How does my 192 AES/EBU input a 20-bit signal when my session is 24-bit??

[jacobfarron]

This sounds dumb, and probably is. Recently I've been wondering how I can record a 20-bit AES/EBU signal with the digital in's on my Pro Tools 192 I/O, when the session bit depth is 24-bits? There's no speak of it in the manual... nothing I can find online. All I can find is sample rate conversion, and I know that has nothing to do with it.

Also, does this mean I can send it a 16-bit signal and it'll record fine? An 8-bit signal (not that I'd want to). I just feel the need to know what's going on in there.

Again, I know it works. I'd just like to know how.

[craigmorris74]

You can record at different bit depth and the system will perform fine because the sample rates should be the same. From what you describe, since your project is a higher bit depth, the system is just adding extra zeroes to your files.

16 bit stuff will sound fine recorded as 24 bit (possibly better, as suggested by some who know much more than I do).

Craig

[jacobfarron]

It sounds simple... add extra zereos to the file, but I never leave "well enough" alone. I suppose I'm looking for a technical document describing the AES/EBU in and out process. It seems as long as I dither, I can output a 24bit AES/EBU signal into a lower bit input. But no adjustment is necessary when recording a lower bit depth into a higher bit depth - as long as my sample rates are the same.

Anyone have specifics?

[craigmorris74]

Check out Bob Katz's "Mastering Audio" book. Very good.

It's obvious why you might need adjustment (dither) when removing information from a source (going from 24 to 16 bit), since something might need to cover up the error (heard as noise) created by just throwing out that information.

Think about this: Why would you need to adjust something when no information is being lost (just meaningless information added when going from low bit rate to high bit rate)?

All the AES/EBU interface is doing is carrying a certain resolution signal (determined by bit depth) take at a certain time interval (sample rate). It's not especially different from SPDIF or other digital forms of transmission.

If that doesn't help, or if you don't trust me, then pick up Mr. Katz's book, it explains all this very eloquently.

[jacobfarron]

Haha, I will pick up the book. I trust all you guys, I'm just a techie like the rest of you. I like to know HOW and WHY... schematics my friends! Then we stuff goes wrong, and it does... I have a good excuse to take it apart.