Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error

Forum Jump  
Time-Stretching: Progessive Vibrato
Last post Fri, Jun 28 2019 by Seventh Sam, 16 replies.
Options
Go to last post
Posted on Wed, Apr 10 2019 19:23
by Pixelpoet1985
Joined on Fri, Dec 23 2016, Germany, Posts 181

Hi,

is it possible to time-stretch the progressive vibrato legato patches in Dimension Strings or Appassionata Strings? It would be nice to have especially [EDIT] faster variations, so that the vibrato comes in earlier.

Posted on Fri, Apr 12 2019 11:27
by MMKA
Joined on Tue, May 22 2012, Posts 312
I recognize the need. The patches could be more useful in faster tempi.
Posted on Fri, Apr 12 2019 16:22
by Pixelpoet1985
Joined on Fri, Dec 23 2016, Germany, Posts 181

Not only in faster tempi, even the normal version is way too slow for my taste. Beyond time-stretching it would be great if the length could be controlled by a CC or that it adapts to your playing speed.

Posted on Fri, Apr 12 2019 17:55
by MMKA
Joined on Tue, May 22 2012, Posts 312

Originally Posted by: Pixelpoet1985 Go to Quoted Post

Beyond time-stretching it would be great if the length could be controlled by a CC or that it adapts to your playing speed.

Probably needless to say: there is of course the possibility of Xfade from a non vibrato patch to a vibrato patch, but I suppose you know this and you should prefer the cleaner way of adapt the length of the non vibrato part of a progressive vibrato patch.

Posted on Fri, Apr 12 2019 19:47
by littlewierdo
Joined on Sun, Apr 24 2016, Posts 178

Im using the Special Edition patches and I cant use any of the solo instruments for this reason. In fact, so many solo instrument libraries have exactly the same vibrato for every note. It drives me nuts. Every note sounds exactly the same. Vibrato is a powerful expressive tool that musicians use to express emotion. I use this video all the time to really demonstrate this, this is a video of a solo violin and a solo oboe backed by a full orchestra playing the theme to Schindlers List. Watch the violin player carefully, some notes, there is absolutely zero vibrato, some notes, vibrato starts at the beginning of the note, some notes, vibrato progressively increases or decreases over the duration of the note, and not all notes start with vibrato, and some notes start with vibrato and end straight. The point is, no solo string library on the market is allowing this kind of expressive control that Im aware of, I suspect because of the purity argument. That is, in order to do this, the mod wheel would have to control vibrato speed, and some other CC would have to control vibrato depth. Vibrato would have to be artifically created (that is, every note is recorded flat with no vibrato and all vibrato would be algorithmically generated).

Some libraries allow some flexibility in this regard, Chris Hein's library allows you to preprogram 4 notes with speed, depth, and start / stop times on the vibrato, but only for four variations.

Schindlers List Theme

Posted on Fri, Apr 12 2019 20:18
by Pixelpoet1985
Joined on Fri, Dec 23 2016, Germany, Posts 181

@MMKA
Of course I know the xfade option, but it's way more musical and expressive to use the progressive vibrato patch. For the same reasons there are also so many dynamic patches (pfp etc.). You can't imitate them with just crossfading. 

@littlewierdo
I know what you mean, but virtual instruments (if not modelled) have their limits, especially with vibrato. There are so many different vibrato styles (slow, fast, narrow, wide and all combinations of them) as you described. For me, a progressive vibrato patch comes closer to reality than any crossfading. I'm not a fan of virtually created vibrato, but I like the concept of the Friedlander Violin (Embertone).

Posted on Sat, Apr 13 2019 15:15
by civilization 3
Joined on Sat, May 16 2009, SF Bay Area, Posts 1558

Originally Posted by: littlewierdo Go to Quoted Post

Im using the Special Edition patches and I cant use any of the solo instruments for this reason. In fact, so many solo instrument libraries have exactly the same vibrato for every note. It drives me nuts. Every note sounds exactly the same. Vibrato is a powerful expressive tool that musicians use to express emotion. I use this video all the time to really demonstrate this, this is a video of a solo violin and a solo oboe backed by a full orchestra playing the theme to Schindlers List. Watch the violin player carefully, some notes, there is absolutely zero vibrato, some notes, vibrato starts at the beginning of the note, some notes, vibrato progressively increases or decreases over the duration of the note, and not all notes start with vibrato, and some notes start with vibrato and end straight. The point is, no solo string library on the market is allowing this kind of expressive control that Im aware of, I suspect because of the purity argument. That is, in order to do this, the mod wheel would have to control vibrato speed, and some other CC would have to control vibrato depth. Vibrato would have to be artifically created (that is, every note is recorded flat with no vibrato and all vibrato would be algorithmically generated).

