Posts Tagged ‘ Composition ’

Proportion and variety

It’s a funny word, but it’s critically important.

Proportion in music best understood as the relationship between the amount of musical material in a song and the length of time that the song goes for. By ‘musical material’, I’m referring to the unique ideas – not counting repeats or slight variations. Another way of thinking about proportion is as the amount of variety in the song. There has to be a balance – too little variety will result in the listener getting bored and too much variety will result in the listener getting confused.

With too little variety, a song will have too much repetition. This is a common problem among beginner dance music composers. You’ve probably heard it before – a seven minute track that only has eight bars of music in it. It’s just different variations and combinations of the same material.

If you’re working on a track like this, you probably have very little source material and you’re trying to squeeze it for all it’s worth. The most effective solution will either be to introduce some more original material (not just variations or developments of what’s already there) or reduce the total length of the track.

With too much variety, a song will have too many different ideas, with little connection between them. You’ve probably heard this as a song that has a lot of good ideas in it, but seems to have a weak identity or doesn’t seem to have anything that binds it together. Instead of being heard as a single focussed piece of music, it comes across as a collection of different ideas.

If this sounds like one of your songs, you probably need to separate the ideas out into two or three (or more!) individual songs. Focus on getting more mileage out of fewer ideas. By varying and developing fewer ideas (instead of simply adding more fresh ideas) your song will sound much more focussed and cohesive.

Of course, the goal is to find the right balance. This is where judgement and experience play such an important role, and why it’s important to listen for proportion in your own music and others’ music. Try to identify when you’re listening to music that feels like it’s repeating itself a bit too much (like a sense of not knowing how long the song will go for) or when you’re listening to music that keeps switching between different ideas (like switching the TV channel or radio station).

There’s no magic ratio here. It depends on your personal taste and your listener’s expectations. Listen to a lot of music and you’ll know it when you hear it.



How to develop a melody from a simple idea

Sometimes you’ve got the beginning of a good melody. Just a few notes that seem to work well with the chords or bass line or other parts in y our song.

Don’t just loop it!

Turn that little melodic motif into a melody! Develop it into something that grows and moves and pulls the listener forward. Make something of it. It’s unfortunate that so many great-sounding songs are let down by their lack of melody. A melody is not just a one-bar monophonic line voiced above middle C. Melodies span several bars. They have contour. They have phrases. They interact with the other parts of the song.

So how do you turn your few notes into a melody?

If you’re stuck for ideas, you can start with the duplicate-and-vary approach. Quite simply – start with the short idea you’ve got duplicate it so it plays twice. Then change the duplicate so that it’s recognisably different and also recognisably derived from the original idea. Simple, eh?

Of course, that wouldn’t get you very far if that’s all you did. The trick is to take it further. For starters, you can create multiple variations and sequence them in a way that makes musical sense. If you’re not sure what to do, think about contour. Some variations will be busier than others. Some will be high-pitched than others. Some will be more recognisable than others. Think about the ways in which you can make variations and organise them into a sequence that makes sense to you.

This might take a bit of practice. Don’t worry if your first few attempts sound a bit weird.

Once that makes sense and you’ve got a good grasp of how it works, you can start to start thinking about other factors as well, such as:

  • Using more than one original idea. Start with two or three different melodic ideas, create some variations and then explore what happens when you combine them in interesting ways.
  • Dividing your melody into phrases. Rather than creating a long string of constant notes, divide your melody into shorter sections (try 4 bars) with each section separated from its neighbours by a beat. This can make a melody feel more natural if it loosely mimics the length of time a singer can hold a note or phrase before needing to take a breath.
  • Harmonic complexity. Think about where your melody uses the tonic (the same note as the key of your song). Phrases that use the tonic a lot will feel more stable than phrases that don’t use the tonic much (or at all). Use the circle of fifths to understand how stable/unstable different notes are.
  • Rhythmic complexity. This is very similar to harmonic complexity. Think about how many notes are on the beat and how many notes or off the beat (or in between beats). Phrases that have a lot of notes on the beat will feel more stable than phrases that have more offbeat notes.

And here’s a little secret: This approach works on more than just melodies. It’s a valid approach to take for basslines, drums, background parts… almost anything. Just keep asking yourself: How can I take this further?


Hierarchy of production, and why mastering is overrated

Mastering is the least influential part of making a recording. It has the least effect on the effectiveness of your creative expression – your ‘sound’.

It might sound obvious, but if you want a particular kind of sound, it starts early in the processes – as early as possible. Every subsequent stage of production has a smaller and smaller influence on the end result. The most influential part of making a recording is the initial concept and composition. A great song will shine through mediocre production, but a mediocre song will bore even with great production.

