Vienna Symphonic Library Forum
Forum Statistics

184,988 users have contributed to 42,372 threads and 255,397 posts.

In the past 24 hours, we have 1 new thread(s), 6 new post(s) and 78 new user(s).

  • feedback for piece - question on Sibelius vs. DAW Wars Theme With VSL/243223

  • last edited
    last edited

    First of all, let me post here some thoughts about music with samples produced by a score program or produced by a DAW.

    Take a baroque piece. We all know that baroque music lives from "heavies" and "lights", "longs" and "shorts" or accents and no accents. Example

    I've never got a baroque score where this style was notated! So most of the composers write their music down for the musicians who will play the piece. Of course composers write sfz or legato or staccato. But the sfz will not be the same from case to case. Sometimes it is very short, powerfull and strong and another time it is probably closer to a swell...

    Now when it comes to play the music with a score program a sfz is a sfz, sustain is sustain and so on. When you play the music with a DAW you choose not simply the sustain - as it is written in the score - you choose perhaps the 5s diminuendo because the sound has got a little swell within the first 2 seconds before it fades out. This is maybe just the right sound for this certain 2s-situation. In other words: With a DAW you choose (hopefully) the sound which fits best and not the name of an articulation. Of course you always have a limited choice with sample libraries but believe me if you choose articulations in a clever way you will get always better results this way.

    To show the difference in an clear way:

    Eine kleine Nachtmusik (with Sibelius, unknown producer)

    Eine kleine Nachtmusik (within a DAW, Kaufmann)

    I don't say that one could not increase the quality of the Sibelius-performance. But I believe if you are doing nothing more than using the notes it sounds at it does here. This is very OK because the composer knows very good that the piece will sound good with an orchestra. Now, a lot of the composers are not happy with the result. They would like to have pieces as we know them from the demos here.

    The piece with Sibelius here shows some disadvantages of using a score program for getting top natural results:

    • The articulations are selected by their names and not by their sound
    • Dynamic differences can sound a bit unnatural
    • "Vary the tempo" (in a musical way) is not as easy produce with most of the score programmes and does not sound as natural as a Daw it can do - so this feature often is not used even if music extremly lives from this parameter.

    And finally: First of all samples are "dead" sounds so to say. The artistic combination of them let them bring to life over all. Changing articulations as often as possible (every new note would be great), vary the volume with every new note, vary the tempo, vary the dynamic on every longer note,, and all these things are doing the real music. So a score program is not really the right tool for beeing so flexible with all the parameters above and for reaching top natural sounding music.

    BUT to be fair: It is not the job of a notation program to deliver such top results - even it would be great 😉


    Now to your QUARTET

    Anand, your piece sounds great and I believe you've got out the most of it you could.

    Improvements with a DAW:

    1. Bar: If I would get your score as the first violin player I would make a little accent on the first of the four 16th (f) and also a lttle one with the following next four 16th on the h. Just for giving my 3 colleges the tempo in a clear way. Unfortunately: The start with the legato articulation is very weak and is often not very suitable with fast figures of 16th. So in a Daw I would try whether a sfz on the f and h would do the job or a "fp", or a "sus", or a "marc-legato", or a "dim 1.5s"...

    Going on: In the second bar the 16th (h) is too weak (quiet). Played by a real violinist it would be a more clear situaton between this 16th and the following half note (in a DAW you would increase the volume of this 16th until it fits between this and the following note) in the score programm you should set a ff which isn't really what you want... no problem in the DAW.

    As a further example bar 10: Even if you notized p for the violin - it is too loud (it should be a bit more mysterious as I feel it)... or the first pizz of the cello is too quiet... how ever the balance of the volume between the instruments could be better in my problem in a Daw to adjust all these volume and balance things.

    ...And so on...

    Preparing all those little things in combination with more tempo variations (and a bit a better mix) would let sound your quartett even better than now - I am sure... Listen to the quartet demos of Jay Bacal in the demo section here at VSL (SoloStrings). They can show the difference as well.

    Hope all these infos help to understand what the diffences could be. Important to know as well: The amount of time for treating a piece will not be lower with a DAW but you will probably be happier... 😉



    - Tips & Tricks while using Samples of VSL.. see at: - Tutorial "Mixing an Orchestra":
  • Great advice by Mr. Kaufmann!

