Mastering the loudness of a track

[Impostor]

At least you have the good feel to be right, thats all we need, right?
The arguments wont count so much here....so you dont want to reduce the volume, just on some peaks, and that doesnt count, right....

Eehhh...

I am the topic starter. Of course I'm right about what I wanted :)

[Senbee]

x42 limiter on ALL tracks
Very high gain, very high threshold, attack depends on the instrument.

All the rest: export wav 44100 16bit, load in Audacity, select all, normalise, loudness -12dB, apply, limiter -1, apply, export.
That's actually what everybody does to make things to sound loud on the radio. It's actually pretty bad, but it's the trend since the end of 80s.

[Death]

If you use compressor for mastering the loudness of a track...
you want to make it less loud - compressors are normally not used for making things louder but for making them less loud.

I don't really agree with this. You would normally make things have less dynamic range but not less loud. Of course, a compressor will usually make things quieter but when compensation gain is applied (which should be done to judge what difference it has made) you will actually make the signal louder.

Yes, this is the core point I've been trying to make but clearly I've been misunderstood.. The compressor will reduce the dynamic range and make the sound quieter, but once you apply makeup gain to bring the peaks back up to where they were pre-compression, the average level will now be higher and thus sound louder. For anyone who is confused by this subject, just look into 'psychoacoustics'. Someone in this thread mentioned Ian Shepherd. He has a podcast called 'The Mastering Show' where they always talk about loudness and how it can trick your ears. It's essential to understand this concept and always be aware of it so that you can make the best decisions when deciding how to, or how not to process a signal.

[Death]

awareness seems to be the key here...
first when playing with a compressor you can feel and hear that it has grip.
Sound gets better at a sweet spot and then gets quite "narrow" and loses somehow soundquality in the high freqs.
using compressor means for me losing soundquality, because as you said you will narrow the "head room".
this is problematic because digital has less dynamic range then analog.
thats why running your audio engine at double precision 64bit float aint a bad idea.
you can use compressor for sound design or you can use it to master the loudness of a track.
there you might experience digital clipping, not because the sound is too loud, then you can make it less loud....nooo....you can reduce the gain because the not so loud parts might dissapear then.
you just want to remove some digital clippings, happenig maybe 3 or 4 times at the loudest parts of the track.
then you can use the compressor, and in the end , yes you make it louder by reducing the volume of the peaks.
the peaks are gone and then you can make the single track of the song project louder.
now we are talking war of loudness.

I get what you're trying to say but I'm not advocating for making things as loud as possible. I hate what the loudness wars has done to music.. All I've been trying to do is explain a bit about how compression works and also why it's so important to loudness match your processed signal with your unprocessed signal to get a much clearer idea of the changes you've actually made when you A/B them.

I'm not really trying to give much more of an opinion than that. But I do think it's a shame when people only see a compressor as simply, an automatic volume control. That's essentially what it is but there are so many possibilities and cool things you can do with them and I just like to get other people thinking about these things because it's so cool once you realise what you can do with them.

By the way, does digital audio really have less dynamic range than analogue? I thought that one of main benefits of digital audio was having a much higher dynamic range and thus, a much lower noise floor which is why we don't have to have all that hiss anymore. I know that a compact cassette tape has the equivalent DR of about 9 bit. I'm not sure about professional tape though..

[erlkönig]

https://en.wikipedia.org/wiki/Dynamic_range

On a decently maintained Tascam 24 Track on 2" you reached -90dB (unweighted) to noise floor, adding some headroom (~ 10dB) you ended at ~-100dB. A fostex R8 on 1/4" tape reached -50dB unweighted, so with headroom you ended at -60 dB.

[Death]

https://en.wikipedia.org/wiki/Dynamic_range

On a decently maintained Tascam 24 Track on 2" you reached -90dB (unweighted) to noise floor, adding some headroom (~ 10dB) you ended at ~-100dB. A fostex R8 on 1/4" tape reached -50dB unweighted, so with headroom you ended at -60 dB.

That's interesting. There can be quite a bit of range then! I think digital still has more than any tape though, right?

you can see things from different perspectives:
for me, analog world has much more fine graining of volume and frequencies...also, you have much more headroom when working analog then working digital...
maybe i am wrong, maybe i am seeing things from a wrong angle...
Mastering means, you will prepare your audio to the distribution media...
reducing peaks and digital clipping by using a compressor,for me this is not mastering the loudness of a track.
then again, who is the master, who the servant?
take the long way home

I really like the analogue world myself, even though I don't own any analogue gear. I do use a lot of analogue emulating software though. I love a bit of analogue style soft clipping and I also add tape hiss into my recordings on purpose There is preference in all things but I think that what we've been talking about here is a lot more on the technical side of things and how things work.

As for mastering, I believe that mastering is simply an opinion made by a third party. A mastering engineer typically works with certain tools but the use of the tools is not what defines mastering. I believe you can master a track without doing anything to it. As you say, mastering is about making sure something is ready for its distribution media.

[erlkönig]

for me, analog world has much more fine graining of volume and frequencies...also, you have much more headroom when working analog then working digital.

The topic is a tecnical one. To get away from personal expirience and to be able to dive into a technical discussion i recommend reading (and, better, understanding) the folowing media:

• "Professional Ssound System Engineering", a book by Don and Carolin Davis (recommended by AES)
  In your profile i saw, that you're probably german
• "Handbuch der Tonstudiotechnik" by Michael Dickreiter
• http://www.sengpielaudio.com/
  The fine thing about physics is: you get away from personal views to reproducible, comparable results. And if you don't understand physics, you can apply the results anyway.

Mastering means, you will prepare your audio to the distribution media...

You're right, but: in audio technics we deal with 3 domains: time domain, frequency domain and dynamic domain. In my understanding "Mastering the loudness of a track" relates to the process of applying a mastering in the dynamic domain.

then again, who is the master, who the servant?

What kind of noise is this??

[erlkönig]

That's interesting. There can be quite a bit of range then! I think digital still has more than any tape though, right?

...yes! Much more at at least 24Bits, 64Bits are unnesecary.

As for mastering, I believe that mastering is simply an opinion made by a third party.

Not only. Mastering has its root in vinyl (or maybe earlier). The excursion of a needle is not linear to frequency, so the amount/level of low frequencies determined the length of audio on a side of record, therefore RIAA curve was applied to records. And to get that bit more of time, the mastering engineer had to tweak by hand. It evolved in the late 60s/early seventies to "make it sound better" (=make it louder mainly).

[erlkönig]

Sengpiel is one of the most compressed sources on audioproduction concerning the technical side in german language. To reduce it simply to the tests leaves out the main content of it, and: the sites' purpose. But the tests are good points to stop and check, if you got it. If you pass them, you're on a really good way in audio-engeneering (concerning the technical side).