Vienna Symphonic Library Forum
Forum Statistics

200,732 users have contributed to 43,209 threads and 259,126 posts.

In the past 24 hours, we have 3 new thread(s), 11 new post(s) and 59 new user(s).

  1. Home
  2. Sample Libraries
  3. Vienna Instruments Libraries (legacy)
  4. A suggestion for 'dynamics' sample lengths

  • A suggestion for 'dynamics' sample lengths

    C
    C

    First of all, I love most of the libraries and have collected quite a few now.

    One thing that sort of annoys me (only a wee bit) is the non-uniform length of the samples in the dynamics section. Some last 1,5 seconds or 2 or 4, others are way longer. It would be great if ALL instruments (as far as it is physically possible of course) have dynamics (both regular and pfp) in 1,5, 2, 4, 6, 8 second sizes, both vibtato and non-vibrato. This way it lends itself for many more combinations without trying to work around it/faking it.

    wkr

    Janco 
  • W
    W

     The dynamic samples are largely uniform, and do not vary a huge amount. They tend to be 6 or 8, 3 or 4, and 2 or 1.5.    You might think if the samples were all exactly the same length everything would be easy.  But actually it wouldn't. 

    Because you will find that each musical phrase you are applying a real time dynamic to will vary in needed length on a moment-by-moment basis depending on the music and the tempo.  So if you have everything exactly the same, that still won't make it match perfectly.  I have often adjusted start and stop times of notes because of this. You can also add expression at the beginning or end of a note. But the best way to deal with it is with crossfade dynamics.  This works so well with any ensemble artticulation that you cannot tell the difference between them.  I once tried A/B-ing real dynamics with crossfade and they were indistinguishable.  For example, if you program a sudden mod change on violin ensemble, ramping down from full in a half second to minimum, you instantly have a sforzando that can then crescendo back up at any speed.  So the ensembles can be used for any dynamics with crossfade.  Also, though many people believe you cannot use crossfade on solo instruments, actually you can usually do it in an orchestral context, because the phasing that sometimes occurs between layers (only on certain bright sounding instruments like trumpet, oboe, etc.) is not usually audible unless it is in an actual exposed solo. 

    One other thing that can be used is of course the time stretching facilities of VI.  I haven't even had to use that because of the combination of crossfade with real.

    The main point is that using real pre-recorded dynamics is never going to be perfect and will require compensations of various kinds even if you have totally uniform timing in the samples.
  • C
    C

    I second that. Cross-fade dynamics (controlled by the mod wheel can produce highly effective crescs and dims.
  • C
    C
    I agree on the crossfades too, but it would be nice if there was just one more dynamic layer because a lot of times trying to get to mp or mf with crossfading is a bit tricky. Otherwise I too find it more useful to crossfade than to use the dynamic patches.
  • W
    W

    You are right that having lots of dynamic layers helps.  I believe that a good approach is to use as much actual dynamics as you can on solo instruments.  It gives an extreme realism, especially when heard together, if all the instruments are playing real dynamics.

    However, on ensemble, it is almost not even audible, the difference between crossfade and actual dynamics.     

    One trick that Herb talked about a long time ago on the Jerry Goldsmith Parachutes demo which involves dynamics, is to use dynamic samples where you don't really need them.  That sounds weird, but if you analyze a real musical line, you will notice many times how players tend to descrescendo on each note a little.  This is represented perfectly by using a slightly too long dynamic sample - for example, a 3 second diminuendo on notes that are about 1 second long each.  They won't sound totally uniform that way, even though it is not for creating a true diminuendo.   I have also noticed this with sforzandos - if you have a fortissimo line, you might want to try the sfz or sffz articulations rather than pure ff samples, because that is another thing that even if not called for in the score players will do - they hit a note really hard, but do not sustain it at that full volume.  In many cases it is physically impossible to sustain the note at full volume, and the attack gives most of the character of the note.
  • C
    C

    Ah...of course...why didn't I think of that before? :)

    It makes the dynamic curves way more flexible. I now tried it with different patches and so far no phasing, works like a charm!

    J
  • C
    C
    Those are some good suggestions William, personally my style is to use crossfade patches and move the modwheel as it sounds natural, which means that practically speaking the modwheel is never standing still in my recordings. So I agree with what you say but I don't use dynamics patches to get there personally.
  • E
    E

    The way I understood the post originally was a request to have *all* of the instruments with their dynamic options fully in tact.

    I have noticed that some instruments do not have 6 or 8 seconds. The solo viola for example, has 1.5, 3 and 4 seconds. I certainly would have liked to see more choices. I belive some instruments have 2 seconds, but not all of them.

    I do not expect the instruments to match exactly, but rather to have more consistant choices in lengths avaliable  so I can choose something much more suitable to a particular line.
  • C
    C

    esperlad, I've never heard a violist play a p-f-p for eight seconds.  VSL took the capabilities of the instruments rather strictly into account, and the truth is some instruments just can't do things the same way as others or for the same length of time.
  • E
    E

    I went through several instruments in order to check their sampled dynamic length. The longest pfp I found is six seconds for the orchestral violins. I believe the cresc./dim. patches also play up to that length.

    I checked many of the brass and wind instruments. The English Horn I, Oboe d'Amore and Clarinet in Bb play a pfp that is 10 seconds. Many will play up to 8 seconds.

    It would be helpful to know the maxium capabiites of the maximum dynanic lenghts.

    The point I am trying to make is that I would like to see a more consistent set of dynamics. Not just pfp, but standard cresc. and dim. patches. 8 seconds may indeed be unplayable, but several of the shorter lengths *are* playable. They are just missing.

    I hope we can get some clarification on some of the missings dynamic lengths.