    To add a somewhat philosophical angle to the topic: imagine working in a DAW as acting as a conductor at a practice. The musicians have their sheet music in front of them and the the trumpet players know that they're supposed to play let's say, a triple tongue passage. But then you halt for a moment and ask them to play the passage a little more snappier, or a little more vigorous, because you're not quite happy with the delivery and ductus. Or you ask the clarinet player to "hang" on the note for a little longer towards the end.

    Or imagine a band practice where you're playing a jazz tune and ask the drummer to "lay behind the beat" a little. That's a "feel" thing that can't be communicated to a notation program, but can be programmed into the DAW. It's all about coaxing the samples into the behavior you want to hear.

    If you listen to some very good and very convincing sample mockups of talented experts - I guarantee that if the MIDI data was to be simply printed out as a score, several passages in there would be a total mess! Because note lengths, their placement along the meter and dynamic indications would be all over the place and would not always correspond to correct notation. Sometimes you cut a note shorter, or draw it longer, or place it too early or slightly too late or whatever, just because the samples behave, or have been edited, in a certain way and you need to work towards their technical idiosyncrasies.

    In a notation program, you can only specify what notes to play In the DAW, you're telling your "players" how to play it.

  • Thank you again Beat.

    The Eine Kliene examples are like night and day for sure.

    Also thanks for the detaild suggestions on the individual places where my piece can be improved. I particularly like your comment about the p on bar should indeed sound much better if this was much quiter. This should be enough motivation for me to start  devoting time to learn the use of DAW.

    (By the way that string quartet demo by Jay Bacal - Beethovens 59/3 - was the very reason I got into VSL, it was very impressive.)

    Also the points by Jimmy are excellent.

    Best Regards,


  • last edited
    last edited

    Hi, Anand

    while I agree with Beat and many others that a notation program like Sibelius doesn't seem to be a priori the best option to work on fine mockups with virtual instruments, there's absolutely no need to be dogmatic about it. If one is proficient with Sibelius, as you seem to be, you already know that the appearence of the score and the playback information are two completely different things. It's extremely simple to set and modify real durations, start times, note velocities, etc. to each individual note without affecting its notation, and it's also very simple to hide the music symbols that affect playback, and show any other symbol you want (e.g. you can hide an active cresc. and show a inactive dim., have legato samples triggered while hiding an active slur or staccato while showing an inactive slur, etc.).

    I'm not saying you shouldn't learn to use a DAW, you definitely should, but if you are used to a "pencil and paper" approach to composition, feel more confortable and hear music more clearly with notation than in a piano roll view, maybe you can also spend a little more time investigating what's possible in Sibelius to enhance both your programming experience and sonic results. If you know how to add custom symbols to the playback dictionary, and spend a little time learning how to make your own soundsets and playback configurations, you can do whatever you want.

    I personally prefer to use Sibelius for the score and the mockup because having both in the same file saves me a lot of time. As an example, Here is a mockup done entirely in Sibelius. Although my mixing skills are not comparable with more talented users in this forum, I think you can nevertheless hear what I mean in terms of musical expression and programming flexibility (the score not being affected at all in the process)

    Hope it helps you form an unprejudiced opinion about it.



  • Hi Servando

    That was an awesome mockup of 'catch me if you can'! Thanks for sharing.


  • Servando that would have been a great simulation for DAW, but incredible for Sibelius!

    Do you have a private e-mail that I can write to you about something? You can reply to me through VSL's Private Message facility if you don't wish to make your e-mail public.

    Best wishes,


  • Thanks Anand and Errikos (you have a pm, btw).

    Sticking to Anand's original question, what I'm trying to say is that many times, when people explain why a DAW is the best (or even the only) option to produce orchestral mockups, they tend to use many innacurate (or plainly false) statements which can very easily lead to all kinds of wrong, prejudiced conclusions. 

    For me, it's just a matter of personal preference and/or appropriateness for a particular type of work; I will never ever use a DAW for an orchestral mockup with VSL instruments if I have/need the score... nor will I ever use Sibelius to mockup a saxophone quartet with a breath controller triggering other instruments which use a very different technology not based in crossfading sampled velocity layers (as I'm doing now). I simply choose the easiest, quickest way for a particular job, depending of its nature. That's what I meant by saying "there's no need to be dogmatic about it", and that's why I recommend that you learn to use both your DAW and notation program of choice to their fullest potential.

  • PaulP Paul moved this topic from Orchestration & Composition on