Greetings! I have been recently starting to score some of my pieces but have realized that I'm having trouble assigning dynamics and especially how the dynamics might affect the other instruments. For example if one instrument is too loud it might drown out another instrument. So I guess what I'm asking is there a general rule or tips/ tricks when adding the dynamics of a piece—like how loud should the melody or secondary melody be in comparison to say the rhythmic or supporting structure? What about percussion or brass, is there some rules about that? I know quite a few questions... Even a resource that can help me with the process. Thanks!
You need to be a member of Composers' Forum to add comments!
Replies
Thanks Bob!
Your absolutely right, the best experience would come from the 'hands on' kind. I have practically no experience with a live orchestra. The only thing I can go by is the virtual instruments sound. How that might translate into a live performance will truly show me what is right for what I'm writing. As of now I don't have immediate access to an orchestra to play my music but I have Sibelius and want to start writing my music down as I only have it in audio format via EWQLSO. Right now I don't really know what a pianissimo sounds like played by a real orchestra or a forte etc. I'm eager to learn all I can, but thank you for taking time to help me out! If you have any other suggestions I would love to hear it, thank you.
I would highly recommend listening to more orchestral works (by live orchestra, not sequenced). If you cannot afford or it's otherwise inconvenient to attend live concerts, I recommend listening to as many recordings as you can for each work.
The first thing you should do is to cultivate a healthy distrust of the sound your computer produces -- because what the computer produces is often very different from what a real orchestra sounds like. Always assume the computer is inaccurate, until proven otherwise (e.g., by comparison with live performances or recordings).
The second thing you should do is to also cultivate a suspicion of recordings: many recordings are the sound engineer's concept of what the orchestra should sound like, not what it actually sounds like. There is a difference. Because of this, it's highly recommended to listen to multiple, different recordings of the same work. That will help you develop a more accurate idea of what the instruments actually sound like. Don't trust in what you hear from a single recording, because it may not represent the real sound.
Get hold of the scores of your favorite orchestral pieces -- you can find a lot of free scores on sites like IMSLP (though those tend to be older orchestral works, because of copyright issues) -- and then follow along the score while you listen to the piece. You may want to listen to the same recording several times, each time paying attention to a different instrument in the score, to learn how the composer makes use of that instrument, and how it sounds like. Then do the same thing with a different recording (even better if it's a different conductor, so you aren't biased by one particular conductor's interpretation of it), and compare them.
I suppose this is far more than you asked for -- which is how to write dynamics correctly -- but writing dynamics is one of those things that everybody has a different opinion about (some people recommend writing brass dynamics a shade below other instruments, others recommend writing the same dynamic throughout and let the conductor/performers work it out). At the end of the day, you just have to figure it out for yourself by studying what other composers have done. And in the meantime, you will have also learned other aspects of orchestration that will probably help you a lot too.
Note that you should never write dynamics just to make the computer output sound good (unless you're specifically composing just for the computer). Such a score, if played by a real orchestra, will probably be completely different from what you have in mind, because quite often, the instrument patches used by the computer were recorded in two different places, with different recording equipment placed at different distances from the instrument, so their relative volumes won't match, thus requiring you to write strange dynamics just to make them balance. But real players following that kind of skewed dynamics will only make a big mess of your music. Instead, you should always write dynamics as if for human players, and then if the computer doesn't do what you want, go to the instrument balance / per track volume controls and fix the problem there. The computer should be configured to interpret dynamics the way you want, not the other way round.
"cultivate a healthy distrust of the sound your computer produces" Love it! It is a very scary yet exciting thing to think about how different your music will sound once actually performed. That is something I have yet experienced but often day dream about. Thank you Teoh for your great insight! Trust me you didn't in the least give me too much info the more I can learn the better. Those were some really great tips! I really found "but writing dynamics is one of those things that everybody has a different opinion about... At the end of the day, you just have to figure it out for yourself by studying what other composers have done." That was really cool! If anyone else has more to add I would love more knowledge! Thanks.
One of things my conductor said to me was that part of his job was to find the melody.
He will often ask a player to change the part's dynamics so as to be heard or not heard so much.
Strings generally play everything mf given the chance, even in good quartets. Have to practice playing pp.
"The first thing you should do is to cultivate a healthy distrust of the sound your computer produces -- because what the computer produces is often very different from what a real orchestra sounds like. Always assume the computer is inaccurate, until proven otherwise (e.g., by comparison with live performances or recordings)."
I have to disagree with that.
I think you should assume that your computer creates a sound environment or sound universe that is thoroughly and totally real, in its own right.
So does an ensemble in a concert hall.
Which one are you working with?
A "recording" of a computer audio file (played over the internet, through your speakers) may not even sound that much different from a recording of a live orchestra (played over the internet, through your speakers). Consider how far removed you are in front of this computer screen from the "source" of the recording, whether it is an audio file or a live ensemble. Think how many "degrees of separation" there are.
