I am taking a class in school called Inquiry Quest, which is an independent research project that takes an entire semester. Because of the freedom to choose any topic I could think of, and because I had been planning for a while to compose for orchestra, I decided my topic was going to be orchestral composition.

So, for this project I am comparing two methods of composition for orchestra, with the goal of deciding which method is more effective, and which results in more interesting/effective orchestration. I am composing a piece using each method, so that I can come up with a theory based on my results.

The Two Methods:

The first method is to compose the piece on piano, and then orchestrate the result using a music notation software (like Finale or Sibelius). This method is the more natural method for me because piano is my main instrument and most of my compositions have been for piano. I have already finished composing this piece, and am almost finished orchestrating it. (Of course there is still loads of editing that has to happen).

The second method is to compose the piece using Sibelius from the start -- so a piano is not used at all for the generation of ideas. I thought this method would be a lot harder, but now that I've started, I'm realizing it is actually just as possible as the first method. For this one, I've only composed about 30 bars so far, but it will be progressing quickly in the coming days. 

I have been collecting data about each of these methods as I compose, and there are clearly benefits and downsides to each method, but I am interested in hearing your opinions about this topic.

If you compose for orchestra, which method do you use or prefer? Are there specific reasons for your method besides the fact that you are more familiar with it? Which method do you think is more likely to result in more effective orchestration? 

It would be great to hear any other ideas you might have!

You need to be a member of Composers' Forum to add comments!

Join Composers' Forum

Email me when people reply –


  • This is a tough question.

    The advantage of composing on the piano first, as you said, is that it's easier to work with if you already know how to play the piano. It also lets you focus more on the music itself, as opposed to getting lost in the tedious technical details of orchestration such as instrumentation decisions, doublings, timbral issues, etc.. However, the disadvantage of composing on the piano first, is that the resulting music may be too "pianistic", i.e., it sounds idiomatic on the piano but hard to translate effectively to the orchestra. This is especially an issue if you rely a lot on piano-specific idioms like using the sustain pedal, piano-style arpeggios, chords that your hand can span, etc.. Orchestral chords generally have very different spacing considerations, and piano-style arpeggios and pedalling simply doesn't translate well to orchestra at all.

    The advantage of writing for orchestra directly is that you are working directly with the timbres and instruments of the final work, so the  music will be idiomatic to the chosen instruments, and you get to work directly with the various timbres in a way that you can't work with effectively on the piano. A good number of orchestral masterpieces were composed for specific instrumental colors, that would not make much sense on the piano at all. However, one problem with writing an orchestral score directly is that it's easy to get bogged down by the copious minutiae that you have to deal with when orchestrating: which instruments to assign to which melody, which timbres blend well, which instruments can play effectively in the assigned range, the distribution of tutti chords, etc.. All of these details can easily consume all of your attention and cause you lose track of the larger picture of the music itself.

    Personally, I lean somewhat towards writing directly for orchestra, but I find that the most important thing is to first hear the music in your head, so to speak, before you start writing notes on your orchestral score. In the initial stages of composition, I find it more important to figure out where the music is going -- to fix the melody in my head, to figure out the sequence of passages, what are the important motifs, gestures, etc.. For example, my initial inspiration may be a short motif or rhythm; my first order of business is to figure out what notes to go with it, to make a melody. I may use the piano to figure out the exact pitches, if I'm not 100% sure; but I try to limit the piano to just helping me find pitches or melody lines -- I try to avoid going into filled-out full piano accompaniment, because that's where piano idioms will creep into the music. The main work is to make the music clear in my head.

    As the piece starts to form in my mind, I'll have a clearer idea of where it's going, and I may hum it to myself, or otherwise "hear" it in my head. At this point, I may start assigning instruments to the music -- not the complete orchestral score, but a general idea of what I want it to sound like -- e.g., the first passage should be played by the strings, this and that chord will be tutti, the next passage features a trumpet solo, etc.. In other words, I assign instruments or instrument families to the main gestures and melodies, but the actual details of exactly who is going to play the accompaniment, who will play the bass notes, the tutti chord distributions, etc., are not decided yet.

    At this point, I might start working on the so-called "short score": 2 or 4 staves, or maybe 6-8 depending on how complex the music is, where I sketch out the main motifs and melodies, with various annotations as to which instruments will play which lines. This is generally much easier to edit than a full orchestral score, and is very useful in early stages when you're not 100% sure about exactly how the piece will proceed yet. Instrumentation can also change easily just be erasing/rewriting an annotation, rather than a complicated cut-n-paste operation while trying to decide where the move the accompaniment figure that the target instrument is already playing, etc..

    It's around this time, that the main melodies and overall direction of the piece gets set, and I might start working on the full score. Since I already know, by this point, what the melodies are, and which specific instruments I might want to play in which passage, it's much easier to make orchestration decisions. In other words, I've already mostly worked out the compositional issues, so I can focus my attention more on the orchestration issues. Trying to do both at the same time is very difficult!

    Of course, generally the process isn't as simple and clear-cut as "step 1: compose, step 2: orchestrate". Often during the orchestration stage I may decide that what I thought was a good idea wouldn't work after all -- maybe the instrument colors I have chosen doesn't work out well, or the instrument I had in mind would be out-of-range during an important moment in the melody, or maybe, after hearing the piece in the DAW, it becomes clear that my original conception of it isn't working out. So then I have to go back and re-imagine a certain passage or section -- so I might take some time off working on the score to explore alternatives in my mind first, before settling on a prospective solution -- then come back to work out the orchestration.  Similarly, sometimes during the compositional stage I may have a hard time keeping track of all the details in my head -- I might need to hear a sample of the passage to decide how to proceed, especially if it's a highly-complex polyphonic passage. In this case I might "cheat" and go straight to writing the full orchestral score for that particular passage so that it can be rendered by the DAW. Sometimes hearing a fully-orchestrated passage, even if it's just a tiny fragment of the full piece, may inspire you with new compositional possibilities, and may cause the piece to proceed in a different direction than originally envisioned.

    So while the general sequence is to compose first and orchestrate second, it's not a cut-and-dry formula, and during composition the two may intermix and play off each other a lot. At the end of the day, it's really about using the piano (or short score, or whatever) as your compositional tools, to help you work with the music, and not get bogged down by too many details -- esp. if you're writing for a large orchestra! During the compositional stage, you don't really want to be worrying about which doublings to use, which instruments are out-of-range, etc.. You may decide on these things for the main themes, but it's too tedious to do that for everything down to the last detail of the background accompaniment. First, you have to sort out where the melody is going, where the music will end up. Once the big-picture decisions are made, then you can sit down and focus on smaller details of orchestration.

  • An excellent and very comprehensive reply, Teoh!

    Well, what can I say? I agree and feel well covered.

  • Thanks a lot for this reply!

    It's a great inspiration for me to hear how others compose, and there is a lot I can learn from your process of composition and orchestration.

    H. S. Teoh said:

    This is a tough question.

    The advantage of composing on the piano first, as you said, is that it's easier to work with...


    Is there a superior method of composing for orchestra?
    I am taking a class in school called Inquiry Quest, which is an independent research project that takes an entire semester. Because of the freedom to…
This reply was deleted.