Stereo mixing from the inside out
To paraphrase DnB supergroup Noisia’s approach to perfecting a track’s stereo width, outlined in a Reddit discussion from 2015 (see bit.ly/Noisia_AMA), a good approach is to tackle a mix’s stereo width “from the inside out”. By nailing the perfect mono mix first, you’ll get the most important elements working well in the centre of the stereo field. From this solid foundation, it’s far easier to mix ‘outwards’ and add your stereo information ‘on top’, safe in the knowledge that the track will collapse nicely to mono.
Some engineers like to write and mix an entire tune in mono first, then pan and widen elements after. Legendary 90s jungle producer Dillinja was known to mix through a single mono speaker for just this reason. That way, you know the mix is solid down the centre to begin with, and all your stereo stuff will be the icing on an already tasty cake.
Here’s where the staple mixing strategy of parallel (specifically send/return) processing comes into its own. Adding any sort of stereo info via a parallel return or duplicate channel is a great way to control overall width, as you keep the main channel totally solid in the mix (unlike an insert with wet/dry), and you can customise the width return by levelling and processing (levels, EQ, M/S, etc).
To illustrate, let’s say you want to widen a guitar part in an already solid mono mix. Slap a chorus plugin on the guitar’s channel, crank its wet/dry mix amount right up, and the guitar’s dry mono signal will be turned down in relation to this wet stereo signal, making that mono information a lot quieter in the mix. Now, when the entire mix is summed to mono, that guitar loses its central power.
To solve this, you can instead place the chorus effect (with 100% wet mix) on an auxiliary return, send the dry guitar’s signal to the aux, then blend this stereo information ‘over the top’ of the dry signal in parallel.
This is a win/win – you’ve retained mono power, and applied stereo information on top.