The Composition Thread

[Jeez]

Wow. Way tired. Another night with no megapost.

 

Sorry guys.

 

Thanks. See you all tomorrow.

 

Forever,

 

 

 

 

Kim.

[Rack]

Great thread Jeez! Your discussion on composition is provocative and well thought-out . It got me thinking of how Keith Jarrett performs in his solo piano concerts. I see how he employs many of the concepts you discussed. And, he improvises all of it!!

[Jeez]

Tap tap... weeeEEEEE.... tap tap... is this thing on?

Ladies and germs, It's that time again! Your daily update on The Composition Thread!!!

 

Tonight we're going to take a look at variation and development.

 

Ooh. Before I continue, I should point something out: Unlike the previous topics I've discussed, this one does not make reference to the chronological order of the music. In other words, tonight we're not worrying about the order of sections, etc. "Chronological" means "in time" (roughly, it's a simple definition, just enough for you to understand what we're talking about here).

 

 

I'll start with variation. I'm sure most of you know what variation is in composition. It's when you take a bit of music (it could be one bar, one instrument, or even a whole section) and you change it. More accurately, you would probably make a copy of that bit of music, and change the copy. The key is that the changed copy still bears some resemblance to the original (ie, we're not talking about transforming it into something unrecognisable).

 

Easy.

 

What then, is development? Development is kinda like a more sophisticated version of variation. If we define variation as "making some change", we could define development as "using a defined process to make a variation". At first glance, the difference may look like one of semantics, but I assure you it's more than that. Before we discuss why we'd want to use development, I'll give you some examples so you know what I'm talking about.

 

Example one:

Let's say we have a drum pattern. Let's say we want to mess with the position snare drum. If we were to make a variation, we might "randomly" move the snare drum hits around, perhaps inserting some or removing some. The variation will not actually be "random" - we'd be changing the snare hits according to what we think sounds good. I use the term "random" because it helps illustrate the difference between variation and development.

 

If we were to make a development of the drum pattern, we would use a defined process to alter the snare drum hits. Defined process? Well, we could do something like move all the hits one sixteenth of a bar earlier. Or we might gradually increase the density of snare drum hits (one in the first beat, two in the second beat, ... , four in the fourth beat). Or we might make the velocity (volume) of each snare drum hit increase as they progress throughout the bar. Or we might do all three.

 

Example two:

Let's say we have some melody, and make a copy and we want to change the notes on the copy. If we were to make a variation, we'd change the notes "randomly" - according to whatever we think sounds good.

 

If we were to make a development, we might do something like transpose each note one step higher than we transposed the previous note. Or we might change all upwards jumps to equal downwards jumps (and vice-versa). You might want to measure "steps" in your favourite scale, to avoid getting "wrong" notes. Or you might like the sound of the "wrong" notes.

 

 

The difference between variation and development, is that for variation we're using a defined process. You may also choose to think of it as a repeatable process. We could take the process that we used, and apply it to some other tracks, or another section.

 

 

From those examples, you might already be thinking about some ways in which development may be useful as an alternative to variation.

 

One advantage (that I've already mentioned) of development over variation is that you can use some process, and then apply the same (or a similar) process to other bits of music. For example, you could perform some development on a drum track during a bridge section, and then do the same thing on the bassline, or the chords, or whatever. Or you might make a development of the main melody, then perform and inverse or opposite development on the bassline.

 

Of course, multiple develpment doesn't have to be just in parallel - you could do them one after the other. For example, you might have a really dense drum pattern. You might have it plain once, then for the next repeat you could use some process to remove some hits. Then for the next repeat perform the same (or similar) process on the previous development, and keep doing that until the drum pattern is empty.

 

Another interesting approach could be to apply a similar process across different lengths of time. For example, you could come up with a process to let you thin out a drum pattern rapidly - so that at the start of the bar it is complete, but by the end of the bar there is nothing left. You could then apply a similar process to the bassline, but across two bars. Then do the same thing to the pad, but across four bars. Then the meldoy, across eight bars.... or something like that.

 

 

Keep in mind that while I'm using typical tradtitional western music constructs (notes, metric rhythm, drums, melody, chorus, etc) for examples I'd like to point out that these principals (all of them) are appropriate to almost all kinds of music. I choose these examples so the majority of readers will be able to understand them.

 

 

Anyway, I hope that gives you all some more ideas for composition!

 

 

Forever,

 

 

 

 

Kim.

 

[Yoozer]

I don't know if you're also getting to this at a certain point, because it deals a bit more with theory than composition - but I can't help to recommend an URL that somewhat deals with both : Chord Progressions, Simple and Complex. It might help you since the letters "A A' B A B B'" might not exactly stir the inspiration in you, so there's your way to fill in the A's, A''s and B's.

