the mixing and mastering process

since the mixing process can be confusing to some like it’s black magic i’m going to explain to you my kind of system of how i go about my mixes. of course, the golden rule is always “use your ear” because the whole purpose of mixing is getting something to sound as good as possible, but that may not be helpful to those who are trying to find a method to do that. this is my approach to it in order.

(fig 1. me playing bullshit at interscope records)

• organization: i gather the stems from the client and organize them into typically 5 to 7 groups. for the record i do use ableton live, and am without a hard mixing console - i’m not sure how other daws handle groups but at the least i will have these track types local to each other if not in groups:

  • drums

  • atmospheres/ambience (sweeps, impacts and pads)

  • fx (if the project is small i will typically have atmospheric elements in the fx group, but typically this includes miscellaneous sounds, layers, or enhancers)

  • melodics

  • basslines

  • vocals

• individual processing: once i have them grouped i will typically work my way down the tracklist and process each of them individually. this helps me get an idea of what each sound is doing in the mix. i will also very quickly solo each track to see if there are any hidden or unnecessary frequencies i can cut off. it’ll also help me find anything that is hidden or probably forgotten by the artist, and will raise those volumes if it sounds good.

  • after doing this, i will typically start working on the overall sound of the mix starting with the loudest section of the song, and the loudest sounds in the mix. this will help me understand how loud the quieter parts should be (i.e. breakdown, intro, build up, etc) because before i work on those i have the loudest section already mixed to use as a reference.

tip - it’s important to note that during my process, i do not mix into anything. in other words, there’s no processing in any sends, busses, or especially the master. during this stage it’s vital to understand what each particular sound is doing without hearing it through any sort of effects (particularly limiting, compression or EQ) as this will affect your understanding of its dynamics, transients and frequency responses.

another tip - during this stage i’m typically mixing at a very low level, 50-60dB at listening position (which is the loudness of a quiet conversation, or rainfall). this helps a bunch when placing things in the dynamic space and avoiding ear fatigue. i spend most of my time at this level. in later stages or when mixing sub bass and catching problematic frequencies, i’ll start mixing around 70-80dB, or jump onto headphones.

another another tip - headroom is necessary while mixing. you should not be clipping in the master while mixing down, the same reason why you shouldn’t mix into any master effects.

• sound placement: after understanding the bits and pieces of the mix, or while i’m doing step 2, i will try and find a place for each sound (i.e. basslines belong in the low end, fx in the stereo field, melodics in the middle range, etc). each sound should have its own residence. most of the melodics will probably have its lows cut to make room for the bassline.

  • very important - find a place for each sound in the stereo field as well. select things that will reside in mono, then some in stereo, then left or right. if everything is wide, the imaging will have no depth and thus will lose dimensionality - same as if everything were in mono. 

  • depending on the sound, sometimes i like to get small sounds in complete mono and pan them slightly left or right (like, 10-20% left or right). most of the time these are small little details (probably minor one-shots or FX). this is a great way to get small things to pop out of a busy mix.

tip - don’t overdo EQ. when boosting certain frequencies such as the sub, it shouldn’t diminish any of its surrounding frequencies. they should be complemented. while bass boosting is a preferred setting of eq, there is a line to be drawn where the sub becomes too loud, and i’d say that’s when the sub becomes the only thing you hear as opposed to its upper harmonics working together.

• dynamics: in the same light as sound placement, we also need each sound to reside in its own loudness. if everything were loud, it would be difficult to understand what sound is supposed to be getting the spotlight (unless the mix is intended to be a wall of sound). in electronic dance music, you typically want the kicks and snares to be the forefront, then the basslines, then the leads, and so on. anything you don’t want to distract from those main sounds you wouldn’t want them to be too loud or forefront.

• mix in mono: this helps you understand how mixes translate on cell phones and other systems that don’t offer a perfect imaging scenario as well as headphones. hearing the mix in mono will also help you hone in on sounds frequency and dynamic wise. gain, eq and compression decisions in mono feel a lot more apparent, at least in my opinion.

• a/b monitoring: if you are on monitors, switch to headphones or vice versa. sound represented by the two can be very different from each other. there are a lot of things that i can’t catch on monitors that i can on headphones, and vice versa. of course, the more advanced your system is, the less room there may be for differences - but i don’t believe the two will ever be the same regardless of how sterile the environment is, so it’s always good to use two monitors.

  • using monitors are an excellent way to understand dynamic space and body of sounds. with a good room, most of the mix can be done on monitors, and will subtract the time i spend going back and forth between headphones and other systems.

