Dithering when creating submix?

[SideBMusic]

I usually mix my backing track and bounce it as one submix or separate submixes for like electric guitar, bass, etc. It is my understanding that dithering can be used when bouncing from 24 to 16 bit. But should I turn off the dithering when I am doing these submixes to use in my vox-recording project file? I leave the submixes at 24 bit, BTW.

[triplets]

Only dither the final mix bounce, which is at 16-bit.

[Eric Cardenas]

I think you'll find this thread interesting:

Logic's Dither Questions

[SideBMusic]

Cool. Now I get it.

[MikeRobinson]

Always bear in mind that when you "dither," you are intelligently and purposely "throwing away digital data." Therefore, as a matter of principle, keep the original data and precision as long as you can. Try to avoid doing things which involved "oxen unequally yoked": data sources of differing bit-widths and therefore resolution. (Hence, not a "submix.")

 

Keep the files floating-point and 24-bit, etc., until the very last step, which is the manufacture of "distribution files" (of various types) from "the one true Master." (Because of your workflow up to that point, the only steps that involve dithering, compression, or what-have-you are non-destructive. "The" Master is untouched.)

 

"It's just better that way," no matter what sort of digital manipulation you are doing – sound, video, CG, photographs ... The price you pay is disk space, but who cares?

 

Maybe it will be "a difference that makes no difference," but the general problem is that, if this turns out not to be the case, you can find yourself rather scroo-o-od. It's difficult and time-consuming to go back and fix the problem, if you can do so at all. So, "stay up at altitude until you are ready to land." (Also be mindful throughout of the order in which you apply digital manipulations, since each one "loses a little bit of sand and picks up a little bit of dirt," some more than others.)