The Composition Thread

[Kirumamoru]

Indeed, most excellent.

Kiru

[Jeez]

Originally posted by Birdienumnum Bravo!

Originally posted by Kirumamoru Indeed, most excellent.

Don't cheer me! If you want to say something useless, tell if it's helpful for you!

Forever,




Kim.

[tucktronix]

I'm glad you've touched upon the idea of perception of time speed. Whenever I write a composition, I always have in my mind how long I would like the song to be. I would add parts to it, but the composition often does not reach the length I would like it to be. There is a sort of a benefit from this though. There is enough parts and motion to fool the listener into thinking that it is a 6 minute song when it is actually a 3 1/2 - 4 minute song. I hope that this is within the realm of what you're saying

[gilbertopb]

Originally posted by Jeez Don't cheer me! If you want to say something useless, tell if it's helpful for you! Forever, Kim.

Thank Jeez.
I have found lots of usefull and practice topics.
I am not an erudit, nor a aplied student, just I try to do something with my one and half neuron.

[Jeez]

Originally posted by tucktronix I hope that this is within the realm of what you're saying

It's similar. Two very different pieces may be both exactly the same length... but that doesn't mean that the listener will percieve them to be the same length!

If you start composing a piece knowing how long you want it to be, perhaps you could try planning out the structure before you even do anything else. For example:

At 120BPM 4/4, a five minute piece will have 150 bars. Knowing this, you could split it up into large sections - maybe bars 1-32 are the introduction, bars 33-72 present the primary thematic material (the main themes), bars 73-130 are some kind of development or alternate 'B' section, bars 131-142 revisit the main theme, and the remaining eight bars (143-150) form the conclusion (outro).

Once you've sorted your outline, you can either go straight into "putting the notes in", or you could split the sections up even further. For example, bars 33-72 present the primary thematic material, but that's 48 bars (quite a lot)... so maybe split that into a ternary structure (A B A), where each theme has sixteen bars.


Anyway, that's one possible approach.


Another way of "filling in time" is to take the process of development further - rather than keep coming up with new sections, develop the ones you already have. This will help you generate more material, and also make the final piece more cohesive. Read my post on variation and development for more details.

Forever,




Kim.

[tucktronix]

Thanks for the info!! I usually do follow the structural approach, laying and labeling each part on Cakewalk. I will try developing some further, as you had suggested. I used some development techniques on a tune I've written called "Guitaricide", with success. Again thanx

[Treves]

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. 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.

 

 

Hey, good trick! =)

 

I wanted to take this opportunity to introduce some theory jargon, and maybe pave the way for some jazz theory discussion.

 

The tones Jeez is talking about are the 7th, 9th, and 11th of the chord. Why do we call them that? It's easy - let's look an the F# minor chord:

F#

G#

A

B

C#

D

E

 

If we assign numbers to these tones (aka the "degrees" of the scale), then we have

 

F# - 1

G# - 2

A - 3

B - 4

C# - 5

D - 6

E - 7

 

So it says here that the G# is the 2nd degree of the scale, right? Well, yes, except now we're talking about extensions. So, let's extend the scale, as if we were to continue playing it up on the keyboard:

 

F# - 1

G# - 2

A - 3

B - 4

C# - 5

D - 6

E - 7

F# - 8

G# - 9

A - 10

B - 11

 

If we continue counting after 8, then we see that G# becomes the ninth degree of the scale, rather than the second. Why is it called the ninth and not the second? Well, a few reasons not worth going into. Basically, a 2 and a 9 are used in different ways.

 

Adding a 7th or 9th is a good way to fatten up a chord (try it on a pad). When you want to impress your friends, refer it to it as an "F# minor add 9 chord!" 7ths and 9ths are also very common in jazz chord voicings. 11ths and 13ths appear in jazz as well, but not as frequently.

 

More on extensions:

In the olden days, a 7th chord was used in a very specific way - the 7th of the chord (the Bb in a C chord) "wants" to resolve down a half-step to an A. So, a C7 to F chord progression is very satisfying, since you have a half-step motion from the Bb to an A, and there's more half-step motion between when the E goes to the F (or Eb if it's an F7 chord). Also, the root note goes down a 5th, which is pleasing to our ears (see Yoozer's chart).

 

However, as music has gotten more "modern," this rule isn't as stricly followed as it used to be. Still, you'll hear the V7-I resolution a lot, especially in slower pop songs in which the 7 has time to be emphasized.

 

(warning - post written with no coffee)

[Jeez]

Sorry no post yesterday, I only had two hours sleep the night before, and thus eved up going to bed at about 9pm (and waking up at midday).

Forever,




Kim.

[Treves]

Just thought of an exercise for my post on extensions - whip up a beat and bassline in your sequencer, then play the extensions of the chords on a pad - so if you're grooving in the key of C, play a chord of D and F (D and A is neat, too).

You could even play the normal chord tones for section A, and then add/play just the extensions for section A'.