 

The simple map:

http://members.aol.com/chordmaps/part3.htm

 

More on chords, how to make 'm interesting and get your A to turn into an A':

http://members.aol.com/chordmaps/part4.htm

 

And the complex map.

http://members.aol.com/chordmaps/part5.htm

 

Hope this is a worthy contribution, I'm following this topic since it started. I sincerely hope we can let the moderators make this a "sticky" thread.

[scubyfan]

i'll tell scott to turn it into a sticky

[Roald]

absolute number one killer thread! you go for my HC KSS forumite of the month award. :D

 

 

I'm always fighting with theory...on one hand, you need it to make your stuff sound interesting, on the other its gets in the way of the spontaneous creative process sometimes.

[Birdienumnum]

 

Originally posted by Rack It got me thinking of how Keith Jarrett performs in his solo piano concerts. I see how he employs many of the concepts you discussed. And, he improvises all of it!!

 

 

The Koln Concert and many others.

[Jeez]

Let's try to keep his on topic. If you want to talk about stickies, there's another thread for that.

 

We should try not to clutter this resource - especially if it's destined for eternity (stickyness).

 

Forever,

 

 

 

 

Kim.

[Acehigh]

I'm not sure how this fits with formal composition, but when sitting at the piano making up tunes, which is what I do most of the time just because it is fun, I will hit a wall, not because an idea can't be found, but because there are so many possibilities. To get around this, I try to first restrict myself to a chord progression, then restrict the rhythm to just playing eight notes, triplets, or whatever, continuously through the chords to see if the mind can sort something out, which can include different chords, patterns, etc. A variation is to restrict the choice of notes to two or three tones that sound good with a chord, and then try to find as many ways to vary the rhythm with these notes.

[Jeez]

Yep, it's that time again. And again, I've left it too late so now I'm tried.

 

Though... I would like to use this opportunity to say something about the issue of Theory vs Creativity.

 

Some people seem to believe that theory and creativity are somewhat at odds with each other. I think the typical line of thinking is that a person who knows no theory is free to compose whatever she hears in her head. Unhampered by preconcieved notions of "right" and "wrong", the composer can get as close as possible to her ideal.

 

On the other hand, a composer who has studied theory is hindred by what is supposedly "right" or "wrong". Perhaps she came up with an idea, but (consiously or subconsiously) rejected it because it didn't fit within certain "rules".

 

 

Now, I'm not going to say that this is incorrect. I am in no position to tell you how your mind works, or how you write music. I would like to say though: It doesn't have to be that way.

 

 

First off, I'd like to point out that there are two kinds of music theory that a composer may draw upon (There are probably more, but I'll split it into two for the purposes of clarifying this discussion). - general composition theory, and "genre" theory.

 

General Composition Theory is what I've been discussing in this thread, and what I'm (personally) interested in. As you can tell by my previous posts, general composition theory is independent of style, instrumentation, size, etc. It is applicable to all composition. The important thing to remember is that it is not a set of rules. As you could probably tell by my previous posts, general composition theory is a set of tools, a set of techniques. It's not what's right or what's wrong. It describes common practices and their effects upon the listener. It's up to you, the composer, to decide what's "right" and what's "wrong" - depending on your context. Feel free to ignore them, disobey them, break them. Feel free to embrace them, extend them, use them to make your own set of tools.

 

In my experience, learning general composition theory hasn't limited my creative strength. On the contrary, it's definately increased it. I still come up with whatever ideas I want, and I still arrange them according to whatever sounds good. But in addition to that, general composition theory has given me ways to develop and extend my ideas to form works of much larger scale. Being conscious of the tools and techniques has allowed me to arrange and develop the material so it is stronger, more complete, and has greater impact.

 

Remember though: as the composer, it's up to you to learn from this. It's up to you to integrate this knowledge into your own style of composition, into your own set of techniques. You have to own it!. Remember: This is art, not science!

 

 

Another kind of theory I'd like to mention is Genre Theory. This includes "how to write trance", or "how to write the perfect pop song". This also includes Classical Theory - the rules upon which "classical"[1] music is composed. Genre Theory is much more often presented as a set of rules - and rightly so. If you want to compose awesome Trance music, you'd better get out that TR909 and stacked supersaw lead. If you want to compose "classical"[1] music, you'd better brush up on your scales and cadences.

 

Unfortunately, many people incorrectly believe all theory is "classical"[1] theory. I imagine this comes from some lower-school education, where this was the only type of theory that was tought. Unfortunate, really.

 

 

Anyway, that's all I've got to say for now. Well, that's all I can think of right now (2:30am). Hope it's useful. See you all tomorrow.

