Dithering?

[kapha]

hey y'all

Had my mastering engineer give me 16 bit masters but the fades were all wonky. I reimported the 16/44.1 masters into ableton and fixed the fades, but my question is do I set it to no dither now for the final at 16 bit? My sound engineering friend said this would be best, just want to double check.

[Tarekith]

If they were already dithered and you just adjusted the fades, you're fine leaving the dither off.

To be honest, you could you could likely dither it 10 more times and no one would ever notice it, so it's not something I would worry about too much either way.

[kapha]

That's what I figured, I also have a bitter meter on the master in the new project with the fades and everything is topped out very distinctly at 16 bit on the master channel.

It is a blank set, literally nothing on the track or master channel when rendering, even removed the bitter meter just in case.

Open to other opinions as well, but I am thinking the no dither option will be fine

[TLW]

Dithering is used when reducing the bit rate - such as a 24 bit recording being converted to 16 bit for CD use.

If the bit rate isn't being reduced all dithering does is add noise.

[kapha]

Right, but I was confused if the output of the DAW would need to be dithered to reduce any new artifacts that could be introduced from the bit depth downsampling, since Ableton runs with 32 bit floating point and wasn't sure what that meant for the output in terms of how I would have to treat the render in order to get the same audio, just with my automated fades in the 16 Bit audio clips themselves.

[TLW]

The effects of the internal floating point resolution of a DAW - 32 bit in the case of Live, up to 64 bit in some other DAWs - is something you can let the DAW worry about.

Where dithering comes in is when reducing the fixed bit depth of recorded or exported audio. Basically consider dithering when Live’s recording preferences are set to record at a bit depth higher than the audio format you need to export from Live.

When it comes to what kind of dither to apply, go for the one that sounds best to you - if you can tell them apart that is, because when anything other than a very simple track is dithered many/most people can’t easily tell one algorithm from another anyway.

[kapha]

So the audio that was imported was still 16 bit but my ableton is set to record at 24 Bit. The audio is not being recorded, so do I still have to adjust this parameter down from 24 bit to 16 bit if I am just bouncing with no effects at all other than the fades in the clips? I.e. should I switch it to 16, reimport the 16 bit masters, and then export with no dithering? Or am I okay with what I have already done importing 16 bit recordings into a 24 bit parameter set and then no dither bounce?

[TLW]

I suggest you try dithering the 24 bit processed version when exporting at 16 bit, export another 16 bit version without dithering and also try setting preferences to 16 bit and doing the whole thing in 16 bit throughout.

Then see which you think sounds best.

The effect of dithering can be very subtle, and is often not something to get too worried about. The only way to decide what works best on a particular track is to try several algorithms along with no dithering at all and listen to them. Preferably in a blind listening test. Or just pick an algorithm and stick to it - most people won’t be able to tell one dither algorithm from another anyway unless the track happens to really expose the process. There are few if any “rules” about this other than “if it sounds right, it is right”.

[oratowsky]

So the audio that was imported was still 16 bit but my ableton is set to record at 24 Bit. The audio is not being recorded, so do I still have to adjust this parameter down from 24 bit to 16 bit if I am just bouncing with no effects at all other than the fades in the clips? I.e. should I switch it to 16, reimport the 16 bit masters, and then export with no dithering? Or am I okay with what I have already done importing 16 bit recordings into a 24 bit parameter set and then no dither bounce?

Just to precise a point, I believe Ableton being set to record at 24 bit means when creating new recordings from an external source like a microphone it will be recorded in 24 bit.

If you import a 16 bit file into Ableton it the file stays 16 bit. it's not related to the recording settings in Live.

Then, any changes you make to that file are done internally at 32 bit.

As others have said dithering is such a subtle process any changes you make to the fades and then export at 16 or 24 bit with or without dithering really won't make or break anything.

But hang on a second, if the mastering engineer gave you something funky with your files would it be best to have them fix the issue on their end ?

[Tarekith]

I was just coming to suggest that, just tell the ME what you needed and let them deal with it. That’s what you’re paying them for, don’t be shy.