dBFS Audio Clipping ???

[DigitalMusicProduction]

Hi

Is it right to say a peak level of -6dBFS is only recommended so as to leave enough space for effects processing such as mastering, which would take up an extra 3dB of space. Therefore seeing as I'm not applying any FX processing as a Master, i assume it would be safe to raise the peak level to -3dBFS ?

[DC-Choppah]

If you are talking about peak level BEFORE mastering, then I don't think it matters where you put them as long as they are safely away from the top.
So to keep it simple I just mix where the levels are safely in the green on the master fader with a few yellow peaks.

Best not even think about peak levels until you go to master the whole track.

When mastering a track, raising and lowering of the level is done often by hand, by ear. So you need to leave room in both directions to allow the music level to be adjusted. I don't think it is so much about effects. It is just about being able to shift the level up and down in different sections of the song. There is no reason to push the peaks in the mix up higher before mastering.

If you are talking about the level of the master, then this: https://www.soundonsound.com/techniques ... k-practice

-3dB true peaks (oversamples peaks) are recommended to make the music sound good if it will be converted to mp3.

[The Elf]

If this question is connected to your question about normalising... don't normalize to set a pre-mastering level. Just leave plenty of headroom and let the mastering engineer sort it out.

[RichardT]

Hi

Is it right to say a peak level of -6dBFS is only recommended so as to leave enough space for effects processing such as mastering, which would take up an extra 3dB of space. Therefore seeing as I'm not applying any FX processing as a Master, i assume it would be safe to raise the peak level to -3dBFS ?

If you're mastering it yourself, which is what I think you're saying, then aim for something more like a peak of -1 dBFS. If somebody else is mastering it, -3 or lower is fine.

[The Elf]

Normalising is a blunt tool - 'maximising' is arguably a more benign approach.

When 'normalising', a stray single-sample peak will set the level of your entire audio file. A 'maximising' limiter, set appropriately, can shave off these peaks transparently. Many limiter plug-ins will take inter-sample peaks into consideration too.

But again, do not do any of this to a mix file prior to mastering.

[DigitalMusicProduction]

To put this more in perspective what I've done is recorded a track and added some minor FX processing tools such as Imaging and Volocity, which technically would constitute as mastering. After this mastering stage I'm still able to peak at -6dB, therefore I'm assuming i could raise the volume level to -3dB or possibly -1dB, would this be safe without causing any issues, the project is for commercial release.

[The Elf]

Read my post again. There is no 'after' the mastering stage - it is the final stage of the process.

Removing the remaining headroom from the mix file is part of 'Mastering' (as is final dynamics processing, EQ, stereo imaging...). Typically you would not do this via normalisation, but if you feel that's the way for you then go ahead.

Hopefully you've considered dithering (or not) in your process, since you do not want to either apply dithering twice, or apply it prior to other processing.

[Ben Asaro]

To put this more in perspective what I've done is recorded a track and added some minor FX processing tools such as Imaging and Volocity, which technically would constitute as mastering. After this mastering stage I'm still able to peak at -6dB, therefore I'm assuming i could raise the volume level to -3dB or possibly -1dB, would this be safe without causing any issues, the project is for commercial release.

To being to understand this, take your finished mix and run it through the loudness penalty plugin. I think it's better to see in real time how much various streaming services will adjust the overall level of your mix. That's a great starting point when you're trying to come to grips with peak levels, LUFS, vs overall loudness of the mix in general.

Without knowing anything about the actual piece of music, it's virtually impossible for anyone to say 'look at these levels'; a lot depends on how much you have in what frequency bands, particularly the low frequency band.

And read what the Elf says very carefully, there's a lot more to his advice than you may be aware of.

[DigitalMusicProduction]

Read my post again. There is no 'after' the mastering stage - it is the final stage of the process.

You are indisputably correct, my words were wrongly stated, what i meant was, at the final stage of mastering should i be peaking as far as -1dB or less? What is recommended.

[The Elf]

What is the destination consumable for your audio?

For streaming services the 'typical' requirement is -1dBTP/-14LUFS, but services differ slightly, so if you have a specific target in mind it might be a good idea to check and match their requirements.

For CD it doesn't really matter so much, but for me that equates to -1dBFS peak. If you want to give rippers the best chance then -3dBFS may be advisable.

So what medium are you 'mastering' for?

[DigitalMusicProduction]

As mentioned the album is for commercial retail (online streaming) I've heard about the loudness war's therefore the standard LUFS volume has to be applied, not overly familiar with the concept at this point, so would need some direction as to how loud my final master should be?

[Drew Stephenson]

If you're aiming for streaming then -1dBFS and -14LUFS are a reasonable set of targets. But you do have be flexible on the latter depending on the material and how it's being presented.
I.e. if you're just uploading single songs at a time that's a reasonable starting point. But if you're creating an album or collection of songs then it makes sense to consider the context of the wider piece; a simple piano ballad probably shouldn't sound as loud as a full-on rock song. There are no rules for this bit though, just preference and experience.
There are plenty of good threads and articles on here about loudness normalisation, it's worth taking a bit of time to properly read up on it.

[RichardT]

For solo piano, you might find a lower LUFS value than -14 sounds better depending on the material, because pianos can have a large dynamic range, and they don’t sound good when compressed or limited - at least if you’re after a more classical sound.

Don’t go lower than -20 LUFS though. I tend to average out at around -16 LUFS.

[The Elf]

Sounds like it's time to get familiar with the metering your DAW offers and read up on metering generally. As Drew says, there are a number of articles and threads that cover all of this.

[DigitalMusicProduction]

What is the destination consumable for your audio?
What medium are you 'mastering' for?

As mentioned the project is a solo piano album, (soundtracks) to be be delivered to an online music distribution site, in my case CD Baby. The intention for the album would be for CD, digital downloads and streaming. Online research states streaming would need to be at -14 LUFS, whilst CDs at -8 LUFS.

Assuming this information is accurate? there would need to be two different final master volume outputs, one for CD, digital downloads, and another for streaming, is this correct?

[James Perrett]

Assuming this information is accurate? there would need to be two different final master volume outputs, one for CD, digital downloads, and another for streaming, is this correct?

Like much of the recording information on the Internet, this isn't necessarily accurate. It may apply in certain circumstances but probably not in your case.

Do you have a way to measure LUFS? If so, how loud is your music currently if you adjust the gain to give you a true peak of -1dBFS? I'm worried that you are chasing LUFS targets that are simply inappropriate for your music.

[Mike Stranks]

Post deleted... duplicated what someone else had already said...

[DigitalMusicProduction]

Do you have a way to measure LUFS? If so, how loud is your music currently

I'm worried that you are chasing LUFS targets that are simply inappropriate for your music.

Yes there is the LUFS meter in Logic, but some piano tracks only peak at -14 LUFS, but float much lower, I've tried raising the gain on the instrument channel strip volume to balance a float at -14 LUFS, but this causes clipping on the dBFS channel strip volume.

All the piano tracks on the album have been mastered to float between -18 and -10dBFS with peaks no higher then -3dBFS, however as mentioned when switching to the LUFS meter the
Tracks float much lower with some only peaking at -14 LUFS, so what to do in this case?

[Drew Stephenson]

There are different measures of LUFS, momentary, short-term and integrated.
The measure that streaming services are looking at is the integrated measure, this is based on the entire piece of music.
The short-term measure can be useful to look at how your loudest sections compare track to track.
I've yet to find any real use in the momentary measure, personally I ignore it.

[Wonks]

Can we assume that you are using the integrated LUFS value for the whole track, rather than the short term or instantaneous LUFS value?