You are here

Crafty Masters (SOS Nov 2017)

For everything after the recording stage: hardware/software and how you use it.

Moderator: Moderators

Crafty Masters (SOS Nov 2017)

Postby BarnabyBass » Mon May 10, 2021 10:41 am

Eddie Bazil's articles are great and very instructive. But did anyone else notice (answer will surely be yes!), that the integrated LUFS of the 3 demonstration audio tracks vary as follows:
Dry: -23.1
Matched -21.0
Finished -17.7
Unless you loudness match these you can't judge, although to be sure there are clear differences notwithstanding.
BarnabyBass
Poster
Posts: 14
Joined: Mon Sep 16, 2002 12:00 am
Location: Winchester, UK

Re: Crafty Masters (SOS Nov 2017)

Postby Wonks » Mon May 10, 2021 5:31 pm

You have a very valid point. It would be nice to have an extra set of sound samples that were closer in overall LUFS to better hear the tonal differences.

But the article was about mastering; taking a mixed track, EQing it broadly in line with a pink noise source and adding a few extra tweaks to add more interest. When mastering, you should be moving from a mix with plenty of headroom to a mastered mix with less headroom, so the sound samples provided clearly show that process.

If, for example, the EQ process ended up adding a fair bit of boost to the low and high ends, then with equal LUFS example tracks, the end result will seem more like the mids have been cut whilst the treble and bass are similar levels. So for understanding the process, the given sound samples are the better ones to have. On the other hand, if the before and after samples were LUFS matched (as if mastered to the same integrated LUFS value by two different people using different mastering processes) then it would help to get a better idea of the resulting sound differences.

So one extra example of the initial mix at the same LUFS level as the final track would have rounded things off nicely. But the tracks as supplied do illustrate the changes made during the process.

I would like to mention that not all pink noise sources are equal. I now use the Melda one after first using the standard Cubase 8.5 noise source module. The pink noise from that definitely did not give a good straight line slope. The Melda one had a few very small variations from ideal but was pretty close). So it’s well worth checking your chosen noise source first on a frequency analyser, before relying on it to EQ your tracks to. And I’d say the noise sample needs to be at least 20 seconds long (I use 30 second samples) before you’ll see a good straight line to your slope. It takes time for the random elements of the noise generation to produce a full frequency cumulative spectrum. Too short a period and the trace will be all peaks and dips.
User avatar
Wonks
Jedi Poster
Posts: 11607
Joined: Thu May 29, 2003 12:00 am
Location: Reading, UK
Correcting mistakes on the internet since 1853