I have created separate scores for playback of each of my recent compositions (a piece for concert band and a piece for wind octet) because of the inconsistent behavior of Human Playback, Percussion sound volume, and so on.

Standard Human Playback is actually substandard, for the following reasons:
1. It does not play dynamic levels consistently; even when the ensemble is optimally balanced with the mixer, it is impossible to tell at any given instance how loud a given dynamic will be.
2. I have an instance in which a group including trumpets plays an accented phrase at ff, and is answered in echo by woodwinds. Even with the woodwinds at pppp, there is little difference between the volume of the two groups. (By the way: the orchestral trumpet sample sounds are godawful when they are accented at piu f or louder; the band trumpet section isn't much better.)
3. Exaggerated dynamics don't do much good: a pppp --> fff hairpin crescendo (for example) will be little, if any, different from a p --> mf crescendo.
4. Same with hairpin decrescendos.
5. However, unspecified cresc-decresc hairpin pairs produce a fairly appropriate swell effect when they start and end at the same dynamic level (changing by approximately one dynamic level); some cresc and decresc hairpins work (by one dynamic level) when no ending dynamic is specified, but not as much as wanted, which is why I need to specify dynamics.
6. It is not entirely clear whether HP cares what ending dynamics and the ends of the hairpins are attached to. For example, if I have a whole note in a measure and the next measure is a rest, the beginning dynamic is already in place, or I attach it to the whole note. Do I attach the end of the hairpin and the ending dynamic to the last beat increment the attachment jumps to, or to the barline, or to the first beat increment in the rest? I've tried all of these, and the results are consistently inconsistent and unpredictable. Would it help to put an unwanted 32nd note in the next measure to attach these to? As far as I can tell from other examples, it doesn't seem to help when crescendos cross barlines.

As I am using VST rather than MIDI, I have to be able to use the resources of HP to finesse playback. Am I crazy to expect Standard HP to behave as if it has standards? Must use Custom Playback to get HP to do what it should do as standard behavior?

Thanks again.


DF