If you are working with a live ensemble in a concert hall, you (or the conductor) can tell a certain instrumentalist to play more loudly or more softly, to get a certain result.
If you are composing with a computer, you compose exactly as you want it to sound, to produce the best aesthetic affect. You can adjust every feature of the instrument sound, including velocity (loudness), pitch, note duration, timbre and many other features.
If many of your listeners are going to be hearing your piece, via the internet, and from an MP3 (that you eventually post on YouTube or Soundcloud) then put your creative energy into that sound file. [Don't rely simply on midi's which compress the data too much.]
The score only becomes relevant when you know for certain that you are actually going to have a live ensemble perform your work. You may ultimately get more people to hear your work, via the internet (on specific sites, like YouTube or Soundcloud) than you will in a concert hall, so why not take the time to make sure your sound file is a good one? Don't treat it 'with skepticism,' any more than you would treat your piano or violin 'with skepticism'.
I don't see the point of making a poor sound file, and saying, "oh, it's just a sound file; people will hear what my work 'really sounds like,' when they hear the live performance" (which may or may not happen).
The sound file is an artwork, in its own right, and if you don't care how THAT SOUNDS then can you really say you care about your music?
One might say the same about the score, which is a kind of work of art. But no one actually "hears" the score, except for a few who might read it carefully and grasp its meaning.
Of course, if you already do have "live musicians" flocking to perform and record your work, on stage or for recording companies, then you don't need the advice of anyone on this site (presumably).
But if you are working on your music, for its own sake, and to produce music that your friends and other musicians can sample, I strongly suggest that the sound file should be taken seriously, and not just thought of as something that "doesn't sound the way your work will really sound" when it's performed.
If the truth be known, even the great composers had actual live performances where the work "sounded nothing like what I intended it to sound like," and that is due to many factors, like insufficient rehearsals, too many rehearsals, conductor philosophy, quality of the orchestra, failure to interpret intent, inability to comprehend the composer's meaning, bad luck, bad digestion of the conductor, failure to pray to the proper deity in an acceptable fashion, etc. Marking the dynamics of a passage may or may not make any difference in some of these cases. The acoustics of a particular concert hall may make a completely different set of markings necessary.
So let's not "cultivate" any skepticism about the sound produced by the computer. It creates sound, and it's what most of us are listening to right now. [You are sitting in front of a computer, aren't you?]
If you should have any attitude at all towards to the computer, is should be confidence, as time goes on, that its ability to make more and more precise sounds, that correspond your wishes regarding the final product, will improve over time, year by year, and even month by month …
"Live performances" are what they are, and computer sound files are what are—they also convey the pleasurable sounds you may wish to hear. And even most of the so-called "live performances" we hear cannot be said to genuinely live when they are 1) recorded (first degree of separation), 2) transmitted over the internet (second degree of separation), and finally 3) poured out into your living room, via your set of speaks (third degree).
Enjoy each set of sounds, and do your best to make whatever set of sounds you produce (by whatever means) have the effect that you think is most beautiful and satisfying.
I agree with Ondib's "general philosophy" on sound whether humanly or machine produced and I think it raises a lot of questions to both approaches. Still, I think that Bob and Teoh gave god practical advise on the human side of things.
Ondib, all you say about sound files is good and valid in my opinion and I wish I had the time to follow your suggestions. It would take me more time and very long learning curve to improve my sound files. Actually much more than to write the music itself, so I dream for the day that affordable technology can deliver the goods for all of us who study and compose music in a traditional way. Unfortunately the education planers did not foresee the need for studying also studio techniques and sound manipulations.
Even so, now that music courses have been updated to include such studies I found that most new students who follow such studies are missing on the musical side of things. I mean, they may know how to tweak sounds and to balance stereo files by panning and a lot of other tricks, but still ignore the basics of a SATB harmonization or imitating a melody in the 5th in strict two part counterpoint, etc, etc.
So if the music is not up to standard in the first place, what's the use in having a perfect sound file for it?
The old computer adage "shit in, shit out" still holds strong.
There is simply no time to study all and less time to do it all by one single person imo.
What do you say?
P.S.
It would be nice, me thinks, if this forum was joint by sound engineers whose aspirations lie in the field of producing perfect sound files. They could take up our miserable sounding scores and render them perfect with their magic!
From my POV, any manipulation of the sound file is secondary. The composer's primary concern ought to be with the music itself. This does not mean the quality of the sound file is not important, of course; but the point is that if the music isn't that great to begin with, no amount of sound file manipulation is going to make it better. If the music is great, but the sound file is poor, then we can talk about how to improve the quality of the sound file.
