A propos du Dither . sous Reason 3

Citation : Dithering is applied to convert say a 24 bits audio file to a lower (16 bits) audio file. As reasons playback engine uses 32 bits floating point dithering was allways applied to the final wav as this is only logical. Now I dont really know much about the various dithering types, but I know its just plain white noise shaped by the output signal. When you choose another type, you change the white noise to pink noise (correct me if im wrong here)

The noise is used to bring the lowest level half bits, either under or above the half (dont know how to else explain it) to the closest bit, =to make it a bit.

When no dither is applied to the signal the bits that are too high (raapie, help me here, im struggling to explain) get truncated/thrown away.

Citation : There a lot of discussion about dithering and which types are the best. I personally prefer Mbit dithering of Ozone3.

You are also right about dithering from floating point into fixed bits. Some people are saying you should always dither, also for floating point. But I guess listening tests are the best.

Probably most people won't have a problem with 16 bit plus dithering exports from Reason now that they can add Mclass stuff on top of it. Dithering becomes much more of an issue when using classical music with long decays or when exporting digital signals to external analogue equipment.