|
|
|
|
|
|
Welcome to the artissimo.gr pages of Blog ! |
|
|
English Language : Music Technology : Blog 
Sunday 14 December 2008 : Appropriate parameterization of the library Special Edition of Vienna Symphonic Library. Part 1: Setting the relative volumes of the instruments |
|
|
About music software A rather bitter history for anyone dealing with orchestrations including many instruments is the setting of the relative volumes: the more the instruments are, the more difficult it gets. There are various problems in the whole process: while playing at the same level of intensity on the keyboard, some instruments can be heard very loud and some other very soft. This result has usually nothing to do with the true nature of the instruments (eg. the bassoon may sound much louder than a trombone: rather absurd, eh!). This problem is solved by changing the volume of the channel assigned to each instrument in order for a balance to come. This balance should match as much as possible the real orchestra conditions (eg. the rolling of the drums cannot sound more gently than a piano from a flute!). Another problem is the different dynamic range the various instruments have that is based on their piano and forte samples, eg. the difference between forte and piano in two different instruments may vary in a great range. Even though one would expect this to meet the specifications of the actual instruments, in practice it is primarily related to the recording of the samples and their subsequent processing: it is simply our world that is not perfect. A solution to this problem is the compromise between the volume of different samples : while playing piano (that is, when playing softly) all should be heard at about the same level, while playing forte, at almost the same level as well. Fortunately, in all the good software synthesizers I have seen so far, different ways of performing with the instruments result in about the same output volume for the same keyboard pressure we apply so we immediately get rid of the burden of further volume adjustments.  However, the situation gets further confused: a) the real ("physical") instruments do not ever play at the same level at their forte and piano (the trombone plays much louder than the bassoon if necessary with both instruments marked as forte at the notation) and b) the samples of the intruments are not always divided into two levels of dynamics (eg. piano and forte), but these can be 3 or even 4 (eg. piano, mezzopiano, forte, fortissimo). How can one tackle all these? The solution to these is simple and practical: one should go to the mixer of the synthesizer and try to put into the higher value (usually this value is the 0 db, beyond this value one can expect phenomena of distortion in a very loud clip etc.) the signal of the instrument that is mostly quiet. This way one can be sure that there will be no problems in the recording. Then, based on this setting - which you cannot change due to technical reasons - you try to set all the rest volumes : you make their volume down so that they can balance with that signal and with all others in an as much as possible musical and matching the specific orchestration way. Eg. the first violins should generally sound louder than the solo flute when the keys are pressed using the same intensity for both channels. The harp should sound very gently, so the guitar - the drums, the trombones, the trumpets and the horns very loudly etc. etc. Of course, various compromises should be made and one will immediately notice that - even after hours of tests and changes in the relative volumes - the result will never be perfect. However, by fixing well the relative volumes for the entire orchestra you can later deal with any other combination of instruments without any problem of this kind. This is because the manipulation of relative volumes is a job to be done when one starts to use a library and then he is having it ready for the rest time of its use (usually long time after that something even better appears and he switches to this solution starting everything from the beginning). As a conclusion, technical problems should be solved as quickly as possible for one to deal with the music problems, which are much more important. |
|
|
|