Vienna Symphonic Library Forum
Forum Statistics

181,802 users have contributed to 42,186 threads and 254,601 posts.

In the past 24 hours, we have 6 new thread(s), 31 new post(s) and 50 new user(s).

  • What comes first? The chicken or the egg?

    last edited
    last edited
    First, a free ad for Beat Kaufmann's 2009 tutorials: this is the missing manual for the Vienna Instrument. Simply brilliant! Thanks Beat for putting them together!

    Now, this leads me to one main question: how do you organise the workflow when you're composing music?

    To me, composing for an orchestra (or smaller ensemble) requires the use of a notation program for one simple reason: I need to see on screen (or on paper) what each instrument is playing or I won't be able to voice correctly. For this purpose I'm using Sibelius.
    Now, even if the playback in Sibelius has improved a lot, after I read Beat's tutorials, I eventually understood why I would not be able to achieve what I want with Sibelius only. I would then have to export my score into Logic for the production aspect.
    My problem is that I'm still at what I would describe to be the dawn of my scoring life, meaning that unfortunately I cannot only write with a pen and a piece of paper: I need to hear what I am actually writing. In other words, I will spend a lot of time in Sibelius to try to get things right so that I'm happy with the music when I play it back. As it never happens, I lose myself into production considerations (including triggering the right articulations as the written articulation isn't necessary the one I want to hear) that should only occur when the piece is written. This prevents me from going any further in my composition, hearing the same 20 bars over and over, not being happy with the result, resulting in me getting tired of what I am working on and eventually giving up.

    Interestingly, this has never happened when working on non-orchestral music. So I blame it on my workflow, which should be:
    1. compose and arrange the piece
    2. mix and produce the piece

    I would imagine that I am not the only person here dealing with this sort of frustrating limitations. So if any of you has some great techniques and pieces of advice to share, please do!!

    Best,
    Denis

  •  FWIW, I am in a somewhat similar place in terms of my own limitations.

    I also start in a notation program (Finale), but while composing a piece simply use the stock sounds avaliable within Finale.  I normally have at least a rough idea of where I want to end up in terms of articulations, and to work on that while composing a piece, I find to be very counter productive.  Only after the piece is completed will I begin to work with VSL sound libraries, by creating a separate staff line (but using the same midi channel) for each part and use that to create all the needed keyswitches.

    After the keyswitches are completed, and I am happy with the articulation choices, I will then merge the two staff lines (the actual part and the keyswitches having been placed in different layers so as not to cause problems when merging), and then import the midi file into my DAW, for much more extensive editing of the musical interpretation.

    At present, it simply is not possible to do the needed detailed midi editing within a notation program.


  • After reading your description, I understand that you are trying to clearly separate the 'scoring' and 'production' stages. Also, you feel that in not clearly separating them, you lose your time and efficiency. Is that correct?


    If the answer is yes, then read on. I believe you should become just a bit more proficient with the 'production' stage. Don't know if it will work for you, but it works for me. I have not much experience with the music production either, just a few years. As I learn the ways to use VSL, or pretty much any other music software, I find the recipes, simple quick ways to make my music sound reasonably well *without* spending too much time. This is important. Indeed, if you lose yourself in polishing a few notes, instead of thinking about the whole score, you lose efficiency. Any polishing should take place only after you're done with the bigger things. But as you learn how to achieve the reasonable sounding quickly, your work in the sequencer like Logic will gradually become similar to the sketching in Sibelius.


    The above implies that you compose in the sequencer right from the start. Of course, there are certain inconveniences in that. Find the ways to overcome them with your software. You might use Sibelius to correct mistakes later, but interchangeably using it with the sequencer is just too much pain, in my opinion.


    And the last point. You don't have to delay the 'production' stage, really, just set up the priorities clearly. You might put some attention to a part that you will re-use later, for example.


  • I often come to the conclusion that keyboard players, like myself, are better equiped, knowledge wise, to compose orchestrated pieces using sample libraries.  Now understand, I'm not trying to be arrogant here I'm just speaking in purely general terms.  I'm sure there are string or wind or whatever players out there who could probably compose circles around me but, generally, keyboard players have it a little easier.  Why?  Because we learned the entire musical range and the complexity of chord structures.  It goes beyond just simple triads as you are only limited by the number of fingers (and sometimes toes) you can put on the keys.  Someone who plays say the clarinet, on the other hand, was trained in only one musical range and probably has a limited understanding of how chords work.  Again, I'm just speaking generally here and I'm not trying to belittle clarinet players.  The point being, if you're not a keyboard player you might be at a disadvantage right off the bat.

    Having said that, all of my works begin as piano arrangements.  For me, if it sounds fine on the keys then it will sound fine orchestrated.  However, it's easier said than done.  Now comes the hard part, for me anyway, arranging and orchestrating.  Do I want to keep it a chamber piece, something bigger, something really big etc.  In the beginning, I would print my piano arrangement out and that became my roadmap.  I would play and/or program the piece in my DAW and that MIDI track became my template for everything else in the piece.  I connected the MIDI track to a simple piano patch just to make sure that the composition and arrangement is how I want it. 