 

Forever,

 

 

 

 

Kim.

 

[1] By "classical", I mean pre-20th-century. I know this is not the strict or accurate definition, but I'm using the term in this more relaxed way to make discussion easier. In this thread, "classical" music encompases baroque, classical, romantic, and anything else I can't think of at 2:30am in the morning.

[Jeez]

Bummer. I can't think of anything to write tonight.

 

Does anyone have any special requests?

 

 

Forever,

 

 

 

 

Kim.

[Yoozer]

Some insight on classical theory (if available) would be very nice - and some of the motivations behind it, too.

[Jeez]

 

Originally posted by Yoozer Some insight on classical theory (if available) would be very nice - and some of the motivations behind it, too.

 

 

Any particular area of classical theory? Where do you want me to start?

 

Forever,

 

 

 

 

Kim.

[Rack]

 

Originally posted by Jeez Bummer. I can't think of anything to write tonight. Does anyone have any special requests?

 

 

 

Yeah Jeez, I want your thoughts on harmony. Pop harmony compared to jazz harmony. Jazz harmony- meaning compositions that are derived from developed theoritical studies; a "school-learned" approach, if you will. Pop harmony - meaning compositions that follow no particular "schooled" approach and are derived from popular cultural trends.

[Kirumamoru]

Originally posted by iamthenumber its fun to take this to the extreme, and become outright confrontational with it. like dropping a half beat off the end of every 7th measure of a 4 on the floor house song - watch the dancing fools get very uncomfortable

Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom Boom BoBoom Booom Boom Boom....

 

LOL

[Jeez]

Hmm. Harmony. Classical harmony. Jazz harmony. Pop harmony.

 

Well, first of all I should say that I hardly know anything about jazz harmony. Only two things: Use Seventh Chords Everywhere, and The Circle Of Fifths Is Your Bible.

 

 

Having said that, I'll focus on basic classical harmony, and how it relates to pop music. I'm not an expert on harmony, so I might say something that isn't quite right (or even outright wrong). If anyone knows any better than I do, I'd be happy for them to educate me.

 

To begin, I'll give a brief history of harmony.

 

 

Ah hell, I'm too tired. I've gotta get started on this thread before midnight.

 

 

Sorry again. I feel like I've let you all down.

 

 

Forever,

 

 

 

 

Kim.

[Neole]

Great stuff Kim, and although I employ all this in my compositions it is good to be made aware of it all over again!

 

I would like your insight into what makes a melody catchy, if you have any.

 

I would also like any theory (if it exists) behind the tempo of songs. It takes me a while figuring out what speed a song (with vocals) would sounds best at. I find myself changing tempos till the end and still not satisfied. Sometimes I feel it would sound great at 100 bpm and at others at 130. And then when I listen to it a week later I figure its too slow or too fast and change it to what finally is a comfortable tempo for the song. (With just instrumental tracks be it techno or jazz its much easier, its the vocals that create trouble).

[Treves]

Originally posted by iamthenumber Deny their expectation. -------------------------------------- its fun to take this to the extreme, and become outright confrontational with it. like dropping a half beat off the end of every 7th measure of a 4 on the floor house song - watch the dancing fools get very uncomfortable

 

Check out this song from Radio Zumbido on Epitonic.com - it's a seriously groovin, kinda-Latin beat, but every once in a while, it a new section is preceded with a measure of 5/4! And it's done very smoothly.

[Treves]

 

Originally posted by Jeez Tonight we're going to take a look at variation and development .

 

 

It occurs to me that DodgingRain's comment about filter sweeps could be applicable here - using a filter sweep on a repeated part but with different settings could constitute variation (though probably not development).

[DrDubbleB]

I have not yet read through this entire thread, but if you're at a loss for something to write about, how about percussion theory and time signatures?

[Jeez]

It seems that I don't get around to updating this thread until after midnight, at which time I'm really in no state to explain The Art Of Composing.

 

So unless someone else is prepared to contribute, this thread may have to die a natural death.

 

Thankyou for listening.

 

Forever,

 

 

 

 

Kim.

[Jeez]

"I love you, you love me,

The Composition Thread will set us free!

We love music oh so much,

We can't help but share the love!"

 

Welcome yet again to The Composition Thread! On tonight's episode, we'll take a look at Perception of time speed!

 

Some of you may have noticed that the music listening experience is often not perceived as metric time. That is, the speed of time seems to vary throughout a piece of music. Some parts seem race by as if the clock were somehow accelerated, whereas other parts feels as if they last an eternity. This is the speed of time.

 

We use the word percpetion because, while we cannot change the actual speed of time we can, through music, change our perception of the speed of time.

 

 

Altering our perception of speed time can be done in many more ways than simply changing tempo. But first we must ask the question: How do we percieve time?

