Bounce settings before mastering

[skycraft]

I'm mixing a project where the source files were all recorded in 16bit/44.1kHz (*sigh*, I know, not my fault, what's done is done). Most mastering houses prefer to get files with at least a 24bit resolution. That being said, Is there anything I should do to improve the quality of the bounce when it is already at 16bit?

Converting to 24-bit, as far as I understand, doesn't make a lot of sense. That would be like converting a 72dpi image to 300dpi. It wouldn't improve the quality, just make the file size bigger. And, because were dealing with audio, I assume that would introduce some weird artifacts.

I don't use heavy compression on the master bus (maybe just a smidge to catch the peaks).

What about dither?

Or should I just leave it?

Thanks

[ajv205]

Your stuck. Keep it at 16/44.1 from here on out to mix delivery. One could argue about upconverting to 24 bit for the mixing process, but that would only make sense if you are going out to an analog console and printing back into PT. No sense upconverting in your case (asuming you are in the box all the way). Leave the dither to the mastering engineer.

[Sven62]

It's already at 16/44.1 so what would be the point in dithering? Isn't dithering the process of changing a file from one bit rate to another? i.e. dithering down to 16 bit for making CDs.

[Murt]

FWIW Digi still recommend using a dither plug in mixing down from a 16 bit session because the session is being processed at higher internal bit depths.

[daeron80]

Well, consider if you had a 72 dpi pic that you needed to do a bunch of processing on. Then it would make sense to convert it to 300 dpi so the processing has more room to do its thing properly, right? Same w audio. If all you're going to do with these tracks is pass them as is to the mastering house, then, no, it won't help to up sample them. But if you EQ, compress, or mix them together, or do anything to them at all, then PT LE is going to do that in a 32-bit floating point engine, and will output the result of the calculations at 24-bit fixed. If you play back a 16-bit file in a 16-bit session and do nothing more than change the gain .01 dB, you have thereby converted the bit stream to a higher depth.

BTW, importing a 16-bit file into a 24-bit session has no effect on the audio - it just tags 8 zeros to the end of each byte.

If you might apply any plug-ins as Audio Suite or if you might print any submixes for use in the mix, then absolutely definitely without any question a 24-bit session will make all the difference in the world - it will provide 256 times greater resolution for those tracks.

If not, then a 16-bit session will probably sound the same, but definitely bounce to 24-bit with no dither for the mastering house. That will preserve the resolution of the mix itself, which is not 16-bit even in a 16-bit session. The bit stream is 32-bit floating, and the output is 24-bit fixed in LE regardless of the bit depth of the "session" (meaning the files being played back in tracks in the session).

For output to 16-bit files for CD or mp3, use 16-bit dither as the last plug-in on the master fader. Just make sure the dither is disabled for output to 24-bit for the mastering house to use.