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



Forum Jump  
Humanizing Dimension Strings to make a bigger sound
Last post Tue, Jul 02 2019 by fatis12_24918, 6 replies.
Go to last post
Posted on Sun, Jun 30 2019 14:00
by Pixelpoet1985
Joined on Fri, Dec 23 2016, Germany, Posts 237

I spent quite a lot of time in achieving the sound I was looking for, in terms of delay and tuning humanization. Especially for Dimension Strings the settings are set extremely high.

But I stumbled upon one question:
The humanization presets are very good. I'm wondering why the values for the individual players go down to zero?

I think there is always a bit delay between the players to make the ensemble sound bigger. Maybe it's because of some natural delay in a room?! I tried a long time and figured out that 5–10ms of a minimum delay between each player is a good value.

DS Violin 1: min. value of 0ms (+ higher variations)
DS Violin 2: min. value of 10ms (+ higher variations)
DS Violin 3: min. value of 20 ms (+ higher variations)

What do you think? I want to hear some opinions. 

Posted on Mon, Jul 01 2019 03:00
by Macker
Joined on Tue, Aug 21 2018, London, Posts 139

Hello Pixelpoet,

it seems you've discovered for yourself the well known "chorus" effect which thickens the sound and which mix engineers have been using for many decades. Up to about 40 mS the human ear doesn't notice any echo (i.e. the Haas effect; well worth reading up in Wikipedia). It appears to me far too few people these days are bothering to do what you've done and experiment for themselves with basic principles. Well done.

As for why VSL's humanisation pitch contours generally all return to zero and quite soon, experienced musicians can tell you far more and far better than I can. I think it's generally understood in broad terms that during the so-called attack phase of a note, the player very quickly and intuitively "seeks" the required pitch, so to say, then (sort of) settles into the next phase, dynamic or otherwise, with a more or less established pitch. But of course that's an idealisation. Many, many times using Melodyne I've examined pitch contours within recorded samples from many sources and I've seen the good, the idiosyncratic, the weird, the bad and the ugly. I'd say VSL have conservatively adhered to good taste as well as practical realism.

Posted on Mon, Jul 01 2019 15:43
by Macker
Joined on Tue, Aug 21 2018, London, Posts 139

[Edited addition]

I'm guessing you're after longer duration of the thickening effect, which is stronger the more that the voices' pitches deviate from each other. The risk of ending up with an out-of-tune orchestra is there, but otherwise there's no reason why you shouldn't experiment with your own more extended pitch contours. Of course it's also possible to offset the fine tuning parameter on each instrument, though that can in some cases sound more artificial than letting humanisation introduce modulated pitch discrepancies between the instruments. The latter technique is especially effective, as you've discovered, when the humanisation contours happen at different times because you've put different delays (all within the Haas region) on the various voices.

I assume you've had to add your Haas region delays outside VIPro. [Edit - Oh wait, sorry, I've just remembered the other delay facilities in VIPro.] Maybe VSL will introduce a new facility in VIPro for easily setting up the Pixelpoet chorus effect!

Posted on Mon, Jul 01 2019 20:48
by Pixelpoet1985
Joined on Fri, Dec 23 2016, Germany, Posts 237

Hello Macker,

thanks for the reply.

I don't have a chorus or delay plugin that can be used in VE Pro, I tested it with Logic Sample Delay and it works quite good. But my template is in VE Pro, so I thought it could be done with the delay in the humanization presets. Is there any free plugin available? I guess so.

Primarily, I mean the delay part of the humanization presets which goes down to zero. I noticed that it sounds "better" or "bigger" when, instead of going down to zero, one of my minimum values are applied. Comparing this to a recorded (sampled) ensemble of the same size it comes closer.

So, it's not "wrong" to do it that way? I don't "destroy" anything? I mean, Dimension Strings was recorded as an ensemble, but it sounds smaller.

Posted on Tue, Jul 02 2019 02:46
by Macker
Joined on Tue, Aug 21 2018, London, Posts 139

Ah, thanks, now I have a somewhat better understanding of what you're doing.

1. First things first. Yes there is a free VST/AU delay plugin I can recommend: MUtility, by MeldaProduction. I use the paid version but believe the delay function works exactly the same way in the free version; and it lets you set delay values in samples or mS.

2. I wonder how Dimension instrument sample patches and presets actually relate to the acoustic realities of renditions given by the ensemble players during the Silent Stage recordings. I certainly don't doubt VSL when they tell us that the Dimension instrumentalists played in ensemble while being recorded individually. But also I wouldn't mind betting that VSL's library editors have subsequently tightened up time alignments across the ensemble's recorded samples, so that (a) it's then easier and more predictable for users to apply their own choices of complex post processing, such as humanisation, stretching, envelope manipulation, etc; and (b) the patches are capable of being played in DAWs with very precise temporal discipline if required. I see nothing wrong with such an 'editorial normalisation' approach. That said, I don't suppose VSL would ever describe their competitive tradecraft in detail.

3. I have only the (Sy-zed) Brass 1 library of the Dimension collection but I know what you mean about the 'smallness' of the unprocessed sound of the ensembles. I guess that's probably partly a consequence of editorial tightening - just as in reality, musically very "tight" live ensembles tend to sound somewhat compact acoustically.

4. I applaud your efforts to undo the temporal tightness and acoustic compactness of Dimension Strings. As far as I can see the only thing you would be 'breaking' is that 'editorial normalisation' which one can regard as a technically convenient starting point - ONLY a starting point. Good luck to you!

Posted on Tue, Jul 02 2019 07:29
by fatis12_24918
Joined on Sat, Dec 16 2006, Posts 270


Well, actually I often did it (something similar to your play with Humanization parameters) to increase organic feeling and human touch, and always got acceptable to excellent results. (In addition to mixing different articulations to get bigger sections, e.g. vibrato sus + progressive vib. or natural + sordino etc.)

I think the whole concept behind Dimension Strings is exactly to give you the power of finding your own effects according your music style and aesthetic taste, so the default very cold and compact sound has not to be considered the only one, nor the reference, according VSL designers themself!

It's also explained and recommended in the documentation and advertising by the way, so you are in the perfect attitude and stream! 

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.