[music-dsp] 32 bit floats >>> 24 bit integer

[Shay Green]

At 9:49 PM -0600 2006.01.20, Doug wrote:
> I just round to the nearest integer.  I assumed at some point that was at
> least better than simply truncating the amplitude, maybe it's not hmmm...

I thought this at one time, but I'm not so sure anymore, since truncating
is equivalent to subtracting 0.5 then rounding to nearest. In other words,
truncating introduces a miniscule DC offset.