I can tell you that it doesn't work with real players, so samples will probably be the same. The problem is that it's not just that one player will affect the way another plays, but it also affects their instrument. There is something abut the reaction between instruments that can't (currently) be replicating by stacking solo players.

DG

What DG is saying is true, and there's even more to it than that!  Solo strings are a style of their own.  The player will intentionally play with a stronger tone and will even use different rosin or sometimes an entirely different instrument to get a more strident sound than in an ensemble and the difference in sound is quite significant, so right there you can see why an ensemble is not a collection of soloists.  For example, none of the strings in Dimension Strings sounds like a guest soloist.  When you solo one player at a time they each sound like an ensemble member.  Additionally every player in an ensemble sounds different, from the sound of their instrument, the sound of their playing style, the sound of their individual vibrato, the way they transition between notes, etc.  Sure you can control some of these things but not dramatically enough to actually sound like eight different players and eight different instruments from one single player on one single instrument, which is what you have in Solo Strings.  Considering the front-and-center tone of VSL's solo strings (and actually most soloists, for the reasons above) you wouldn't get a very pretty-sounding ensemble even if you succeeded in making them sound like different players.  It would sound like everyone's trying to get into the spotlight because that's exactly the intention.

 Thank you guys! Now I better understand why this just won't work. Glad we got Dimension Strings [:)]

I recently recorded a violinist and cellist for a trio piece (with piano) I was working on. I don't work very often with real players (they don't come cheap), so I started by recording them separately, one at a time, like you would do with MIDI parts, because that's what I'm used to.

When the recording was done, the players – who know each other well and often play in large ensembles together – told me they could do much better if I recorded them together. Especially for my piece in which the piano was recorded with no click and with lots of rubato.

I'll just say that I was thrilled by listening to them play the two parts at the same time. Dynamics, intention, rythm, it was really something else. They really do play differently, and much better, when they have another part playing simultaneously next to them.

I ended up using the separate tracks because they were good enough and to be able to do some stereo separation and EQ (I'm talking really low-budget here...) but to me it proves DG's point.

Just my 2c.

I hope the OP does try this, because it's instructive. It won't work for the reasons described in this thread. But when you hear the conflated mush of 12 sampled solos playing at once, you gain insight into why VSL does what it does and how it does it. 

I suspect every veteran here has tried this once in his MIDI life. It's a sampling rite of passage. I thought once we could randomize pitch, attacks, and set solo samples in their own MIR seating, I could pull it off. Nope. Still didn't work. 

Playing the same sample 12 times would sound like one violinist playing 12x loud, not like 12 violinists playing in unison.

I was under the impression that they weren't planning to play the same sample but use different articulations and the pitch-shifting technique.

@Casiquire said:

I was under the impression that they weren't planning to play the same sample but use different articulations and the pitch-shifting technique.

Using different articulations seems weird to me, but ok.

Using the same articulation but pitch-shifting 12 notes would result in an octave span between the extreme pitches being shifted. Actually 2 octaves, for libraries where the sample changes every whole tone instead of semitone. That's gonna move formants all over, but ok.

That's why I'm not trying this any time soon. The thing about the players being recorded at different times doesn't totally deter me, despite it being consensus here. I mean, I use brass, winds, strings that were all recorded at different times -- we all do -- and it's not always a disaster.

Besides the approach of people in an ensemble, the mentality of it and all of this, the psychoacoustics and the expectation is not met by piling on single instruments.

"I thought once we could randomize pitch, attacks, and set solo samples in their own MIR seating, I could pull it off. Nope." is the truth. Its appearance is fake because it's fake, essentially.

Thanks guys. Very insightful. It's very clear to me now that solos won't work in this regard. I was recently in a live orchestral event, and I was observing string each player, and they do play differently compared to a solo performance.