There was a recent thread on how to get better at playback. Here’s a thread to collect more advanced techniques, which may have already been touch upon or covered elsewhere.
I’ll start with tails on longs versus shorts. As Christian says @ Spitfire Audio
DON’T APPLY THE SAME AMOUNT OF REVERB TO ‘SHORT’ AND ‘LONGS’ SAMPLED ARTICULATIONS.
Because of the nature of virtuosic playing styles, all world class players will allow long notes to die off a little before becoming silent. This is called ‘diminuendo’* and is the reason short samples often sound wetter than long. It is imperative therefore that you track-lay and / or apply differing amounts of blending reverb to shorts (less) and longs (more).
As much as I agree with the playing style (and have always personally done it, it’s just human nature and not virtuosic), I hated putting the patches on different reverbs. One, it’s ugly and two, it felt like a filthy hack. But now with Dorico I think we have a proper solution - modify the note length in “Playback Options Override”, and add a volume/expression scale down on CC1 or CC11 (typically, library dependent)
Here’s a mock up using my first instrument (clarinet) and HALion
This sounds very natural, and makes the clarinet much more realistic and livens up HALion. In this case I didn’t set the playback option and just manually lengthened the tenuto note. With wind instruments you just let your breath die away which is vary natural, to give a long tail. The trick is to make sure you end the niente before the start of the next note or you’ll get a quiet hiccup. Obviously this should only apply to notes which have space before the next note, which also highlights my issue with a brute force reverb, which applies it to everything. Reverb is a separate topic that maybe we’ll cover in a different post.
Super advanced The final natural note cuts off (at least with HALion) too much I think for a real clarinet. That air column doesn’t naturally want to die off that much. There are a couple of solutions. One which is the naturally wet libraries such as Spitfire will give that to you (and trouble you on the shorts presumably). Or two, you can also lengthen and to a faster niente. I would only do this on a critical part of a major solo.
- n.b. actually that’s wrong AFAIK, it’s called “niente”, diminuendo is an explicit notation to decrease in loudness, not necessarily to silence. Also don’t they mean “that’s why LONG samples sound wetter than short”? At any rate, some libraries have niente samples, such as EWHO (East West Hollywood Orchestra). The advantage of a true niente sample is that tonal character of an instrument changes with volume, and assuming they sampled at the different volumes all the way to silence, with one of these you can capture that spectral timbre changes.
Daniel “Auto niente” would be a fantastic feature to add. I would think an override, on a per expression switch basis would do the trick.