Similar behaviour here with the french oboe. A couple of years ago the outcome of a discussion on this forum (or with vsl, I don't remember) was to try to avoid the notes where this happened too prominently,

regards

Gerard

Not the notes, but the x-fade velocity values. Usually it is only a very small range in which actually 2 samples are heard. So, say, at vel=70 you hear the mf layer and when you raise the velocity value it simply gets louder, until you reach, say, 95. By further raising the vel value the ff layer comes in, while the mf layer fades out, at a certain point, say, 105 you only hear the ff layer.

All you have to do is to decide, which of the layers matches your desired level of expressiveness, then x-fade either to the highest mf value, or the lowest ff value.

@wrathy said:

...I am hearing "two" oboes rather than one. If you release the note you are...

Hi

Unfortunately this is correct.

If you increase the level for example X-Velocity mixes the sound from one layer into the next.

Further, becauses this fade needs to be soft is always a part where two layers play the note.

So you are right: There is a part with two oboes so to say. You can't do anything against this effect.

X-Velosity-fade works good with ensemble sounds because there are already more instruments and that's why you can't recognize this blend effect as good as with the solo instruments.

But often this "fault" with solo instruments disappears a bit in the context of a whole orchestra.

Beat

@wrathy said:

[...] 2) Possible workaround, or "just live with it." [...]

The way these instruments are designed, you're actually supposed to use the real recordings of fp, sfz and the like (a.k.a. "Dynamics"). That way the performance will consist of the most authentic building blocks. ... using crossfades is easier, without doubt. 😊

-> http://www.vsl.co.at/en/211/496/802/819/1096/706.htm

HTH,

Without sounding like a broken record, I'll once again suggest using Lemur. You can create a multiball that controls Velocity XF on one axis, and Expression on another (as I do on my Youtube video "Composing using Lemur and sample libraries", at around 04:55). That way, you can decide what type of volume automation sounds best. I often use a combination of the two. It's far more than a workaround...it's actually a better way to work. It's great for quiet but heavy brass pads, like those found in Strauss's music. 

HTH

MOH

P.S. I agree with Dietz. If you are doing a quintet with lots of exposed solo instruments, you should as much as possible use the dynamics patches. Exposed instruments are the hardest to make sound realistic. I stopped trying...

I'll just echo MOH's comment, not necessarily about Lemur, but just about switching to using expression rather than velocity crossfade when the doubled sample issue bugs you. It's not ideal, but you just have to pick which you can live with.

@mohurwitzmusic said:

...P.S. I agree with Dietz. If you are doing a quintet with lots of exposed solo instruments, you should as much as possible use the dynamics patches...

I agree with Dietz as well. And when you support these Dynamic-Samples a bit with the Volume you will get excellent and really natural dynamics.

If you are using VIPro  you are able to adapt the diminuendos and crescendos in the length...

Listen to the example below. It is only played with such dynamic samples...

http://www.beat-kaufmann.com/VSL_New_VI/BK_Bach_BWV_622_O_Mensch_bewein_VI08.mp3

1st part: Solo Strings, Viola (0:00 - 1:11)

2nd part: Solo Solo Strings, all (1:12 - 2:17)

3rd part: Chamberstrings (2:18 - 3:02)

4th part: Orchestra Strings (3:03 - 3:44)

5th part: Appassionata Strings (3:45 - End)

The trick is, that you not always need the whole sample length because...

Normally Diminuendos start with a little crescendo first > so use just this first part of the sample

Or, it is not necessary to use Crescendos until their ends

... it is really a play with all the samples. Important: Don't chose the samples because of their names but because of how they sound.

This is another piece, which I played in 2006. Listen mainly to the strings. I only used Dynamic Samples as well and no X-Velocity ... because I didn't know them at that time [:$]

http://www.beat-kaufmann.com/VSL_New_VI/BK_Betlehem_Strings_VI_06.mp3

Hope these pieces give you some new incentive to try it as well with dynamic articulations for bringing life into the music.

Best

Beat

(Edit)

Appendix

Sometimes I used the possibility of blending from sample 1 into sample 2. Example: You can start with a crescendo and after a while you blend into a sustain sample.

Fact is, that nice music with samples always needs a lot of work.

So for good results you should have the sounds of your samples (in all their layers) in mind and you also should have a lot of time and patience.

Lemur?

Nothing against Lemur for getting a first result with it.

But when it comes to the "fine-adjusting of samples" and bring them to their "maximum music" Lemur can't do magic.

My experience is that one has to be very close to each sample for such tuning tasks.