I found the following resource helpful: http://music.tutsplus.com/tutorials/the-ultimate-list-of-tips-for-p...
It backs up what Bob said, that while the default string patches in your software may not sound that great, it is possible to improve it quite a lot without resorting to shelling out big bucks for the biggest, fanciest string library out there these days. (And even if you did shell out the bucks for one of those libraries, it still depends on whether you know how to use them effectively. If not, you may not get a result that's significantly better than a plain ole MIDI patch.)
H.S. said,
"From my POV, any manipulation of the sound file is secondary. The composer's primary concern ought to be with the music itself."
I don't quite understand that. Why isn't the sound file to be thought of at all as "the music?"
What you say seems to me like telling Bach, just before his solo performance on an organ, "Don’t' worry about the instrument, or the sound it makes. Your primary concern is the music itself."
But what the organ gives us is, in an important sense, "the music," just as what the sound file gives us is also "the music."
Or, to give a more modern analogy, it would be like telling Stravinsky, as conductor, and his sound engineer, before performing the Rite of Spring,
"Don’t worry about the recording, or the microphones or the acoustical effect produced by the orchestra, or how things sound on tape. Your main concern is the music."
The recording is the music, just as the sound file is "the music." And why not?
Perhaps we are not aware of the changes that are taking place within the minds and habits of large numbers of composers. In fact, at least one new member of this forum, in his profile, emphasized the importance of the computer as his primary tool for composition and the creation of the musical product itself.
"Your main concern should be the music."
So perhaps you could explain: Why isn't the sound file "the music?" If it isn't why isn't it, any more or less than a score, a particular recorded performance, or the idea in the composer's head?
From a philosophical point of view, we can say, the actual "music itself" doesn’t exist in the recording, the score, the performance, or even in the mind of the composer. The essence of the music may reside in an eternal world, resembling Plato's "realm of ideas," along with mathematical and musical entities that have their own independent existence, apart from all material and even mental manifestations. But I am not sure you wish to look at this philosophically. In this post, I am just examining the nature of the physical entity we call a sound file, or a "recording, something human beings hear and experience phenomenologically.
You said,
"This does not mean the quality of the sound file is not important, of course; but the point is that if the music isn't that great to begin with, no amount of sound file manipulation is going to make it better."
I don't understand that either.
That's like saying to Tchaikovsky, if your first draft of your score isn't "that great to begin with," then no amount of revision, or manipulation of the themes, or the orchestration, or development will make it any better.
It doesn't seem to me that you believe the sound file can start out sounding one way, and then become a completely different and far superior piece of music, at the end of a long process of revision and reworking—perhaps even more effectively than what is produced after the revision of the score. I think you see the sound file as a mere means to an end.
What you have said is like saying, "the music sounded bad during the first rehearsal, which leads me to believe no amount of rehearsal will make this sound any better, no amount of changes in the score, hiring a new orchestra would do no good, nothing can help this." You do not appear to be considering the premise that all music starts out in an embryonic form, and develops with all kinds of alterations, and that most of these, perhaps nearly all, or all of these, can take place through the development of a sound file.
Everything can be changed in a sound file: the complete form of the work, the melodies, the duration of notes, total orchestration (from one instrument to well over one hundred instruments, of every conceivable description), harmonies, the tuning, timbre (altered radically or with subtlety), absolute and relative volume for each instrument, attack, decay, how long any individual note is sustained, all tempos and rhythms for each bar, key signature, mode (from major to minor, from Phrygian to Mixolydian, or the invention of new modes a la Messiaen), quantization, syncopation, the creation of new instruments (through sound sculpture), not to mention the countless synthetic effects of old-fashioned or new methods of electronic sound production.
The sound file is sometimes just seen as semi-convenient way of "giving an idea" of the sound of the piece, which is subordinate to the score, or the so-called "live performance" (which, as I said, is no longer live, when we hear it, but mediated by the recording, the transmission over the internet, reception by your computer, and sound effects produced by your speakers or headphones).
It seems odd to me to treat the sound file so lightly, especially when such a large number of us are hearing the vast majority of pieces (written by composers here) via some kind of sound file, produced by a computer program, such as Sibelius, Logic, Cubabase or even Garageband, to mention just a few.
The world has changed, and the way we compose music and listen to it has vastly changed, and it is changing even more rapidly, as we engage in these sorts of conversations. Shouldn't that fact be taken into account?
Haha my favorite conductor used to say the same exact thing to us about finding the melody in the piece. That is so true. With orchestral composing it can be challenging to write and constantly be going back and adjusting sections and parts for dynamics/expression etc., and sometimes it can kill the vibe (or make it) having to labor over this...but I find keeping a tidy house from an organization and dynamic standpoint throughout the creation of your cue, is very helpful and can be helpful for monitoring your part's dynamics.
Michael Lofting said: