Masters Of The Art Of Mixing

What would you hear — and what might you learn — if eight top producers mixed the same project their own way?

loudness-matched-masters.zip

loudness-matched-stems.zip

Many engineers are refreshingly open about their craft, laying bare their decision‑making processes in SOS interviews and online videos. But a problem remains for students of mixing: they’re watching different cooks making different dishes with different ingredients. There’s never really been an opportunity to compare how different world‑class mixers would tackle the same multitrack. Until now!

We commissioned eight of the world’s top engineers and producers, whose professional experience spans everything from classical, jazz and pop right through to extreme metal, to mix and master the same song. It wasn’t a mix shootout, as such: rather, the aim was to reveal more about the different decisions, approaches and techniques of each engineer when presented with the same material and the same brief.

The track is ‘In Solitude’ by UK Metal Merger and, while it’s in the metal genre, the techniques, approaches and processes involved will be just as relevant to most styles. We invited all eight engineers to mix and master the track, with a brief that the music should sound ‘powerful’, ‘heavy’ and ‘well‑produced’, but that they should otherwise approach it as if they were the producer.

Needless to say, all the engineers delivered great results. But it was really striking just how diverse their approaches were, and how different their mixes sounded. We can’t cover every aspect of each engineer’s mix in fine detail here (that would fill a book!), so what we’ve done is to pick out some of the most interesting details about their approaches, and tried to tease out plenty of learning points along the way.

In case you do want to dive into the details, though, we’ve put many more resources online. You can hear the finished mixes (including one by Mark who made the initial recordings), download stems from each producer’s mix, and find a video below in which Mark explores the different mixes. If that’s not enough, there’s still more available at www.modernmetalacademy.com, including video interviews and screencasts that walk you through all eight producers’ mixes in detail — that’s all free to access, by the way.

Buster Odeholm: Templates & Tone‑matching

Buster Odeholm.Facilitated by a comprehensive mix template with pre‑configured routing, Buster Odeholm’s aggressive/in‑your‑face production aesthetic is achieved by optimising every aspect of the mix in meticulous detail. Many of his techniques and approaches are highly unconventional, and some even directly contradict commonly held ‘rules’.

For example, most rock/metal producers keep the low end of the bass clean, using EQ or filtering to apply distortion only to the mid and upper frequencies. By contrast, Buster’s highly effective bass sound in this mix (which, incidentally, was crafted entirely from the DI) involves heavy distortion of the bass sound’s low end. He explained that this produces a sound that’s not only very dynamically stable, but also “super‑fat”.

On the subject of sonic weight, Buster says the kick drum is his number one priority, as it’s paramount for driving the rhythm. Ensuring the kick always remains dominant means not only that the super‑fat distorted bass textures can reside in the lows, but also that the guitar tones can have significant weight. In a similar manner to EDM mixes, where the kick typically ducks almost every other signal, Buster’s kick triggers side‑chain compression on almost every signal, including the drum room mics, cymbals, bass, guitars, vocals, orchestration and effects! Importantly, though, he avoids the ‘pumping’ aesthetic of EDM by keying only a single band of a multiband compressor (in most instances everything below 120Hz) rather than full‑band compression.

A single band of multiband compression, triggered with every kick hit, is applied to almost every signal in Buster Odeholm’s mix.

Another interesting aspect of Buster’s approach is his extensive use of tone‑matching EQ, with FabFilter’s Pro Q 3. Once he’s captured the tonal ‘shape’ of parts such as the rhythm guitar sound in successful mixes, a given sound in a current mix can be quickly treated to give it a similar spectral curve. Every mix is different and needs to be treated as such, but this is a fast way to work and, rather than being someone else’s preset, it’s still his own sound. The approach frees up time and mental energy to focus on other important details of the mix, and it’s a tactic that’s well worth consideration if you want to streamline your own mixing workflow.

Finally, Buster’s master‑bus routing and processing chain provide an interesting contribution to his sonic signature. An instrument pre‑master bus has every mix signal routed to it other than the orchestration and vocals, and is heavily treated with EQ, compression, multiband compression and, courtesy of FabFilter Saturn, stacked tape distortion. Conversely, a second pre‑master bus comprising just the orchestration and vocals is left without any processing whatsoever. These pre‑master buses are then routed to a master bus, where a final round of tone‑match EQ, intelligent EQ (Soundtheory Gullfoss), tape emulation and limiting is applied.

Adam ‘Nolly’ Getgood: Space, Bass & Clarity

Former bassist for math‑metal band Periphery, Adam ‘Nolly’ Getgood is a progressive metal engineering and production specialist. During his interview, Nolly spoke passionately about his ideas around the perception of ‘heaviness’, which he equates to a literal sense of heaviness, in this instance represented by the sounds of massive objects or phenomena, such as large machinery or natural disasters. This association between heaviness and physical size/space is very much apparent in his highly three‑dimensional mix of ‘In Solitude’. Importantly, though, alongside a very natural sense of cohesion, he managed to retain the all‑important clarity.

Adam ‘Nolly’ Getgood.

Space and cohesion are recurring themes throughout the interview (which you can watch at www.modernmetalacademy.com). As one example, talking about sample‑based drum sounds, Nolly states: “They haven’t approached it thinking of the kit as a whole cohesive element that was recorded simultaneously, and in a space where it was recorded simultaneously. For me, there’s a loss of integrity that I’m looking for when I hear music, like, if I subconsciously or consciously can relate it back to a real event that happened — a drummer in a room playing.”

Adam ‘Nolly’ Getgood: We’ve got some boosts, one around 900Hz and one around 2k. Those will both be bringing some kind of clunky clacking character...

Nolly achieves space and clarity through emphasis on how things sound in context, and careful consideration of the impact of sounds on each other — almost entirely disregarding how things might sound when soloed. For example, when discussing his EQ on the bass bus, he says: “We’ve got some boosts, one around 900Hz and one around 2k. Those will both be bringing some kind of clunky clacking character that, to be honest, I didn’t find super pleasant to listen to on its own... but in the mix, is what really allows it to cut through.”

The EQ curves applied by Adam ‘Nolly’ Getgood to the bass and guitar buses, respectively.Interestingly, Nolly’s mix uses all five bass signal options made available to him (in contrast, Mike Exeter’s mix, which also exhibits a great bass sound, used just the one). To allow the bass and guitars to coexist and still do their job, Nolly applied intelligent EQ moves to both. Both exhibit relatively broad cuts in the low midrange, while the 80Hz boost in the bass is mirrored with attenuation in the same region on the guitars. Likewise, the 1.4kHz cut in the bass is mirrored with a boost to the guitars around 1.1kHz, and the broad 2‑5 kHz boost in the bass by three separate cuts to the guitars in that region.

Likewise, on the relationship between the cymbals and vocals: “I dynamically ducked the cymbals around the ‘s’ region of the voice. That’s something I found when working with bands with vocalists that need to be really upfront, you end up in this trade‑off: there’ll be cymbals that are too loud, clouding the articulation. This is where dynamic processing can help. So, instead of having to use broadband compression, making the cymbals quiet, or doing some really painstaking automation, approaches like this just allow the voice to not have to be super‑loud relative to the cymbals, and yet they can both be heard at the same time.”

A further example of Nolly’s consideration of space is in his use of Mid‑Sides EQ on the stereo strings track. He applied a roughly 5dB cut to the low‑mid portion of just the Mid, with a broad Q so it attenuated frequencies all the way from 100Hz to 3kHz. Not only does this gesture afford more space and reduce the potential for spectral masking in the centre of the mix, but also, by accentuating the panoramic extremes, it creates a wider image. The main takeaway point is that when putting together the audio jigsaw puzzle of a mix, we’re only ever going to create the overall picture we seek if we ensure the various shapes don’t overlap and obscure each other.

Dave Otero: Stacked Saturation & Multiband Expansion

Dave Otero.Feeling that the use of mix templates risks stagnation, Dave Otero prefers to start his mixes from scratch, and be guided by the song. Yet, there are still processing tactics that he’ll typically adopt for most projects, most notable of which is ‘stacked saturation’ on almost every channel. In this mix, this takes the form of Slate Digital’s Virtual Tape Machines (VTM) and VCC Channel plug‑ins, the latter set to the SSL 4000E channel, and with the Drive relatively high.

This stacked saturation tactic is a sort of ‘pre‑emptive strike’ to counter the potential for digital recordings to sound slightly ‘sterile’, with the various mix elements sounding somewhat disconnected. In some instances, for example on his kick bus when mixing ‘In Solitude’, signals that have already been subjected to this stacked saturation receive a further dose. In this case that came from Kazrog’s True Iron plug‑in, to add a little more ‘squish’ or ‘push’ to the sound’s low end.

