The previous code is only correct if bytesPerFrame is a power of two. If it's not (e.g. 2-channel 24-bit = 6 bytes per frame), then the computed buffer position is corrupted. This can lead the code to use the wrong buffer half, resulting in glitchy audio.
The previous code is only correct if
bytesPerFrame
is a power of two. If it's not (e.g. 2-channel 24-bit = 6 bytes per frame), then the computed buffer position is corrupted. This can lead the code to use the wrong buffer half, resulting in glitchy audio.Fixes #763