User Manual

Crumar Seven User s Manual '
5.1. VOICE ALLOCATION: MODELING.
This is common to all physical modeling engines, meaning that each single note is like a complete synthesizer on its own and plays
independently from all other voices. Thus, each note is equivalent to a voice - exactly like in the real instrument - leaving no need for
extra voices. The only correlation between notes is the generation of sympathetic resonances, but this is a different topic that will be
discussed later in this manual. Let's take the “Tine” piano for an example: when the sustain pedal is kept depressed and the same note is
played over and over (ribattuto), only one voice is allocated, and all the elements that make a voice (hammer, damper, tine, tonebar,
pickup) are reused each time the same note is played. In this case we can call it “Fully Polyphonic”, means that there's no limit to the
polyphony as the instrument is capable of playing all of its possible notes simultaneously.
5.2. VOICE ALLOCATION: R.R. + I.V.S.
R.R. stands for Round Robin and I.V.S. stands for Intelligent Voice Stealing. This is the allocation model explained above, common to
most synthesizers and sample playback engines. In our instrument, the voice stealing algorithm always steals the quieter voice.
5.3. VOICE ALLOCATION: MODELING (HYBRID)
This is only used for the Electric Baby Grand and the Acoustic Grand Piano engines and is a mix between two different voice allocation
models. The stringboard is split in different parts because some of the strings are free to resonate even when they're not struck by the
hammers, other strings only vibrate when they are required to do so. This method offers the possibility to obtain the behavior of a true
polyphonic model just like in the real instrument.
Crumar Seven User's Manual - Page 12