A hierarchy might look something like this:

  1. Concept. This is the initial set of decisions around what the recording will sound like. The decisions at this stage are (or should be) the driving force behind the direction taken at every subsequent stage. This is where the creative direction is established.
  2. Composition. Call it songwriting, call it beatmaking, call it programming. This is the stage where the individual notes are chosen.
  3. Performers / collaborators. These are the people who play the music. Sometimes there is one person that plays all the instruments who is the same person that composes the song. Other times the composer might not perform any of the instruments on the recording.
  4. Instruments. Now we start getting into the sound. Notice that the first three items are all about the notes and the performance. It’s only after these have had their effect that the sonic choices start to matter. The choice of instruments includes the choices of which family of instruments to use (e.g. guitars vs keyboards) and which variety of instruments to use (e.g. Strat or Tele).
  5. The recording engineer. The recording engineer is the person who is responsible for capturing the sound of the instruments. This includes making creative (and practical) decisions such as room acoustics, mic choice, mic placement, initial processing chain and recording medium.
  6. Recording tools. The relationship between the recording engineer and her/his tools is similar to the relationship between the performers and their instruments. While it is the tools that we ultimately hear, the decisions around which tools to use and how they’re used are more important. Recording tools also include the recording medium (e.g. 44.1k vs 96k or disk vs tape).
  7. The mix engineer. The mix engineer is the person responsible for balancing the sounds captured by the recording engineer. As a reader of this blog, you are probably a mix engineer (possibly one of your many hats). Even though there are some neat tricks (like reamping or pitch correction) at the mix engineer’s disposal, ultimately this job is limited by what was captured by the recording engineer and what was played by the performers.
  8. The mix tools. Noticing a pattern here? The relationship between the mix engineer and the mix tools is just like the relationship between the performers and their instruments and between the recording engineer and the recording tools. In this case, the mix tools include the console/DAW, outboard/plugins and mixdown media.
  9. The mastering engineer. Once the mix is done, the mastering engineer prepares the mixdown for distribution. This requires a different set of skills and different way of listening (compared to mixing). In many cases, it also requires different tools. The mastering engineer is the person who makes decisions around how the mixdown is prepared – usually involving changing the tone and level (and sometimes, dynamic behaviour) of the mixdown so that it compares favourably with similar commercial releases.
  10. The mastering tools. And this is the end. The mastering tools are the least influential part of the production process.

In this context, you can see that if you are responsible for making creative decisions, your efforts are best spent on having a clear creative direction, guiding (or participating in) the composition process, and ensuring the performers are all contributing their best.

Trying to achieve a certain type of sound through mastering is approaching it from the wrong end.


How long should your song be?

This is a deceptively simple question. Will some of you will answer “3:30” without hesitation? Do you even have a standard length that you aim for?

Nothing wrong with having a standard length, by the way. Some genres call for it. Hard to have a club hit that’s 2:30 (unless you have an extended version just for DJs). Hard to have a pop hit that’s ten minutes long (unless it’s the extended video version). Chances are your saga about wizards and dragons won’t fit in less than five minutes (fifteen if you’re wearing a cape).

But even within stylistic constraints, you need to pay attention to length. This is because the length needs to be considered together with the quantity of musical material in the song. This is called coherence. Basically, it’s the trade-off between musical material and time.

  • A song has low coherence when it has a lot of musical material squeezed into a short time.
  • A song has high coherence when it has not much musical material stretched over a long time.

You should aim for a balance between the two.

A song with not enough coherence will have too much material and not enough time to develop it. The song will sound fragmented and unfocussed. Listeners will be confused and disoriented.

Conversely, a song with too much coherence will have too much time and not enough material (or development) to use it effectively. The song will sound boring and undeveloped. Listeners will be bored and will probably not wait to listen through to the end of the song.

What’s the right balance? You have to be the judge. You have to use your experience and judgement to decide what the right balance it. Fortunately, finding that balance isn’t difficult – after a lifetime of listening to music you should have a pretty good sense of what level of coherence to aim for. If you haven’t got it right in the past, it’s probably because you simply haven’t been aware of it.

So what if the song you’re working on isn’t quite there yet?

If your song is sounding fragmented or there isn’t a clear musical language, you probably need to increase coherence. You can either increase the length of the song (to let the material develop a bit further) or you can remove some of the musical material (which in turn gives the remaining material more time for development).

If your song is sounding boring or your musical material is dragging on, you probably need to reduce coherence. You can either reduce the length of the song (to make it tighter) or you can add new musical material (which adds more breadth and variety).


What makes a good melody?

Most music uses melodies. Often they’re sung by the lead vocal but can also be played by other instruments, such as s synthesiser or guitar. Melodies play a big role in how memorable a song is – it’s usually what listeners hum to themselves after hearing the song. Melodies also form a big part of the song’s identity and character.

But not all melodies are created equal. Some seem to effortlessly soar or speak for the music. Others aimlessly meander or get stuck in a loop and go nowhere.

If you’re composing a melody, you should keep the following guidelines in mind:


Maintain the listeners’ interest by keeping the melody in motion. Momentum and development are key here. Make sure it feels like it’s going somewhere – not just repeating the same loop over and over again. A repeated two-bar loop is not a melody – it’s an ostinato pattern. While I’ve written about development and momentum in the context of overall song structure, it applies equally on the smaller scale of melodies (and on the larger scale of EPs and albums!).


