.png)
Mastering is the final step of the music production process where an engineer will optimize the overall sound quality of a mix, ensuring that the music translates well across different playback systems and environments at a competitive level. Ideally this process enhances the mix. At the minimum, a master should come back sounding the same as the mix. A master should NEVER sound worse than the mix. However, because the mastering process almost always involves artifact-inducing dynamic range reduction with tools like compression, clipping, and limiting, this isn’t always the case. Therefore, when evaluating a master, it is vital to compare it to the pre-mastered mix.
But there’s a catch. As listeners, we tend to like music more the louder it is. Since the mastering process usually involves making the music louder, a direct comparison between the master and mix is not particularly helpful as a master that plays back louder than a mix will almost always sound better.
So how do we make accurate comparisons between a mix and a master? We need to make sure each file is loudness matched. There are two possible ways to do this. First, we could turn up the mix to meet the perceived loudness of the master. But because mixes are generally more dynamic than masters (meaning the level difference between the loud and quiet parts of the signal is greater), turning up a mix often results in clipping as the audio’s peaks will exceed the point of distortion. Mastering engineers are able to turn up the level on their masters with little or no clipping because they use dynamic range reduction tools like the ones mentioned above to reduce peaks while increasing the average level overall.*
If turning up the mix to meet the master’s loudness isn’t a good option, that means we need to turn down the master to match the mix’s loudness. There are two ways to do this. The first involves using your DAW:
Create a new session in your DAW and import both the mix and master onto new tracks.
Determine the loudness of both your mix and master. To do this, insert a loudness meter on each track that measures integrated LUFS (like this free one).** Start by looking at the meter on your mix track and bounce the audio. Once the audio is finished exporting, the meter should display an integrated LUFS value for the mix. Write this number down. Now, repeat the process with the master.
Turn down the master until its integrated LUFS value matches that of the mix. You may need to do this with the clip gain or a trim plugin before the loudness meter as the inserted loudness meter will likely measure the pre-fader level. dBFS are generally a decent proxy for LUFS in this application, so you can start by calculating the difference in LUFS between the master and mix (subtract the mix LUFS value from the master) and then turning down the master by that amount in dB. Finally, double check the integrated LUFS of the mix and master as described in step 2 to make sure they match.
With this level matching technique, you can now toggle between the mix and master and directly compare the two without being fooled by differences in loudness, instead making assessments on the basis of tone and feel.
The second level matching technique involves the use of a free online tool called “Loudness Penalty.” Almost all streaming services normalize their audio meaning all music plays back at the same integrated LUFS value.*** In Spotify’s case, that number is -14 LUFS. So, if Spotify receives a master at -9 LUFS (a very typical loudness), the platform will turn it down by 5 LUFS. Determining the amount of reduction the various streaming services will apply is what Loudness Penalty is all about. A quick note – the word “penalty” is a bit of a misnomer as it evokes punitive connotations. In reality, almost no commercially released popular music these days is mastered as low as -14 LUFS. Don’t be spooked if you process a master with Loudness Penalty and see a big reduction figure.**** Unless the music sounds slammed (over-compressed/limited), you likely have nothing to worry about.
Loudness Penalty is a great tool for level matching mixes and masters because it allows you to play back each file at the same integrated loudness. Simply open two Loudness Penalty tabs, process each file, and compare. Loudness Penalty will turn BOTH files down to match whichever streaming service you select.
Another convenience of using Loudness Penalty is that you can compare your mastered music directly to music on the streaming services to hear if it competes in terms of loudness. What you hear on Loudness Penalty is almost exactly what you’ll hear on streaming. Of course you can do this in your DAW too as shown above, but it’s more complicated than Loudness Penalty’s click of a button.
One thing to be aware of is that certain streaming services will not turn up your mix if it arrives below -14 LUFS. Therefore, if you process a mix with Loudness Penalty that is below -14 LUFS, you’ll be unable to audition certain streaming services. You’ll also run into the issue of clipping/limiting for the services that do increase the loudness on songs below -14 LUFS. All told, I recommend using the first loudness matching method if your mix’s loudness is below -14 LUFS.
That’s it! Now a couple closing thoughts. First, remember that integrated LUFS is a measure of the average loudness of an entire song. Also remember that the mastering process usually reduces a song’s dynamic range. Take a simple song made up of one verse and one chorus. If the mix’s verse plays back at -14 LUFS and the chorus plays at -10 LUFS, the integrated LUFS value for the entire song will be -12 LUFS (the average of the two). Now let’s say in mastering, the macro dynamic variation between sections is consolidated, so that when the mix and master are loudness matched on the basis of integrated LUFS, the master’s verse now plays at -13 LUFS and the chorus is -11 LUFS. The integrated LUFS value is -12 again. The mix and master will display the exact same integrated LUFS readout when measured, but the master’s verse will be louder than the mix’s and its chorus will be quieter. Put simply, if you’re loudness matching a mix and a master and you notice loudness differences between sections, you may have to chop up the song and use the DAW loudness matching technique from section to section.
Finally, it’s a little unfair to compare a very dynamic mix against a much less dynamic master. Part of the mastering process is about bringing a song up to competitive levels. As I wrote earlier, mastering engineers achieve this by using dynamic range reduction tools. But those tools do impart artifacts as they reduce a song’s peaks. If your mix is sitting significantly lower than the master in terms of integrated LUFS, but the peaks are still hitting 0 dBFS, you may want to consider “meeting in the middle.” That is, apply a bit of limiting to your mix and loudness match after that so the master has a fair fight. In a pop context I might suggest limiting the mix if its peaks are hitting 0 dBFS but its integrated loudness is -18 LUFS or lower, but that’s just a ballpark number.
*Clipping is a tool that is often used in the mastering process, but it's done in a controlled and precise way, almost always before the final output limiter.
**LUFS is a measurement used to quantify a piece of music’s loudness. The higher the negative value, the quieter something is. For example, a piece of music playing back at -17 LUFS will sound quieter than a piece of music playing back at -8 LUFS. Integrated LUFS refers to the average LUFS value of an entire piece of music or playback span.
***Unless the master’s integrated LUFS is below the normalization target in which case most streaming services will leave it as-is. Most streaming services will also preserve the relative loudness between songs on the same album.
****Loudness Penalty makes it very clear that when you “upload” your music for evaluation, you’re not actually uploading it so there’s virtually no risk of your music ending up on some mystery internet server (at least according to them).
