How real do you make it?

[Mattias Westlund]

(07-11-2022, 06:13 PM)Nayrb Wrote: I've noticed that, with brass, two instrument sections have a distinct sound, and section sizes above that also have a distinct sound, but that sound doesn't change all that drastically after going beyond three or four players. I suspect, in a live setting, there is a good reason for adding more players, otherwise no one would have bothered to do it in the first place; but in the world of samples and microphones and studio recordings, is there any real reason to have more than, say four horn players, or three trombonists? Is it just sampling gimmickry? I get wanting to capture a certain sound, but if that sound can be achieved with fewer players, why would they even bother?

The cynical side of me would like to point out that we have nothing but the developer's word that a certain section actually is the size they claim it is. It's not like we have any way of counting the players. So a 12 horn section (or 24 or 66, or whatever) might only be a 6-piece section layered with itself and/or with additional processing. That could also explain why the sound hardly changes at all. And, of course, ridiculous section sizes is what sells libraries to the "bigger is always better" crowd

[Samulis]

(07-12-2022, 06:51 AM)Mattias Westlund Wrote:

(07-11-2022, 06:13 PM)Nayrb Wrote: I've noticed that, with brass, two instrument sections have a distinct sound, and section sizes above that also have a distinct sound, but that sound doesn't change all that drastically after going beyond three or four players. I suspect, in a live setting, there is a good reason for adding more players, otherwise no one would have bothered to do it in the first place; but in the world of samples and microphones and studio recordings, is there any real reason to have more than, say four horn players, or three trombonists? Is it just sampling gimmickry? I get wanting to capture a certain sound, but if that sound can be achieved with fewer players, why would they even bother?

The cynical side of me would like to point out that we have nothing but the developer's word that a certain section actually is the size they claim it is. It's not like we have any way of counting the players. So a 12 horn section (or 24 or 66, or whatever) might only be a 6-piece section layered with itself and/or with additional processing. That could also explain why the sound hardly changes at all. And, of course, ridiculous section sizes is what sells libraries to the "bigger is always better" crowd

Very true! It's trivial to layer two sets of recordings together; in fact, I did that very thing for some of the violin section articulations if I remember in VSCO 2 because two of the five players had to leave early, so I think trems are two separate sets of samples combined in post.

The only recent-ish thing I can think of is that absurdly large cello session Spitfire did for that Hans Zimmer thing years ago, where they did actually film the sampling session and it literally is like 40+ cellos. I don't think it was really that nice sounding though. XD

For some reason, mostly curiosity, my friend Joe managed to dig up 6 other trombone players for the trombone section session and I agreed to do a 7-player section, which is of course abnormal (normally 2 tenor + 1 bass, or 3 tenor). I remember Christopher Byrum Harris mentioning in his review that the larger brass sections get, the more of a "choral" sound they get, rounder and thicker with less definition, and I think that is a good analysis. Basically things blend so much that it becomes like a string section is compared to solo violin. There certainly is some form of diminishing returns, where each addition sounds less and less distinct, but there definitely is some difference between say, 3 trombones and 7 trombones or 6 trombones and 18 trombones, albeit a small one. There also is a very clear difference between 5 violins and 10 violins, or 8 violins and 18 violins, again, something we did in VSCO 2 where we provide a "large violin section" patch which combines the two RR together to make a 10-player section.