Another fascinating Otero mix tactic comes into play on the drums. Most Sound On Sound readers will be aware of the challenges posed by the cymbal spill captured by close mics used on the shells. An aspect that’s often overlooked, though, is the relationship between the region where this cymbal spill in snare or tom tracks resides, and the way this region of the snare or tom sound naturally decays. Rather than gating or using Strip Silence, Dave employs multiband downward expansion (using FabFilter’s Pro‑MB) to great effect. This first involves generating MIDI notes for all snare and tom hits. These are then sent to the side‑chain input of the multiband expander. He typically divides the frequency spectrum into four bands. A fast release setting is applied to the uppermost frequency zone, thereby pulling down the accompanying cymbal spill the fastest, with the release time getting progressively longer in the lower zones.

Dave Otero makes clever use of multiband expansion to minimise the impact of cymbal spill on the drum close mics.

Josh Middleton: Drum Samples & Turbo Tone Referencing

Whereas Andrew Scheps (discussed below) didn’t use any drum samples for his ‘In Solitude’ mix, producer and Sylosis (ex‑Architects) guitarist Josh Middleton went in the opposite direction: he chose to replace the kick, snare and tom spot mics with drum samples. Crucially, though, Josh invested huge attention in the sample selection, sample blending and multi‑velocity triggering processes to ensure that, from both a performance and tonal perspective, the results sounded natural.

Josh Middleton.

Especially for the more pitch‑based snare and toms, Josh spent a long time auditioning samples before selecting the candidates that he felt were the most pitch‑coherent with the original acoustic sources, and which had the closest sustain/ring qualities. He then worked to further optimise the pitch coherence, using the ‘tune’ (pitch) feature of Steven Slate’s Trigger 2 plug‑in, including by experimenting with the pitches and polarity settings of the drum samples when they were blended. No doubt there will soon be AI‑based tools that can carry out this sort of analysis of pitch, polarity, phase and attack/decay/sustain/release qualities, but for the moment, there are no good substitutes for time and patience!

Josh Middleton uses Slate’s Trigger to replace most of the close mics with carefully chosen and pitched samples.

It was no surprise that Josh also delivered some great guitar tones. In the video interview we recorded after hearing his mix, he explained something about his approach to guitar tone more generally, and it’s worth highlighting here. He uses references extensively, and goes into painstaking detail to analyse the sound and figure out how to create a similar tone. He’ll often find a section in a song where the guitar sound is exposed, and will learn to play that part accurately himself. Then, he’ll run his DI’ed sound through his signature STL Tones Tonality: Josh Middleton plug‑in, which emulates his signal chain (including his preferred pedals, amps — two Peaveys and one Mesa Boogie — and cabinets), and try and recreate the tone of the reference.

By level‑matching then A/B comparing the two versions, he establishes the way his own tone differs from the intended production in terms of parameters such as drive level, midrange emphasis, brightness, presence and so on. And he can then use that tone in his productions, and knows just how to tweak it to sit it in a particular mix.

Fredrik Nordstrom: Technique Over Technology

Fredrik Nordstrom.In a world where we’re continually being sold new and better plug‑ins and many of us suffer with gear acquisition syndrome, the concept of ‘ear over gear’ needs highlighting, and Fredrik Nordstrom’s mix and accompanying interview are a revelation — well worth studying.

Fredrik has a mantra: he doesn’t want to impose his sonic signature on a band he is working with, but instead works to capture and translate the band’s vision and collective personality in a recording. To that end, he adopts keep‑it‑simple‑stupid, less‑is‑more processing tactics. To craft this wonderful mix Fredrik used, with just a small handful of exceptions, stock Pro Tools plug‑ins. Yet he managed to separate the signals so they all ‘sing’ in an effective way, without interfering with the ability of other sounds to do the same — and did so in such a way that no mix automation was needed either!

Some guitarists think that investing in a top‑of‑the‑range Mayones guitar will help their picking technique. It won’t. Only practice will. And the same advice applies when you’re seeking to improve your mixing: direct your focus toward mix technique before mix technology; achieve the desired sound with minimal processing, rather than 27 moves using the latest weird and wonderful plug‑ins, and the results will almost certainly sound more organic, and allow the listener to connect with the performances without the production getting in the way. So why not grab a single parametric EQ, a couple of compressors and a de‑esser, a couple of verb and delay units, and mix as much music as you can? If you think this approach would be too restrictive, take five minutes to listen to Fredrik Nordstrom’s mix of ‘In Solitude’. You won’t regret it!