    Now, let's say I want to start with the 1st violins.  I would copy my template track and paste it in the 1st violin MIDI track.  Then I would go in and delete the notes that I didn't want the 1st violins to play.  Next I would assign keyswitches.  At this point, you have a choice, you could record the first violins and bounce the track or you can move on to the rest of the string section.  You don't have to worry about the voicing because you already took care of that in the template.  So go to the 2nd violins, copy and paste the template into the 2nd violin MIDI track and erase the notes you don't want the 2nd violins to play.  Then go on to the violas, celli, DB, etc.  You may have to change your octave range here and there but that's easy just highlight and transpose.  As a technical sidenote, do not connect VSL to the MIDI track(s) until the track is ready to record otherwise VSL get confused and the engine shuts down.  At least it does in Sonar I don't know about logic.

    Let's say you want to do something that's not in the template like add a woodwind arpeggio or something and you're not sure of the voicing.  Well, that's where the paper score comes in handy.  You just figure out what the chord is for that measure or whatever and you base your arpeggio on that chord.

    Another thing I like to do with the template or master MIDI track is create counterpoint that I didn't know I had.  Let's say you have some four part harmony playing.  You can delete everything except the soprano notes here, the alto notes there.  Maybe the bass in one spot here and the tener in another spot there, then transpose everything to the same octave in the range of the instrument you're editing and there you have it.  Instant counterpoint.  It isn't perfect.  You might have to make some adjustments here and there and the results may not be necessarily desired but it works.

    I don't mean to go on and on but this is generally how I've worked since I purchased my VSL libraries.  Only recently have I tried to put myself in the mind of a string player instead of a keyboradist and compose in the string realm.  The result was the Chamber String piece, "Child's Play" which is in my sig line.  A very challanging piece for me as I found it difficult not to go back to the keyboard for harmonizing.  Old habits are hard to break.

    Don't despair Denis the more you work with it the better you'll get   

          


  • What might help you is making the playback in Sibelius sound worse (well, not quite so, but make it sound very different to the end product). Either only use piano samples for each and every instrument. Or use midi sounds, not samples. The trick is that this output gives you the right pitches etc. but shouldn't distract you in the way of wanting to fiddle around to improve it. Of course this approach is not suitable to check instrument colours, blending and such. But maybe it might help a bit


  • Denis,

    Did you try Notion3 or SLE...

    Intuitive and easy to use and learn...

    (The best world betwen a sequencer and a Music engraver)

    I'm like you I need to write music on score...

    I'm actualy composing for orchestra and I'm very pleased with the N3 playback...

    (I'm using a standard notation...)

    So,

    I'm available for more details...

    Alain


  • Notation is nowhere near an absolute language. Notated rhythms, are just signs that require interpretation to get to be musical. The piano roll in a regular sequencer will give you accurate, real durations and timings. I would advise moving to the piano roll and getting familiar with it as a basic composition tool. I can't imagine anyone being happy with the result from a notation application, it just won't get it. Beyond that fact, writing it as a score is a kind of middleman in the process of creating, you are naming things in front of the process. In a real DAW, you can get the actual sound that you will be producing up at the outset of the work. A lot of composers who need scores will do it after composing in the DAW.

    I need the sound to compose, I don't make a clear division between composing and production, except there are some cases where I master from stems as audio-only projects. I'm mixing levels anyway, and I like the sound to really be there, so I mix as I go (I use VE Pro on a slave machine and return a submix to the sequencer host); once the midi is more or less sussed, I fine tune the mix and premaster, deal with the spacial aspect and reverb.room reflections etc... To me, music isn't abstract, and the DAW with a piano roll is inspiring in a way that a pencil on paper or a program dealing with notation after that paradigm never would be.


  • Thank you all for your comments and pieces of advice. I'm back from holiday and all this seems a bit blurred in my head at the moment. But I've got some new ideas to develop and a new classical project, so I'll be back here soon.

  • Hello,

    Essentially, I agree with Dominique. Since I am an arranger/orchestrator for a living and the majority of my work is for live bands/orchestras, I am most comfortable writing to score. For virtual work, the most efficient approach for me is to write the whole piece to score exactly as I would for a live group. Since I am old, I spent most of my life as a paper and pencil guy and you get very used to hearing the music in your head that way. 

    IMO, Notation programs were not concieved or originally designed as true mockup production tools like sequencers and samplers of today. I purposely have not worked on improving the sounds of my notation program (Sibelius). I want to hear the notation, etc., and some approximation of the instruments but I don't spend one second trying to get the 'sounds' closer to a good quality mock up. I actually prefer them to sound a little 'bad' because, for me, it makes what I've written very clear, very in-my-face. I know what the actual instruments will sound like. It's all about writing the piece well first. Then I go to the production stage which I view as a completely different mind-set from the writing.

    I realize that, for some, the two work together but that just doesn't  work for me. I can't think about all the editing and other details of good production while I'm writing. I just write the music the way I always have (except on a computer now) then switch to the production part of the brain. There are times when I've done multiple pieces for a client where I've done all the arrangements first then started on the production. This is how you would do a studio recording of multiple pieces. You would write a number of arrangements, then go to the studio and record them all, then do all of the mixing and finally all of the mastering. You wouldn't write one piece, record, mix and master it, then go back and write the next one, etc. Even with only one piece, I always completely finish the score, with every articulation, dynamic, tempo change, etc., everything, exactly the way I want it. Of course, it doesn't play back the way it will after the production work, just as it doesn't sound the way it will with live musicians. I don't expect or need that.

    I'm certainly not saying this is the best approach. Only that it is what works for me, most likely because of my background before the orchestra sample options became available.

    Be Well,

    Jimmy


  • PaulP Paul moved this topic from Orchestration & Composition on