This little guy has metallic strips of varying length. Striking any one of the little strips will make a pitched sound. Each strip makes a different sound (i.e., note) because each one has a different length. Sound familiar? This is basically a mini-piano. A piano has strings of varying length and thickness that also make different notes when plucked.

Second, there is a cylinder with small bumps on it. When this cylinder rotates, the bumps pass the different strips of metal on the mini-piano. Except instead of just passing by the strips, they actually lift up the little strips and "pluck" or "strike" them as they pass - thereby making the pitch of that particular strip. The bumps are arranged in a very specific way. In the case of the video above, the bumps are arranged so that the cylinder will play the Pink Panther theme as it is rotated. We can say that the cylinder has been "coded" with the theme song or that the theme song "information" has been "saved" onto it.

Third, you have the crank, gears, housing and the rest of the claptrap that is designed for only one purpose: to allow the cylinder - or information - to interact with the mini-piano - or instrument - to play a song.

OK. So how is this an analogy for MIDI?

The mini piano is an analogy for a MIDI-controlled instrument. This would be a virtual instrument such as a synth plug-in that came with your DAW (Digital Audio Workstation), a drum machine, audio samples that you want to trigger or even a hardware synthesizer.

The cylinder is MIDI

MIDI is the "information" that we can combine with some virtual instrument to create a sound. It does not contain any audio and it cannot make any sounds on its own (just like the cylinder could not play the Pink Panther Theme if you removed it from the music box).

It is "smart" because it knows a song, but it cannot play it without help. The synth or mini piano is not smart. It does not know any songs. However, it can make sounds if you can feed information into it.

Finally, the crank and other clap trap is simply the system that combines the two: your computer!

Remember: MIDI files do not contain any audio! They only contain information that can be fed into a virtual instrument to create audio.



Part 2: What Information Does a MIDI File Contain?


OK. So we understand the real world analog to MIDI: our music box cylinder with bumps on it.

In the digital world, you can think of a MIDI file as a small computer program.

We can create some computer code that has instructions and feed those instructions into a virtual instrument. The virtual instrument then reads those instructions and generates audio.

Actually, if you think about it, it's a lot like handing a piece of sheet music to a professional musician who is then able to perform the piece.



Don't worry though, you don't have to know anything about sheet music to understand the information that is contained in a MIDI file.

Pretend you had to tell a robot how to play a song on a piano. What pieces of information would you have to give the robot?


You would have to tell it:

1. Which key(s) to press

2. How "hard" to press the key (this is known in MIDI terms as "velocity")

3. When to START pressing the key(s)

4. When to STOP pressing the key(s)

Take a moment to think about how powerful that list is.


With just those four pieces of information, you can have the robot play absolutely any song reasonably accurately.


These are the exact pieces of information contained in a MIDI file.


Let's have a look at how this information will be displayed in our DAW and break the display down by each of the four pieces of information we discussed above.

In the image below you see there are little colorful blocks on a grid. These blocks are the bumps on our music box cylinder. They contain all four of the pieces of information we will be discussing.

1. Which piano key(s) to press


On the left side of the picture, there is a keyboard. This keyboard is a representation of the notes of our virtual instrument.


To the right of the keyboard you can see what looks like graph paper - or a grid.


Each row in the grid starts at one of the notes on the keyboard and travels horizontally to the right. Each row represents the note that it connects to at the keyboard.


2. How "hard" to press the key (this is known in MIDI terms as "velocity")


On the bottom of the image above, just below the keyboard, you can see a small text box that says "velocity".


Velocity is simply an indication of how hard the note is to be played. If you've spent 30 seconds with a piano you know that pressing a key hard makes a different type of sound than pressing that same key lightly.


The same is true for virtual instruments (in most cases). Each velocity bar you see in the lower portion of the picture corresponds to the note marker (colorful block) directly above it.


In this example (which is from Pro Tools) you'll also notice that a bigger velocity bar corresponds to a darker shade of the note marker above it. This is to give you a quick visual indication of how high a note's velocity is - even if you're not looking at the velocity display (perhaps because you've minimized it).


Each note's velocity can be adjusted individually - something we certainly couldn't do had we just recorded the audio coming out of a keyboard.


3. When to START pressing the key(s)


The vertical red line above represents the time bar. If you press "play", this bar will start moving from left to right at the speed dictated by the tempo of your song.


As the time bar runs into the starts of notes, the instrument will start to play those notes at the velocity indicated by the velocity bar.


Note that multiple notes can be played simultaneously as chords. We will occasionally run into synths or virtual instruments that are only capable of playing one note at a time (i.e., they are "monophonic"). However, this is not limitation of MIDI. It is just the nature of that specific synth or virtual instrument.


4. When to STOP pressing the key(s)


As the red bar continues to move to the right and encounters the ends of note markers, the instrument will stop playing those notes.

Here is what those screen shots will sound like when combined with an electric piano virtual instrument:



Now that we understand what MIDI and what information it contains we can talk about how to fit it into your creative workflow.


That's what Part 2 is for!


Vadim is a mixing engineer and producer. His passion for recording and mixing was sparked through playing guitar and writing music as a teenager. Today he operates Calm Frog Recording where he helps artists get their songs to sound as good as they possibly can. He loves sharing his passion for recording and producing music with others through this blog and the DIY Recording Guys Podcast.

For many more tips, check out Vadim's FREE eBook at www.howtorecordyourband.com