Man...such a good discussion of ideas in this forum right now, but I'm so far away from my studio!!

[Jeez]

Originally posted by Treves C D F D A A A'

 

 

You really should make it clear when you're referring to notes, or chords, or sections.

 

Something like "chord C, note D, note F", or always specifying major/minor (if appropriate) for chords... Simple really, just to avoid confusion.

 

Forever,

 

 

 

 

Kim.

[Jeez]

Sorry about that. No post last night because I was busy with another project. I've got another post coming up for tonight though.

Forever,




Kim.

[Treves]

Fair enough. I've edited the post.

[Caleidoscope]

I want to teach you all a little trick. Instead of writing down chords like this: C-major, A-minor, F-major, G#9b-minor, you could do it liek this: C, a, F, g#9b. Get it? Capital letters for majors, small letters for minors. Piece of cake, short, and easy to read .

[Jeez]

Originally posted by Caleidoscope I want to teach you all a little trick. Instead of writing down chords like this: C-major, A-minor, F-major, G#9b-minor, you could do it liek this: C, a, F, g#9b. Get it? Capital letters for majors, small letters for minors. Piece of cake, short, and easy to read .

Good idea for chord charts or anywhere else where chords are assumed.

However, in this thread we may refer by letter name anything from notes, to chords, to keys, to scales, modes, etc.

That's why I've requested that we make it clear. A few extra seconds of typing is worth the added clarity.

Forever,




Kim.

[Jeez]

Quiet, quiet, sit down...

Good evening, class. Tonight's lesson will be on Expectation and Excitement - particularly how it relates to drum programming.

Before we begin, I'd like to make clear that while we'll be discussing one-bar drum patterns, these prinicpals apply to all aspects of rhythmic composition (including basslines, melody, etc). Additionally, these principals are applicable to drum patterns of arbitrary (any) complexity, but for simplicity we will be primarily concerning ourselves with the four-on-the-floor kick drum pattern often heard in popular club music.

I'm also assuming equal velocity (volume) for every note. Variable velocity will make this ubject far more complex than can be discussed in a single post. I'm already pushing it as it is.

A little more on drum patterns of arbitrary complexity. The important thing to remember here is that everything is relative to the normal. That may sound kinda obvious, but I'll explain further. A static (looped) drum pattern can only be interesting for a limited amount of time. After hearing the repetition several times, we (the listener) know what to expect. However complex this drum pattern is, repetition makes it the normal. If, after a few iterations, we change the loop, we (the listener) will be surprised in some way... but if this variation is then looped, it becomes the new pattern - it becomes the new normal upon which we build our new expectations.

So basically, we will discuss two patterns - the normal, and the variation. The normal is what has been repeated, and what the listener expects. The variation is a new loop that is very similar to the normal, but different enough to surprise the listener.

Ok, enough semantics. Let's get on with The Real Thing.

Expectation

We can create expectation by removing notes from our variation. The sense of expectation is created because the listener expects (from the normal) a certain note to exist, but it does not. You might say the listener "wants" something to be there, but it is not, so the listener is kept "wanting".

For example:

 the existing note that it associates with (variation, beat 4), the effect is that of strengthening the existing note, or elongating it (making it longer).

I'll discuss some special cases in my next post.

Forever,




Kim.

[Jeez]

What? Does no-one care about this stuff anymore? Don't let this thread get buried!

Forever,




Kim.

[Jeez]

Ladies and gentlemen, we are approaching some special cases of adding and removing notes for excitment and expectation. Please fasten your seatbelts and remain calm.

Tonight I'd just like to append to last night's post by discussing two particular special cases of removing notes (for expectation) and adding notes (for excitement). These special cases are Moving notes back and Moving notes forward.

As you may have gathered from last night's megapost, adding or removing notes from a pattern not only dynamically changes the tension and stress positions in the pattern... adding or removing notes can also change the density of the pattern. In many cases, this is desirable because alterning the density can contribute to a general change of intensity that is suitable for the part.

However, interesting effects can be created by changing the tension points without changing the density. If we look at creating expectation, an alternative to removing notes could be to move notes back. This is very similar "pushing back" - mentioned in my (more general) discussion of expectation. The difference here is that this time we are applying the principals to individual notes, rather than larger sections.

Similarly, if we want to create excitement in a drum pattern, an alternative to adding notes could be to move existing notes forward. Again, this is similar to "pushing forward", only applied to individual notes rather than whole sections. I won't discuss this in much more detail, as the principals are very much the same.

Listening to Tori Amos is making me tired, so I think I'll call it here.

I will say one thing more before I sign off though:

This thread is sinking like a stone. My impression is that people enjoy reading this, but choose not to contribute much. That's fine... but when I'm the only one who posts here in two days, I start to wonder whether there's anyone actually still here. This thread dropped off the bottom of the list earlier today. Now, I'd really like for this to stay near the top so more people can see it and (hopefully) learn from it. To keep this thread alive though, I'd like your help. Some ideas:

• 
  
  Forever,
  
  
  
  
  Kim.