Some libraries allow some flexibility in this regard, Chris Hein's library allows you to preprogram 4 notes with speed, depth, and start / stop times on the vibrato, but only for four variations.

Schindlers List Theme


Well, apart from SE which may be just that limited, we use the different vibratos in the libraries; mentioned here was crossfading slots to get some variation still holding one note. 

The fully articulated libraries have choices of vibrato, then dynamics patches with an evolving vibrato over those times, then in VI Pro there's time stretching. I don't have a Dimension Strings product, is it not in VI Pro with the time stretching? The first type is suited for strings. I wouldn't go past 15% +/- personally.

 I don't expect VSL to start doing that artificial kind of vibrato trick anytime soon, if ever. 

"Purity argument", whatever, these libraries are after realism and a generated vibrato glommed onto notes ia anything but that. 

MacBook Pro 15,1: 2.9 GHz 6-core i9
32GB 2400MHz DDR4
OSX 10.14.5
VE Pro 7
Posted on Wed, Apr 17 2019 18:43
by civilization 3
Joined on Sat, May 16 2009, SF Bay Area, Posts 1558

Anyway, if it's the original Dimension Strings in VI Pro, you have time stretching for this.

 

I'm all for as much vibrato variation in a product as possible. It tends to cost more.

MacBook Pro 15,1: 2.9 GHz 6-core i9
32GB 2400MHz DDR4
OSX 10.14.5
VE Pro 7
Posted on Fri, Apr 26 2019 13:17
by MMKA
Joined on Tue, May 22 2012, Posts 312

Originally Posted by: Pixelpoet1985 Go to Quoted Post

Hi,

is it possible to time-stretch the progressive vibrato legato patches in Dimension Strings or Appassionata Strings? It would be nice to have especially [EDIT] faster variations, so that the vibrato comes in earlier.

Two days ago I remembered in a bright moment this thread with a possible solution, unfortunately just for the SYNCHRON-ized Dimension Strings, so the one for the Synchron Player. I remembered namely, that in the Synchron Player a Start Offset of 1 second is possible (in VIPro only 50 ms). There are also several possibilities on which tones this Start Offset is applied: all, first, rep, leg, rep + leg.

I took a little violin melody from a piece I'm working on this moment, and recorded that with eight first violins (2 times, the second time starting on a lower note). After that I played a few seperate tones.
I did the maximum start offset of 1 second and used the setting "first", so only the first note of a melody has the start offset. I had to use the global attack slider (c22) with the "first" tones.
The Start Offset Attack can set to a maximum of 100 ms. Without using c22 I think in this case you need that maximum of 100 ms.

Example Start Offset 1 sec for "first" notes

Posted on Fri, Jun 21 2019 12:53
by Pixelpoet1985
Joined on Fri, Dec 23 2016, Germany, Posts 181

Originally Posted by: MMKA Go to Quoted Post
Two days ago I remembered in a bright moment this thread with a possible solution, unfortunately just for the SYNCHRON-ized Dimension Strings, so the one for the Synchron Player. I remembered namely, that in the Synchron Player a Start Offset of 1 second is possible (in VIPro only 50 ms). There are also several possibilities on which tones this Start Offset is applied: all, first, rep, leg, rep + leg.

I took a little violin melody from a piece I'm working on this moment, and recorded that with eight first violins (2 times, the second time starting on a lower note). After that I played a few seperate tones.
I did the maximum start offset of 1 second and used the setting "first", so only the first note of a melody has the start offset. I had to use the global attack slider (c22) with the "first" tones.
The Start Offset Attack can set to a maximum of 100 ms. Without using c22 I think in this case you need that maximum of 100 ms.

Example Start Offset 1 sec for "first" notes


I totally forgot the thread! :)

Your idea sounds good, I have to try it, but I only have the VI version. 

Posted on Fri, Jun 21 2019 15:16
by Seventh Sam
Joined on Sat, Dec 29 2018, Posts 99

Originally Posted by: Pixelpoet1985 Go to Quoted Post

Hi,

is it possible to time-stretch the progressive vibrato legato patches in Dimension Strings or Appassionata Strings? It would be nice to have especially [EDIT] faster variations, so that the vibrato comes in earlier.

Here you go:

Matrixes for the Violins, Violas and Cellos of the Appassionata Strings.  Each one contains three different time stretched variations of the progvib legato patch (plus the original), tailored so that they don't interfere with the legato samples or mess up the speed of the vibrato in any way.  The time stretching simply shortens the time between the straight tones and the onset of the vibrato.

