@noldar12 said:

In the long run, it is best to create separate parts anyway, as that is how composers generally write for the orchestra.  Doing so allows each part to be independent of the others.  If your main interest is creating string pads, then independent parts are not so important.

What noldar12 writes should be law for anybody striving to create professional sounding strings arrangements with VSL. Independent parts are absolutely essential.

I am also in agreement with this.  If you're only interested in a quick hurry up and give a decent result mockup you can plunk down huge chords and run with it.  However, if you want the pristine real sound it is most certainly easier to achieve one part at a time.  I was working an accompaniment this weekend (and prior) and initially I had plunked down the parts at the same time (played violins 1 and 2 at the same time on the same track).  I then decided this weekend to play violin 1 in, then created a new track for violin 2.  Recorded that line in live while playing back the violin one part (I should mention I had done my piano solo prior to this so I had a performance track to follow).  The results were absolutely stunning compared to the plunking chord style!  Because I was able to have each track manipulated by their own velocity crossfades and expression changes.  I didn't duplicate the same patterns on both tracks so it gave them both their own individual sound.  Not to meantion, that it gave a real and natural quantize to the song because I naturally came in a little sooner, or later than the previous recorded line of Violins 1.

The chord style works nicely in certain areas of songs when it's nothing but sustained background filler.  However, if you stray from that and start having them harmonize one another, I think you might find it hard to beat doing individual performances.  Not to mention the amount of editing you will have to go back and do to create realistic attacks, cutoffs etc. which I was able to produce instantly on the fly for each part.  Being a bit redundant, on my prior performance song I recorded my 4 part chords seperately instead of block styled on one track and the results were so superior I have decided to no longer do it that way.  It takes a little more time up front, but the results it gives me in one pass is absolutely 5 million times better then the days I would have spent quantizing, editing velocities etc on every note to make them not sound synthy.  Besides, there is absolutely no way in real life every player of every section of the string orchestra can simultaneously crescendo at the same time (speed, attack/bow strength etc.)  You will have late bloomers, early risers, even if only to a degree.  Doing them seperate gives each section it's own life and distiguishable beauty.  If you had plunked 4 notes at once in violins and raised the xfade, it would be to perfect.  Which in essence would come across synthy and false to your ears (at least to mine).  But as ALWAYS, if it sounds great it is great!  Not everyone's going to agree with me, so do what works best for you.

**EDIT - I also should have mentioned, that this is specific to legato/sustained to me.  Because during staccato and other short performance articulations VI Pro handles everything perfectly for me (it randomizes the humanistic playback, you can change each patch to have a slightly different delay, panning etc).  When it comes to those fast styled performances I find it easier to chord style them (just as you see in the cool VI Pro videos) and it saves me TONS of time.

Maestro2be

@bhartmann said:

I am just now starting to use Vienna. In some sets (like Orchestral Strings | Violins) there are both Sustain and Legato articulations. The legato seems to be monophonic, while the Sustain for the same set is polyphonic. Do I have that right? If so, I will need to run five instances of Vienna in Sonar to create 1st violins, 2nd violins, violas, 'cellos, and basses. (Additional instances for each wind and percussion instrument.)

No not necessarily.  I'm a Sonar user as well and instead of loading 5 instances of Vienna Instruments I could load 1 instance of Vienna Ensemble, set my string arrangement up however I want it (panning, reverb, effects, etc) in Vienna Ensemble then bounce to five separate audio tracks; basically work with one track at a time.  Vienna Ensemble can load 12 instances of Vienna Instruments all on only 1 MIDI channel.  And since you mention Orchestral Strings, I do the same thing with my Solo strings to layer the string sound.  In total there are ten tracks for strings.  Sometimes 15 if I layer in the Chambers too.

The nice thing about using Vienna Ensemble is that you can save your settings as a template and load them later for another project saving A LOT of time.  There are some downsides such as unless you pan everything straight down the center in the Bus view if you change your mind about a setting or panning you'll have to reactivate the instance and rerecord it.  However, instead of having 5 synths in your synth rack for each instrument, you'll only have one for each instrument family or section.  Much more efficient and tidy.

BTW, The reason the legatos are monophonic are because there is a little sample connecting the two consecutive notes. 

 Actually, you can load 16 instances of Vienna Instruments into 1 instance of Vienna Ensemble.  My mistake.

@cgernaey said:

 Besides, there is absolutely no way in real life every player of every section of the string orchestra can simultaneously crescendo at the same time (speed, attack/bow strength etc.)  You will have late bloomers, early risers, even if only to a degree.  Doing them seperate gives each section it's own life and distiguishable beauty.  If you had plunked 4 notes at once in violins and raised the xfade, it would be to perfect.  Which in essence would come across synthy and false to your ears (at least to mine). 

Very important point. These are the exact reasons why "chording" a string orchestra will always sound bad, and especially bad if the piece has a very polyphonic texture. In such cases you have to work with separate parts to get even half-decent results.

For the ubiquitous chugga-chugga patterns (sorry, can't help my pejorative reflexes on this subject [:'(]) this is not always necessary, especially if they are far enough in the background and can be sufficently masked by the rest.

Hi!

I have quite similar issues

http://community.vsl.co.at/forums/t/28311.aspx

I think the SE demo project are quite good for getting an general idea, but especially for long and slow phrases there is something missing ...

Beats sfzlegato sounds good, but is also not suitable for longer phrases, my experiments by crossfading between legato and detache sounded somehow better, but still far away from perfect ...

Gabriel

Try using ONE instance of Vienna Ensemble instead of several instances of Vienna Instruments (see my previous post).  You say you want your projects efficient?  Vienna Ensemble was created with efficiency in mind.  With Vienna Instruments (VI) you're going to have seperate synth folders opening up for each instance of VI you create.  With Vienna Ensemble (VE), you only have one instance with only one synth folder.  More tidy and less clutter.  It might be a little faster to get your projects going too because even if you create a template using separate instances of VI, Vienna still has to authenticate each instance separately.  This takes time.  With VE there is only one authentication process. 

As far as the volume issue is concerned, bring your attack fader down but that's just a shot in the dark.  I think the faders default to 50. 

@bhartmann said:

Hey, I appreciate the great information, but I still feel like something is missing. It shouldn't be so complicated just to create a sustained note with a fast attack, or a sustained note with a medium attack. What am I missing? On a somewhat related topic, sometimes sustained notes will suddenly increase in volume after a few beats. This is odd and requires a lot of editing (that is, finding the exact instant to insert a series of volume controllers). Again, I feel that something basic is missing in the instructions. Can anyone help with these two points? Thanks.

I. Use the Attack Slider (you can f.e. automate it via MIDI)

II. This is not normal behaviour and is completely unknown to me. Sounds like you have to check both Volume and Expression lanes for the MIDI data you are sequencing - don't forget to erase any unnecessary volume changes.