Mike Exeter: Accentuating Stereo Width

Mike Exeter.Given his daunting level of music‑industry experience and long‑term, high‑profile clientele such as Black Sabbath and Judas Priest, we were incredibly keen to learn how Mike Exeter would tackle ‘In Solitude’. Mike’s thoughts about low‑frequency content are fascinating. He argues, very convincingly, that high‑pass filters tend to be over‑used when ‘dialling in’ the sound in this region. “You’ll notice I never use filters,” he told us. “I don’t like them. I think they’re indiscriminate in what they do. So, I tend to use shelves, and I turn down the bottom end. It’s probably why there’s so much bottom end in my mixes... I don’t want to just indiscriminately go ‘everything below this frequency is useless’ because it’s not. High‑pass filters are a very, very severe, unmusical‑sounding thing, dismissing everything down low, and you don’t necessarily need to dismiss it.”

Mike was also the only producer to program a sub‑bass synthesizer to add more weight and heaviness. But perhaps most fascinating is his method of optimising stereo width, often through extensive use of automation. Many of us use volume automation or clip gain adjustments to ensure each tom fill delivers the right impact. But how many of us use pan automation? Mike automates each tom with a different pan position, according to the number of toms used in each fill — if a fill uses only two toms, he’ll pan those two toms wider than in a fill featuring all four toms.

Further tactics for optimising stereo width include the use of different guitar amps for each side of the mix. The greater the difference between left and right, the greater the perceived stereo width, and Mike further accentuates these differences through his EQ choices. He also uses Brainworx’s bx_console 4000G plug‑in, in which different channel types can be chosen to provide non‑linearity, across all 14 of his submixes. “Dirk from Brainworx will tell you that they’ve gone in and analysed every channel in an SSL. What they’ve actually done is randomised every single component, or certain key components, by about one or two percent. When you choose a channel strip, it’s changing values. If you were to put two identical things up left and right, let’s say two guitar sounds, and then you went into this analogue mode, you’d feel a width change.”

Jens Bogren: Vocal Texture, Impact & Clarity

Jens Bogren.A production area seemingly sometimes neglected by metal mix engineers is the vocal texture, impact and clarity. Jens Bogren, though, believes that the vocals are “always the most important thing”, and adopts a pop‑like approach, automating almost every syllable of every word. Jens begins in this mix by routing the vocal through three hardware compressors: a UA 1176, a Teletronix LA‑2 from the ’60s and an Empirical Labs Distressor. He believes that the software versions still don’t generate quite the same harmonic content, which is as important to him as dynamic stability.

In a metal context, says Jens, vocal signals require distortion to help them blend with, yet at the same time cut through, the guitars. “I think that I tend to use compression and distortion, and distortion by compressing, to the point where I put it in the mix then use EQ to fix the rest. A big part of the frequency response and the EQ curve comes from compression — you compress it with fast release times and make it distort. If you use pretty extreme compression like the Distressor, for example, I have this British mod switch that makes it distort, creating more treble, getting rid of low mids, and suddenly you have a mixable vocal track. If I start by trying to EQ and EQ and EQ, it will never end, I get this spiky vocal track that’s really hard to get in there.”

Jens Bogren’s vocal EQ settings.

Because of these tonal changes brought about by compression, only subtle EQ moves are required on the vocals. A Baxandall‑style curve (delivered using peaking EQ) lifts the ‘air’ frequencies upwards of 15kHz, creating more of a pop‑like sound, and this is joined by a number of moderate cuts within the upper mids and highs, a moderate low‑end lift around 150Hz, and a brightness/intelligibility‑enhancing boost around 5.5kHz.

Bogren also used a de‑esser, a Waves C4 multiband compressor — almost like an auto‑mixer to tame the frequency areas where the vocal performance is most prevalent at any given instance — and Soundtoys’ Decapitator for a touch more harmonic distortion. But even then Jens’ forensic approach to the ‘In Solitude’ vocals was not complete. Sending the vocal to the side‑chain input of a Sonnox Oxford Dynamics compressor inserted across the rhythm guitar bus (which, in this case, was the mix component most likely to mask the vocals) allowed the guitars to be ducked slightly by the voice. This ensures that the vocal remains ‘foregrounded’.

