The simple adjustment of the overall volume relative to another section of a song is a powerful tool in the finalisation of audio.

Not only can we change the feel of the macro dynamics of a given piece of music, but importantly achieve this without actually changing the original mix. Though perceptively we do hear a tonal balance change between sections, as we are no longer hearing the transition in the same way. The ‘volume’ difference makes us perceive the transition when positive as brighter / more powerful. Less gain would make it sound duller than the previous version, though it still may make the overall tone sound more balanced depending on what outcome is intended.

Over recent years I've noticed an increase in the frequency of when I need to use volume as a corrective tool to balance the macro dynamics of a track.

This is mainly in the area of a verse to chorus and return transitions. The chorus being overly dynamic in difference to the verse, often making the vocal sound too quite in the verses or lacking presence. In reality this is the volume difference we are hearing in most cases and not a tonal issue with the vocal. To correct with equalisation would be a poor course of action, as this will tend to ‘thin’ the audio. We'd be incorrectly boosting the high mids to achieve presence / brightness in the vocal, but this would be more apparent in the volume difference by the boost made. If we compared at equal loudness you'd perceive the EQ'd version of the verse as thin or weak relative to the original audio.

In the case of a few supplied tracks for an album recently I was required to correct the verse to chorus balance by raising the verses overall by 1.8db. A noticeable amount of volume by any scale. So why was this apparent to me during mastering and not the mix engineer?

I would surmise in both cases the need to use manual compression as a corrective technique lies in the bad habit of mixing with a limiter on the mix bus. I'm sure in the mind of the mixing engineer the idea proliferated that using a limiter will enable the dynamic to sound similar to that when mastered is a positive. The reality is the mastering engineer will have to apply manual compression to compensate for the mix engineer pushing the louder sections in to the limiter, hence their reduction in the transient detail with have perceptively made the track sound duller during mixing. Thus inherently the mix engineer compensates for the apparent 'dullness', or lack of attack, by adding more volume to these sections to achieve the tonal redress. Without realizing this is just an affect of the overall volume difference. I.e. from perceived loudness.

Fooled again, our ears are our own worst enemy and best friend. We must be ever vigilant in how and why we are manipulating audio. 

This issue is often missed by the mix engineer as when asked to submit the final mix without any master bus processing (see my article on session mix submission requirement for details) they are bouncing out and not listening to the outcome. Never submit anything to anyone without listening in detail. This is a fatal flaw in method and inevitably at some point something will go very wrong.

Remember, just because the computer said it did it doesn’t mean it did! Listen to it!

I should make note at this point the transitions between these ‘manual compression’ volume changes general always works better with an equal power crossfade of 25/35ms, though the transition should be adjusted to suit the change. Straight cuts in the audio can lead to audible clicks, often not notice till the final master. The overall dynamic range will have been reduced and details in the audio brought up. We also perceive the ‘shift’ in level greater in this regard.

To summaries, no matter what anyone proclaims about the virtues of mixing into a limiter on the master bus in reality they are being fooled by loudness. You'll achieve a better more cohesive mix every time by listening to the 'real' dynamic of the music.

Not only are they not hearing the transients correctly during the mix but in the worst case scenario submitting a mix for mastering with this unrequired processing. This will never ‘sound’ better than the mastering engineer having full control of the dynamic and reduction of transient detail.

None of the above is a negative reflection on any engineer or their practice. It's always a difficult aspect commenting on another's work. Equally these points are best raised after the production is complete as to try to re-address in the moment would be unproductive. 

I always think of every job as learning development, knowledge gained is understanding that can be applied in the next audio journey. 

Jp Braddock

Mastering Engineer @formaitonaudio