Comment by adzm

note that floating point audio very often exceeds [-1.0, 1.0] within the pipeline, just to be tamed at the very end of the mix to fit within those bounds. this is pretty much why every modern DAW uses floating point these days.