0% Interest for 24 Months! Learn more »
(800) 222-4700
  • Español: (800) 222-4701
Microphone Month 3

What happens to 16-bit dithered samples when recorded into a 24-bit session?

Today’s Tech Question is pretty tweaky,, but valid. Those who are faint of heart may wish to take today off and return tomorrow. On the other hand, read on, it may interesting.

I‘ve been unable to get a suitable response from Emagic on this one so here it is:

The scenario: I use Logic Audio TDM, discrete output mixing, and have realized that mixing down back into Logic is the most direct method and quickest to get to the CD burning stage. I record at 24-bit into Logic via Apogee AD8000 converters. When I mix, I simply route all channels from my console into a pair of Apogee inputs and record the mix back into the same Logic session at 24-bit. I then solo that stereo pair, apply Waves L1 for dither and “bounce to disk” to create a stereo interleaved file to burn to CD-R.

The question: As I would prefer to use Apogee’s UV22 rather than the Waves L1, does a file dithered externally to 16-bit, then recorded into Logic set to 24-bit, get recorded as a full 24-bit file or a file with 16-bits of data and the remaining 8-bits of “nothing too important”? IOW, what is the nature of an audio file that has been dithered by a converter to 16-bit but recorded into a DAW at the DAW’s 24-bit setting? Do I still need to re-dither when bouncing to the interleaved file or simply bounce at 16-bit, no second dithering pass, knowing that the audio file has already been dithered and contains only 16 “significant” bits?

BTW-– all conversion passes are A/D and D/A. There are no digital transfers in this process.”

In this scenario the AD8000 is supplying 16-bit words to Logic that have been dithered from its 24-bit converters. Logic is then forced to interpret those 16-bit words as 24-bit words in which the eight least significant bits are all zero all the time. Thus you should be able to burn 16-bit discs from those files without any adverse effects. The software will just truncate those last eight bits for you. I would not recommend additional dither because you can cause a build up of noise when you dither your audio too many times.

That said,, this is all theory as it sits right now (I haven’t tried it). I would strongly recommend you experiment with this and some other scenarios that seem workable to you and compare the results. See which one “sounds” the best. Since this is your final mix the subjective sound quality is the most important thing, not how you got there.Bit), and Truncate.

Share this Article