Jens also told us that for vocals, he prefers delay to reverb, and also ducks the vocal delay return, allowing the delay to swell in level at the end of lines, when there is more space for it in the mix.

Andrew Scheps: Parallel Drum Distortion

Last in our round up, but actually the very first contributor to this project, is the great Andrew Scheps. Really, he should need no introduction, but just to be on the safe side: his Grammy Awards for Best Rock Album with Red Hot Chili Peppers’ Stadium Arcadium (2006), Album Of The Year for Adele’s 21 (2011), and Best Reggae Album for Ziggy Marley’s Fly Rasta (2014) should provide a broad sense of his calibre and credentials. Other artists he’s worked with include Alanis Morissette, Johnny Cash, Jay‑Z, Adele, Lana Del Rey, Lady Gaga, Beyoncé, Metallica, Muse, Linkin Park and Black Sabbath.

Andrew Scheps.

Andrew has worked across many genres, then, so it’s perhaps not surprising that his approach to mixing ‘In Solitude’ ignores many modern metal production conventions. The usual emphasis on ‘hyper‑reality’ is largely replaced by more natural tones and a more organic listening experience. One of the principal reasons for this is his decision not to use any drum‑sample reinforcement, as we alluded to earlier.

Scheps is very well known for his use of parallel compression, especially for drums. Yet here, in response to the wall of down‑tuned, harmonically distorted bass and rhythm guitar tones, his tactic for giving the drums the necessary punch, sustain and tonal density involved greater emphasis on parallel and direct drum distortion. This approach, he says, creates a drum sound that sounds more three‑dimensional, with lengthened attack and sustain from the snare and toms, and their energy being refocused towards the midrange and ‘presence’ frequencies.

“Harmonic distortion,” he explained, “gives you more audio to work with when you’re EQ’ing. So that’s why I like it. You can do a half dB in the midrange on something that’s distorted, and that is a gigantic difference. And without the distortion, it’s sort of only the notes that have things close to that frequency unless you’re doing a really broad EQ move.”

In practical terms, the devil really is in the detail and, as always, context is everything. Andrew describes the lengthened attack and sustain resulting from parallel distortion as almost providing an alternative to reverb. Nevertheless, drum reverb is used as well, albeit with relatively short reverb times of less than a second. Bespoke snare and tom aux reverbs were created using Avid’s Revibe II, with 835 and 876 ms decay times respectively. He also used a global drum reverb, using Liquidsonics’ Seventh Heaven: this had a decay time of 750ms, and was routed directly to the mix bus.

Andrew Scheps used Liquidsonics’ Seventh Heaven as a global drum reverb.

In addition to relatively heavy saturation from his own Waves Scheps Omni Channel plug‑in, which was applied directly to the kick, snare and toms aux buses, Andrew used Soundtoys’ Devil‑Loc Deluxe in the first insert slot of the global drums aux track, with a relatively subtle mix setting. This processes the post‑insert sends from that aux channel, which are routed to multiple destinations: the global drums reverb we’ve already discussed, a parallel heavily EQ‑sculpted ‘drums dirt’ channel (which is distorted using iZotope Trash), a further parallel instance of Devil‑Loc Deluxe (this time with a fast release and the Mix setting on 10), and UAD‑modelled dbx 160 compressor and Empirical Labs EL7 Fatso Jnr saturator.

Overall, Andrew’s focus on the acoustic drum recordings, which he heavily enhanced through parallel distortion and a significant emphasis on ‘shared processing’, results in a highly cohesive and original drum sound. Oh, and as a parting snippet of info that we personally found interesting: when it came to ‘mix glue’, as an alternative to master bus compression Andrew used an Eventide reverb on his master bus, with a mix setting of 15 percent.

Takeaway Time

So, what can we all take away from this experience? As you can hopefully see from the discussion in this article, there are many possible tactics, tools and techniques that can be used to achieve a good mix. There’s no one ‘right answer’ when presented with a set of multitracks. What every producer brings to the table is their own unique combination of musical and sonic tastes, different experiences and years of developing their own mixing techniques. It’s this that results in such a diverse but wonderful‑sounding collection of mixes — and while any student of mixing can learn a lot from studying each in isolation (and we’d absolutely encourage you to watch the video interviews!), it’s arguably more important to find out which techniques suit your own personal style and way of working.