In this post, I’m listing three feature requests, in order of importance or utility, at least to me, regarding dynamics playback. From the user’s perspective, these are related enough to warrant a single discussion, although I realize they might involve disparate sections of the program.
- Fix the dynamic loudness model
I describe this in my previous posts at Single Dynamic Point Level Inconsistency and Make Single Point Dynamics MIDI Value Table Available to User, & More. The first post elicited several responses, the second, so far, none.
This should include the ability for the user to assign either MIDI or dB values to each of the immediate dynamic symbols, and to enter user choices for these symbols, such as “m” or “poco f”. Composers who do not use, for example, mp or mf, should be able to either skip over these symbols when assigning MIDI or dB values to the symbols they do use, or at least assign values to the skipped symbols that do not change or otherwise interfere with the values assigned to the used symbols. The current “power curve” concept, with all immediate dynamic symbols listed and their relative MIDI value spacing hard-coded, precludes this.
- Fix the rendition of accents to reflect how they are actually played on orchestral instruments
Currently, accents are rendered as an increase in a volume CC’s MIDI value that persists for the duration of the note. This is realistic for plucked or struck instruments, but not for instruments that can vary loudness on a held note, such as bowed strings, wind, brass, and now (via Cantai) voice. Players of these instruments make only the attack or the initial transient temporarily louder, and then drop back to the marked dynamic.
- Add different crescendo and decrescendo curves, mathematicised as Bezier curves, to reflect, for example, “flared” hairpins, or to produce constant loudness cross fades. These might be best handled ad hoc in the Player’s Key Editor.