 

Imagine for a moment that you do not wear a watch, and you cannot see any clocks or other time-measuring devices. How do you measure time?

 

You measure time by remembering events per time period.

 

Now this is not exact - "events" means anything that happens, and "time period" is your short-term memory. Both are variable. Before I move on, I'll ask you to consider two situations.

 

The first situation is that of you sitting down, doing nothing for five minutes (or even better - watching the clock!) Have you every been in this situation? Did you notice how slowly time seemed to progress?

 

The second situation is that of cleaning your messy desk in five minutes. Very busy, moving everything in its right place. Have you ever been in this situation? Did you notice that the busier you were, the faster time seemed to progress?

 

The same principal can be applied to music.

 

If you want to slow down your listeners perception of time, use less events and introduce less changes per time period.

 

If you want to speed up your listeners perception of time, use more events and introduce more changes per time period.

 

For example, we could focus on a section of music, and look at how many times the drum pattern changes, or how many chords there are, or how many notes (or note events) are in the melody.

 

It's important to note that while {more,less} events is a useful prop, controlling the rate of change is key. This is somewhat similar to the (more "classical") notion of "rate of presentation of material".

 

Another interesting way of looking at it is: Instead of measuring events+change per time period, look at time period per change. Approaching it from this angle, might count the bars between each change, or look at the length of each section. A "faster" bit of music may have more sections, each shorter; whereas a "slower" bit of music mayb have fewer sections, each longer.

 

That's all for today.

 

Forever,

 

 

 

 

Kim.

[Jeez]

I may have given the impression that while I'm happy to discuss the more "philosophical" aspects of composition and music theory, I have been sidestepping the call for more "concrete" music theory.

 

The truth is that I do know a fair bit of classical theory (I am studying at university), but quite frankly, I find it boring to talk about. Concepts such as modes, scales, keys, chords, etc are nothing to talk about and the real magic is when you actually play it for yourself. It's like someone asking "How can I use oscillator sync to improve my composition skills?" Well, I can tell you what osc.sync is, or how it works, and it might take a paragraph. The rest if up to you. There's really not that much to talk about if we're discussing the more "concrete" theory.

 

But for those who need somewhere to start:

 

A pitch set is a collection of notes that you use in your composition.

 

A tonal centre is the "root note", or the base note. It's often what the bass plays (especially in techno).

 

A mode is a set of pitches in combination with a defined tonal centre, to form a restricted field in which to choose notes for your composition.

 

A scale (or key) is a pitch set of seven notes, each no more than a semitone or tone apart.

 

A chord is a collection of different pitches all played at once, often by the same instrument.

 

A Triad is a chord with three notes in it. These three notes are usually seperated by major or minor thirds.

 

Forever,

 

 

 

 

Kim.

[Jeez]

What's all this semitones, thirds, whatever??

 

I'm measuring the distance between two notes.

 

A semitone is the smallest distance between two notes (on a normal keyboard instrument). For example: E to F is a semitone. C to C# is a semitone.

 

A tone is simply two semitones. C to D is a tone. F to G is a tone. B to C# is a tone.

 

Some other names:

 

 

Play them on any available keyboard and listen to them. You'll quickly figure out the pattern.

 

Forever,

 

 

 

 

Kim.

[Jeez]

Often, music that is based on scales and triads can sound harmonically simple. One way to make your sound more complex is to extend your chords.

 

An easy way to do this is to keep stacking thirds and fifths on top of your existing chords. For example, if you have a C-major chord the notes would be C, E, and G. If we wish to extend this, we could add B, then D, then F. To keep it as clean as possible, don't let any notes be less than a minor third apart - add the additional notes in the octave above.

 

Another example - you're working in F#-minor. Your triad is F#, A, C#. You could extend it by adding E, then G#, then B.

 

How do I choose these notes? Just keep alternating between major thirds and minor thirds. In the C-major example, C to E is a major third, E to G is a minor third, then G to B is a major third, then B to D is a minor third, etc, etc.

 

At first it may sound a little strange, but you'll get used to it.

 

This method is useful (I use it myself) because you still retain your tonal centre (your bassline doesn't get quaked), and the additional complexity is added gradually - meaning you can control how complex you want your chord to be. If you just want a bit of added complexity, just add the next note above. If you want more, add another note. Or another. Or another.

 

You could extend this idea by having a different instrument play the extentions. For example, you could have your favourite thick pad playing your triad, then have the extentions played by a thin airy pad in the background. This gives you more ways to balance the simple with the complex.

 

 

Wow. Four posts in one day. Do you think I've made up for the ones I skipped?

 

 

Forever,

 

 

 

 

 

Kim.