The tools you need for this are all ready for use within VI.

So why programming a controller first that it shall make a diminuendo longer when it is possible to do this directly within VIPro as well.

Today I don't use any controller - even for "drawing" velocity curves - because I always had to fine tune them afterwards by the mouse.

So today I use the mouse just from the beginning and I believe that I'm - nevertheless - faster than in the past.

In other words:

The answer from my side to the question "Lemur?" is:

Why not trying this unit but don't believe that it will solve all your problems.

Sure is that it will take another bunch of time on the way to good music with samples.

Musical results need the chose of the "right samples" in any moment and this for each instrument in the orchestra, tempo variations, a natural dynamic, an arrangement which goes along the sample possibilities, later on of course a great mix, the perfect use of effects for treating all the instruments and finally a correct mastering job.

3 minutes of music can lead to 5-10 days of work. If all things are perfect the result is unspectacular in a way it appears just as MUSIC.

Try to find out where Lemur can help in the upper listing. Important isn't one more tool (even if it looks very nice) but a lot of knowledge "how to do all these different tasks".

Beat

The point was that if you want to avoid using dynamics samples and just stick with expression and/or modulation, Lemur's multiball will help you find some combination of the two controllers that works. 

MOH

Pyre,

Sorry I don't know too much about VSL's own iPad app. Lemur is set up to deal specifically with my workflow. I use it not only to control my VSL samples, but also to control things in Cubase directly, adjust faders, adjust monitoring levels in my studio, control guitar FX racks, control synths, make my coffee (kidding, but that would be great...). So in that way, it's far more powerful because it's customizable. Steep learning curve though...

HTH

MOH

On most solo instruments the crossfade will create some chorusing.  However this is mainly because of the programmer being too lazy to do dynamics the right way - with dynamic samples. You will obtain a far better and more realistic change of loudness. Also, using Note-on velocity changes combined with CC 11 expression can be done for many situtuions (an example being how timbrally a mf sample sounds o.k. for either mf OR a p sound because it is not too bright and with simple amplitude adjustment can be used as either). 

You can often use the crossfade anyway if any other instrument is playing because the chorusing starts to disappear - or become very hard to hear - when anything else is playing. 

 One other thing about velocity crossfade is that it is a complex situation in general that sometimes works poorly  and other times excellently.  An example of it working perfectly is  if you have three solo instruments playing together.  Though each one may chorus a little, the ear cannot distinguish where the chorusing is coming from and so it will sound perfect. In fact, I A/B'd actual dynamic samples on each of the solos compared to velocity crossfading and they were not distinguishable. 

Also, the oboe is an instrument particularly suceptible to chorusing in crossfading because of its extremely bright thin sound. If you do the same thing with a flute it will not be detectable.  Another example is the recent Recorder library which has absolutely perfect crossfading at all levels.

I think there is really nothing to gain from velocity cross fade for the solo instruments, actually. you have the velocity from VELOCITY and if you have the dynamic held notes... I use CC11 and track volume otherwise. this is an idea I keep seeing talk about but I think it has no real use for the solo instruments. I'm baffled that people believe there is something here.

@civilization 3 said:

I think there is really nothing to gain from velocity cross fade for the solo instruments, actually. you have the velocity from VELOCITY and if you have the dynamic held notes... I use CC11 and track volume otherwise. this is an idea I keep seeing talk about but I think it has no real use for the solo instruments. I'm baffled that people believe there is something here.

@mohurwitzmusic said:

Without sounding like a broken record, I'll once again suggest using Lemur. You can create a multiball that controls Velocity XF on one axis, and Expression on another (as I do on my Youtube video "Composing using Lemur and sample libraries", at around 04:55). That way, you can decide what type of volume automation sounds best. I often use a combination of the two. It's far more than a workaround...it's actually a better way to work. It's great for quiet but heavy brass pads, like those found in Strauss's music. 

HTH

MOH

@wrathy said:

...but how do you suggest to create a line of say two bars of 16th notes or 8th notes that cres/dim over those two measures?

With short notes I always use the midi velocity.

Advantage: You often have 4 different layers with staccatos.

The case of a crescendo over 2 bars: Have a look that you get a crescendo over all but also that the rise of the velocity is not linear.

In other words: Even if you have a crescendo you should vary the rise of the velosity (as players do it in reality).

Further you could set accents on 1 and/or 3 or so.

Example: This piece only got one type of sample: Staccatos - to be precise; for each instrument!

It only is a variation of velocities.

BTW

If the dynamic doesn't react enough you can increase the dynamic within VI (try it out).

Best

Beat