Civ,
Perhaps I didn't communicate well, but I think you misunderstood me. "Real human performance requires imperfection" is not a debatable statement. It's about as plain blunt fact of music making as it gets. A human being is not a machine and a real performance will not sound exactly the same every time, a sample library can. I don't doubt you understand that, but that's the only thing I meant by that statement- to point out the obvious. I said it to make another point, which I think is where the lack of communication is. I don't want a bad performance at all, I prefer VSL for this. But if I want a less perfect performance at moments, in order to make a convincing mock-up- then obviously humanization has it's place. The point- I wasn't trying to suggest VSL record things imperfect, but that sequencing imperfection isn't as convincing as recorded imperfection (being real human performances)- that I want VSL to accomplish this better. You may feel that you can recreate humanization as well, I don't feel like I can (maybe it's my approach to it?)
Loose staccato and tight staccato are both recorded, why not 'perfect performance and not-so perfect performance' also? That's the only point I was trying to make by that. If you still disagree, I welcome the opinion. I just wanted to make my opinion clear (hopefully I didn't make it more confusing, lol). I definately don't think that VSL should stop recording the 'perfect performance', I just want more convincing humanization in using VSL. Maybe that's my approach, maybe VSL should record both, maybe more software features can help. I'm not claiming to know the exact problem, just that this is what I'm wanting from my library.
-Sean