A knowledgeable mastering engineer can develop an appropriate chain of processing in seconds from experience and, importantly, an intimate understanding of their processing tools. The latter being a key element in achieving an excellent master. To understand how to approach effective tonal balancing there are a few concepts we must first take on board.

Every action has an effect – By this we mean in applying a change to one part of the tonal spectrum it will have the opposite effect in another. i.e. By applying positive gain to the high mids the low mids will sound less full. Equally when apply a dynamic process the same reactionary effect will be evident.

Every process has an artefact – i.e. Applying parallel (upwards) compression increases the bass content, downwards full (single) band compression will dull a mix. In considering the processing chain these should be used to our advantage to negate their effect and, in effect, achieve two positives from a singular process. In doing this we will also reduce the number of processes which improves our ability to create a less intrusive processing chain.

An equalisation (EQ) boost is not the same as a cut – The application of analogue EQ will have considerably more artefacts when boosting rather than when cutting. These artefacts we describe as phase distortion. It is the main contributing factor as to why we like the sound of 'good' analogue EQ and not 'poor' EQ, which blurs or fuzzes the sound. A good EQ introduces minimal phase distortion when cutting as it is in general a serial passive process, hence the initial statement. In practice applying a shelf boost and comparing to an equivalent shelf cut, plus overall gain boost to match at equal level this effect becomes clearly apparent. This must not be confused with digital linear or minimal phase EQ which, by its name, negates this effect. However, because this sounds cold or transparent it has more use in corrective rather than creative tonal processing.

Shelving EQ should be the first port of call - In the nature of shelving it will shift a large area of a mix's tone evenly. Introducing this tonal shift will be a minimal change to the mix balance, unlike the bell curve which will proportionally change gain differently over every frequency band effected but with the added disadvantage of a focus point to the phase shift. Both these effects can be reduced by increasing the 'Q' width or 'feathering' the EQ but the wider the 'Q' the closer to a shelf we get. Feathering is the method of dispersing the focus of the phase distortion apparent in an EQ boost. By halving the centre frequency and creating two 'feathers' on the EQ, one above and one below proportional in frequency, and boosting each of these by a quarter of the original boost. i.e.; 100hz boost of 2db we would create a feather set of 75Hz 0.5db, 100Hz 1db, 125Hz 0.5db. In this it can be clearly seen the same amount of gain has been applied but the focus of EQ has been dispersed over a range of frequencies softening its effective.

Applying dynamic processing can significantly change the mix's tone – Dynamic tools are inherently designed to effective the envelope of the given program material. Hence if we change the attach/release slope of the given material we effect its tone overall. Apply full band downwards compression dulls the tonal balance. Applying full band parallel compression increases the low weight of the session mix. Applying an aggressive high shelf boost before compression will enhance/excite that frequency area and so on. The latter can also be use to create effective stereo width enhancement. I'll discuss stereo enhancement in a future article.

Apply full band downwards compression dulls the tonal balance. Applying full band parallel compression increases the low weight of the session mix.
— JP

Taking all these points into account when listening to a session mix, if it requires no corrective processing (see previous article on 'Good frequency balance in mastering'), we should decide on what is out of balance in the tonal spectrum from the ideal 'flat' or 'full range' sound we are trying to achieve. This is likely to contain dynamic and frequency balance issues. Obviously from what we've stated above we have to take all these points into account when deciding on a processing order chain as every tool/process applied will have an effect on another and a differing aspect of the mix.

To achieve a 'good' master we must decide on a processing order that interrogates its interrelationship. Experimenting is not the way forward, just trying a process will not stand you in good stead – the more you can listen, list what you wish to change, decide on tools to effect and their process order, apply, evaluate using comparative techniques (see previous article on 'Comparative techniques in the mastering process') the better the decision you will make. Basically, we should minimise the amount of contact with the audio, keeping our ears fresh and our original prospective focused in mind.

To achieve a ‘good’ master we must decide on a processing order that interrogates its interrelationship. Experimenting is not the way forward, just trying a process will not stand you in good stead
— JP

There is no magic mastering button, process chain or especially 'all-in-one' software plug-in. Good tonal balance comes from experience, a good knowledge of tools/processes and a wide understanding of musicality. Always remember, a fantastic sounding master comes out of a great mix, a great mix out of an emotive performance and recording and none of this exists without the original creativity of the artist. Above all a mastering engineer needs the ability to be humble with an artist's hard work and their mix. We're mastering, not mixing, making the song sonically balanced and fit for purpose. i.e. sounds great in all playback environments.

Comment