As important as variety is, a melody that is constantly changing can easily feel as if it’s aimlessly meandering. It’s just as important to give the melody some shape. An example is melodies that follow the traditional advice of starting at a low pitch and peaking at the highest pitch about two-thirds of the way through. Not all melodies need to follow this shape, however. As a composer, you’re free to choose any shape – up, down, mountain, valley, wiggle, etc – but you must choose a shape! Often it makes sense to divide your melody up into segments – called ‘phrases’ – and give each phrase its own distinct contour. This will help the listener follow and understand the melody, which in turn helps in recognising and remembering it.

Something special

A good melody doesn’t just follow the rules. A good melody brings something quirky or special to the song. It doesn’t have to be totally unique – but it does have to have some character and identity. This could be musical – such as an accidental from a borrowed scale/mode, or an interesting rhythmic motif . It could also be in the sound itself – such as an unusual instrument or a distinctive synthesiser patch.

…and if you’re thinking to yourself that your music doesn’t have melodies – maybe it’s hip hop or abstract IDM… These guidelines apply to any foreground part. It can be spoken word vocals, chopped up samples, additive spectral noise, or anything else that you want your listeners to focus on.

If you want your listeners to be interested in it, you have to make it interesting to listen to!


What it takes to write a killer bassline

Ever had trouble coming up with a bassline? Maybe you’ve got a beat going, or maybe you’re starting with a chord progression. Maybe you’ve already got the start of something going, but you’re not quite satisfied and you want to take it to the next level.

Generally speaking, there are three main aspects to laying down a bassline: choosing a sound, composing and performing. Ignoring any of these will make it very difficult to realise the potential of the music you’re working on.

  • Choosing a sound is important no matter what kind of music you’re working on. Whether you’re producing a teenage punk band’s first EP or composing an epic orchestral dance film soundtrack, you shouldn’t skip this step. Make this deliberate – take the time to think about the sonic character of the bass. Is it loud or subdued? Clean or dirty? Heavy or funky? P-bass or Stingray? Sawtooth or squarewave?
  • Composing the bassline is also something that need to be done deliberately. Don’t fall into the trap of settling for something simple just for the sake of it. Sometimes a simple bassline is the right choice, but not always. Don’t be lazy! Think about how rhythmically complex the part needs to be (how much syncopation, how much groove). Think about how melodically complex the part needs to be (how many notes, how consonant/dissonant). Every song has its own balance, its own sweet spot.
  • Performing the bassline is just as important as choosing a sound and composing the part. Pay attention here – whether you’re the performer or you’re directing someone else. The performance is the difference between ‘yeah’ and ‘meh’. If the song calls for electric bass, sometimes it’s worthwhile hiring a decent session player – especially if you need groove! Even if your bassline is sequenced (and meant to sound that way), there is performance in adjusting synthesis/effects parameters. Don’t just draw the changes with your mouse as automation – grab some knobs and perform it!

With practice and study, you’ll be able to rely on your experience and music theory knowledge to reliably and consistently come up with good basslines.

The experience part is being able to hear in your head the main features of the bassline – the timing and groove, and the harmony and melodic movement. It’s a skill in being able to say ‘oh, this song goes like *this*, it needs *this* kind of bassline’. It requires having a wide range of musical taste and knowledge, and being able to draw upon that in a musical context.

The music theory knowledge comes in when it’s time to choose the individual notes. It’s a skill in being able to work out what the harmony context of the song is (what the chords are, the scale/mode, the melodic language of the song, etc) and using that to know what notes will work well and best support the music.


How to get faster without speeding up

You’re working in a sequencer, and you have a fixed tempo for your song. Think you’ve got a perfectly regular grid of time events? Think that with everything neatly arranged, that the speed stays the same throughout the song?

Think again.

Don’t get confused! What I’m referring to here is the difference between tempo and pace.

They’re not the same thing, and you’ve probably noticed it yourself. For example, a song might have a very fast tempo, but feel slow (for example a lot of electronic dance music). On the other hand, a song might have a slower tempo, but feel quicker (for example, a lot of pop or hip hop).

Tempo is a familiar concept to anyone with musical experience – either as a musician or as an electronic music composer. It’s the speed at which beats are counted. Faster tempos mean more beats in a set period of time, and less time between each beat. Pretty straightforward.

Pace is a related but distinctly different concept. While tempo is about how fast the grid is, pace is all about how fast the music feels. There are several factors that influence pace, such as groove and rate of change.

The most obvious example of this is half-time feel. This is where the tempo might be quite fast, but the song feels like it’s actually at half speed. Instead of the snare drum sounding on the 2nd and 4th beats, it only sounds on the 3rd.

So how do you get faster without speeding up? Keep the tempo steady! For the section of your song that you want to feel faster, use composition techniques that increase the pace. Use rhythmic/groove elements that emphasise the upbeat or give a double-time feel. Also increase the rate of change – not just by making sections shorter, but melodic/harmonic elements too (such as chord progressions or harmony parts).

Deliberate changing the pace independently to the tempo is a great technique for emphasising the contour of the song. Sections with a faster pace increase anticipation and excitement and sections with a slower pace provide relief and contrast. Read more about contour here: