* fixed event handling bug which was introduced by the recent synthesis
optimizations (events were only processed for the first active voice)
* added global volume attenuation of -9 dB (0.35f) to prevent clipping
which can be overridden with --enable-global-attenuation
|