  • using headphones will isolate you from any room artifacts nearfield monitors may produce (i.e. standing waves, reverberation or early reflections). you may feel that harmonics representation may be more accurate than on speakers, however the dynamic space might feel a bit more congested (transients might not be accurate or may be represented differently). in many cases, i often feel that my mixes feel empty on speakers after mixing on headphones for most of the mix.

  • optional c monitoring: listen through a cell phone, car speaker, or a horrortone single woofer monitor. i would typically do this at the very last step right before mastering. this will get you a better idea of how a mix is represented on consumer systems. while i would probably make only 5% of mix decisions, if any, on c monitoring, it’s useful to know what your mix is doing on something you’re familiar with.

• reference music: before you feel totally confident in sending off your mix as ‘final’, you may want to compare your mix to your favorite music. this is a massive refresher since many factors can affect your mixing decisions, and therefore affect the overall tone of the mix. room, reference monitors and ear fatigue are a few of many factors. often times when i produce without thinking about the mix, it ends up lacking highs and punch so it feels overall dull. i suspect it’s because it’s easier on the ears after listening to it a billion times in the production process, or it could be that my room or listening position warrants a lot of high end. in any case, whether psychological or audiophysical, using reference tracks are a great way to notice where your mix is lacking, or having in excess. while using these tracks as sort of a ‘bar’, i’ll have a checklist of comparisons that also serve as criteria on whether the mix feels finished to me:

  • spectrum balance (is there enough high end? low end? mids?)

  • clarity (can you hear each instrument? does anything feel like they’re cluttering each other?)

  • wideness (do you feel immersed in the mix while at the same time feeling like you can grab a certain sound in 3 dimensional space?)

  • dynamics (does the mix breathe? are the drums punchy? do they sit in front of the mix and aren’t buried by other elements?)

  • lastly, loudness (do you think this would be mixable in a set, such that it wouldn’t be jarringly loud or quiet compared to the next or previous song?)

    • it’s important to note that while LUFs (a pseudo-quantifiable measurement of loudness of how it’s perceived by a human rather than the actual amplitude of a signal such as true peak) is useful in comparing mixes to references, a common mistake is to try and apply that same loudness to all tracks - certain compositions can warrant louder mixes, others not so much. for example, while dua lipa’s music hits -5 or -4, it’s common for darker tracks such as ones by cirez d to sit at -6 or -7 because dua’s loudness comes from her vocals and energetic instruments, while cirez d’s mixes require that space for transients and more apparent low end.

  • during this phase it’s common for me to go back and forth between making adjustments and listening to reference tracks. this could take just as much time as the previous steps altogether. it’s very easy to get trapped in a hole of “is it good enough?” during this stage. often times once you’re making half-decibel changes you should be just about finished.

• mastering: as for the very very last step before sending it off to distribution, if everything was done right in the mixing process, the only thing you might have to do for mastering is to raise the gain for loudness. most modern music sits at -7 LUFs and above, and typically at this loudness the signal will have to clip (or hit some sort of limiter, if you don’t care about transients as much). if you were mixing with headroom where the mix doesn’t currently clip, you will have to raise the gain to the point where it reaches a desired loudness above -7, inevitably clipping, ideally without audible distortion. there are some music where this isn’t necessary. warmer tracks such as acoustic music, older band performances and especially cinematic music where dynamics are crucial, don’t require any master changes. in extremely quiet tracks, where drums aren’t apparent or are just generally ambient, gain may be modified to either peak normalization (max signal of the master hits 0dB), or to fit the loudness appropriately depending on how it sits with the next or previous track.

  • limiting is something i’ve almost completely ditched in my mastering process for louder, modern genres. in most cases, limiting either ruins transients, ruins loudness, or both. the only times i’ve found myself using limiting are where the mix is dark, and dynamic sounds start to audibly distort the mix at a certain loudness. ideally, the mix should be fixed to remove that problem and so that you wouldn’t need limiting, but if it’s absolutely necessary, limiting should be automated for just that problematic section. 

  • when rendering a master, the standard format for submitting to distribution is 24-bit, 48khz wav. in rare (and in my opinion, very weird) cases, i’ve had labels ask me for 44khz, 16-bit wavs, but it does seem that more distributors are leaning towards the former setting.

i know that’s a lot and i’m probably missing a bunch but i just wanted to get the basic layout of how i approach my mixing and mastering. there are certainly layers to it and some definitely warrant its own article. in any case feel free to ask me any questions if you are at all confused with any part of it!