[Kirumamoru]

Hey Jeez, I'm still reading it, I just feel like it's your show and I haven't really thought of anything to contribute. You yelled at me the last time I just said "Hey, good work" so I've been keeping my mouth shut, too.

Kiru

[Neole]

Reading regularly. Marks attendance register.

[Phil]

mmm....ok, im adding one of my own,
i admit that i havent read the whole thread closely, so the way i go about composition may already have been touted, and its my results arent always v good anyway.

but still....

MY best pieces come from playing on the piano. Composing on a keyboard is very difficult to me, unless i am just inspired by a particular sound . There's something about the way a piano creaks and the echo when u put the pedal down that relaxes and inspires me.

often i will have one part of a piece done maybe a year ago, and only just find a part that fits it this morning,

but as it is an ongoing process, one part fitting another part means i always have a lot of options if i just cant compose a bridge on the fly which matches the current song.

The elements which my best works have are, : a pulse, a real ebb and flow to the music, and plenty of space for the music to breath.

If i simplify my ideas down i find i can arrange them much better for synths instead of just a piano, but i still have problems converting a piano tune to a full arrangement.

anyway, that's my poorly described process, and if someone has already said this, thats fine.

It's a great thread, and i try to read a little everytime i'm on.

Thanks,
Phil

[Alex9992001]

I second (third? fourth?) the "reading with great interest" motion. I only stubled accross it today, so haven't read it all yet, but it's good stuff, and has inspired me to start composing again, having had a break to work with technical stuff and set up my studio.

Keep up the good work and I'll post something exciting, when/if I get inspired.

[Jeez]

Well, I didn't post last night - I had an early night. Trying to catch up on sleep

Anyway, in a (weak) attempt to make up for it, I'll discuss a technique that I've personally been exploring for a couple of years - sections of variable length.

Often I've found that using sections of "metric" lengths (four bars, eight bars, sixteen bars) can often give a piece a very stodgy, predictable pace. No matter how exciting or cool the actual musical material is, sections of metric length can really weigh a piece down.

This is because the listener knows (or can guess fairly accurately) when each change will occur. In her/his mind, the listener has heard a sigificant amount of the piece before it's actually been played.

Let me give you an example. Let's say that so far, every section has been sixteen bars long, and it's very obvious whether each section is static or transitional. Within a few bars of hearing a particular section, the listener already knows what the rest of the section sounds like - sometimes to the point of not actually having to hear the remainder of the section. This is the point at which the listener becomes distracted, starting to talk, or getting bored.

I've found a particularly effective way to reduce this effect - variable section lengths. Instead of making each section a "metric" length (four bars, eight bars, sixteen bars, etc), the idea is to make them "odd" lengths. This has two main implications:

How you come up with the lengths is up to you. I composed a piece a couple of years ago where each section length was a Fibonacci number - the sections were all lengths like 5, 13, 21, 34, etc.

Another piece I composed had section lengths chosen by rolling dice.

Of course, it doesn't have to be random. You might choose prime numbers, or the date of every Monday in the year, or anything else. You could even choose the lengths as you compose the piece, depending on the flux in the piece.

It's really just about making the sections have lengths which aren't even multiples of four or eight.

Try it out.

Forever,




Kim.

[Kirumamoru]

Agh! The thread is falling. The thread is falling!

Oh wait, no, here I come to bump and save the day!

I like this last topic. It puts me in mind of a project I did once. I used to hit up the fiddling circuits in New England back in high school, and did pretty good for myself. One year I couldn't find a nice waltz for the contests. This was the summer after Final Fantasy VIII was released for PSX (or maybe the summer before?) in America, and I had downloaded the ball-room cutscene (like two thirds of the 8-18yo male population of the US, I'm sure), and I really liked the waltz that they played in the scene. So I transcribed the melody line and got someone to help me write the chords for it, and it was a HUGE hit with the audiences. People kept coming to ask me where I found it, and I just told them I copied it off of an old tape that I'd gotten my hands on.

I called it "Rinoa's Waltz", after the name of the female lead in the game.

Anyway, fiddle music is pretty much 8-bar plus 8-bar, AB music, since it's all dance music, and you have to be able to coordinate dancers, and I remember that the piece was in fact not based on eight bars at all, so when I went to transcribe it I had to alter it a little. There was a bridge part after the B that went back to the A, too, I think, which I left out. I think it retained it's character quite nicely though, in a standard 16-bar waltz. I was very pleased with the results (though I never managed to do better than 3rd place in any contest, these two local favorites always beat me out oh well, who said fiddle contests are fair).

Kiru

[Jeez]

Come on guys, share your secrets!

Forever,




Kim.

[gilbertopb]

Originally posted by Jeez What? Does no-one care about this stuff anymore? Don't let this thread get buried! Forever, Kim.

 

 

Oh Lord! We are here, at basics, reading and becoming wonder when listen what happens using that on the keys.

 

About the chords notation, I think a mix of capital letters won