The differences are subtle and amount to a time differential of 1-2 seconds depending on the patch.  Any faster than this is 1) not possible with the limits of the VI Pro time stretch function and 2) irrelevant since the other legato patches amount to the same sound anyhow.

Enjoy!

- Sam

File Attachment(s):
AppstringsTimeStretchedProgVib.zip (5kb) downloaded 4 time(s).

You cannot view/download attachments. Try to login or register.
Posted on Fri, Jun 21 2019 19:41
by Pixelpoet1985
Joined on Fri, Dec 23 2016, Germany, Posts 181

Sam, thank you very much! 

Unfortunately there isn't any documentation on the time-stretch feature. The VSL speed matrices are good, but don't cover all cases, especially the progressive patches.

Your patch souns quicker, indeed. But does the start offset would make it even more quicker? Or does it any harm to the patch?

But, thanks anyway. Very appreciated, I have to test it thoroughly.

Posted on Fri, Jun 21 2019 23:10
by Seventh Sam
Joined on Sat, Dec 29 2018, Posts 99

Originally Posted by: Pixelpoet1985 Go to Quoted Post

Sam, thank you very much! 

Unfortunately there isn't any documentation on the time-stretch feature. The VSL speed matrices are good, but don't cover all cases, especially the progressive patches.

Your patch souns quicker, indeed. But does the start offset would make it even more quicker? Or does it any harm to the patch?

But, thanks anyway. Very appreciated, I have to test it thoroughly.

The start offset won't make much of a difference as it's only 50ms, plus it'll cut off some of the valuable sonic information of the sample.  The matrix I made is about as quick as you can speed it up without sacrificing audio quality, afaik.  The only better option is to record another sample of the ensemble leaning into the vibrato quicker :)

Are you talking about the normal speed matrixes or the VI PRO speed matrixes with time stretching?

Also, have you thought about just X-fading between the progressive vibrato patch to the strong vibrato patch?  This seems to me the simplest and most controllable way to get at what you want: a faster fade into vibrato.  You could even time stretch the progressive patch so that it's slower, giving you roughly 8 or so seconds of non-vibrato time to have a crossfade of your liking in.  To my understanding, this is more or less how the Synchronized App. Strings do it - loop the non-vibrato part of the progressive vibrato and then allow you to subtly fade into either the legato patch (norm. vibrato intensity) or strong patch (molto vibrato).

EDIT: Sorry, didn't read the above posts thoroughly...

- Sam

Posted on Thu, Jun 27 2019 21:37
by Pixelpoet1985
Joined on Fri, Dec 23 2016, Germany, Posts 181

Hi Sam,

I had some time to test the patches.

Primarily, I wanted to use them with Dimension Strings, they don't work with them, because after the second re-bow the loop stops. Maybe a different curve is needed, I don't know.

I really, really love Dimension Strings, but the lack of an "instant" espressivo is annoying. Do someone from VSL know how to approach this, please? I hoped for an espessivo patch like in the Solo Strings or Chamber Strings. I do like the progressive vibrato, but it's too slow. 

Posted on Fri, Jun 28 2019 00:11
by civilization 3
Joined on Sat, May 16 2009, SF Bay Area, Posts 1558

one thing you can try is in the humanize window, FX, there are 5 'LFO' presets and you can manipulate those

MacBook Pro 15,1: 2.9 GHz 6-core i9
32GB 2400MHz DDR4
OSX 10.14.5
VE Pro 7
Posted on Fri, Jun 28 2019 00:54
by Seventh Sam
Joined on Sat, Dec 29 2018, Posts 99

Originally Posted by: Pixelpoet1985 Go to Quoted Post

Primarily, I wanted to use them with Dimension Strings, they don't work with them, because after the second re-bow the loop stops. Maybe a different curve is needed, I don't know.

I don't have Dimension Strings (yet), so I unfortunately can't mess around with those patches to try and find a solution.  Although, you could buy it for me and then I could help :)

All joking aside, it sounds like the curve isn't the issue, it's that the sample isn't looped.  A time stretch curve should be able to speed up the time between the attack and the onset of the recorded vibrato, which will of course make the unlooped sample end quicker.

I'd try taking the curve I drew for the appassionatas, take the last point in the curve, and bring it back along the x axis until it gives you a faster fade in but doesn't end the sample too fast.  I guess it's a compromise?

- Sam

You cannot post new threads in this forum.
You cannot reply to threads in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.