The chance to mix a Rick Rubin production was too good for Billy Bush to pass up — but Rubin's minimalist tastes presented plenty of challenges.
Jake Bugg has enjoyed the kind of rise for which the word 'meteoric' was invented. The Nottingham native was discovered in 2011, aged just 17, when he appeared on the Glastonbury Festival Introducing Stage. A deal with Mercury Records followed, and a year later, his eponymously titled debut album reached number one in the UK. This was not only an impressive feat because of Bugg's youth, but also because of his arguably rather unhip musical direction: American roots music influenced by folk, blues and early-'60s rock — it had critics referencing Bob Dylan, Buddy Holly, Donovan, Johnny Cash and even John Denver — with some punk-ish Britrock thrown in for good measure.
The success of Jake Bugg was partly down to the work of Mike Crossey, who was featured in December 2013's Inside Track talking about his work with the 1975. Among those impressed by the album was legendary producer Rick Rubin, famed for his work with everyone from the Red Hot Chili Peppers to Kanye West, and Johnny Cash to Metallica, and so it was that Bugg found himself in Rubin's Shangri-La Studio in Malibu, just north of Los Angeles, for some try-out sessions in the summer of 2013. The sessions went better than anyone hoped, and more songs were written by Bugg, with occasional help from Iain Archer (Snow Patrol), guitarist Matt Sweeney and Brendan Benson (the Raconteurs). An all-star band consisting of Sweeney, drummer Pete Thomas and bassist Jason Lader was amassed, and by September, 14 tracks for a second album were completed. At the API desk at Shangri-La studios were David Schiffman and Dana Nielson, while mixing was outsourced to Andrew Scheps (see SOS September 2013 issue for his mix of Black Sabbath's Rubin-produced 13) and Billy Bush, best-known for his work with producer Butch Vig and the band Garbage.
"Mixing half of the Jake Bugg record really took me out of my comfort zone,” recalls Bush, "because most of the things I have worked on are very dense, layered, soundscaped kind of productions, with lots of instrumentation and effects, tons of vocals, and so on. When I first looked at the Jake Bugg sessions, I was like, 'Wow, this is really stripped-back.' There'd just be drums, bass, one or two guitars and a vocal, and sometimes just a guitar and a vocal. It really was an exercise in restraint for me to not get in the way of the song and the performances, while still making it sound unique. There wasn't any particular brief, it was more like, 'Do your thing,' though they sent me the two singles which Andrew had already mixed ['What Doesn't Kill You' and 'Slumville Sunrise'] as a reference. They were much more punky rock than the tracks I was given to mix, so Rick assured me that I didn't need to make the songs I was mixing sound the same.
"They were going for a rootsy '60s feel, but you can't literally make it sound like that, because it wouldn't sound right in the context of today's music. It really needed an extra level of polish to make sure it also sounded modern, while still making sure it retained its authenticity and sense of realness. This was tricky enough in its own right, and became even more challenging when I found out that Rick doesn't like reverb or in fact any effects that he can hear. This meant that I had to find ways to create space for things that made sense and that made it feel like a finished record, without using many of the stock tricks that I would normally apply. It was a really fascinating and challenging process to figure out how to mix a track without doing anything that would be obvious or noticeable. On top of this, Rick came back to me with notes that, on the surface, seemed somewhat vague and innocuous and that left me scratching my head. For example, one note read, 'It needs to be lighter on its feet.' But when I then listened to the track in question with that note in mind, I'd completely get what he was talking about.”
It was the first time that Bush had worked with Rubin, and there was a learning curve involved for the mixer on that front as well. He got the gig as a result of taking part in a mix shoot-out in early 2012 for the Linkin Park album Living Things, even though that gig eventually went to one the world's biggest names in mixing, Manny Marroquin. "A few weeks after the shoot-out my phone rang,” explains Bush, ”and it was Rick, saying that he wanted me to know that he really loved what I had done and had voted for me mixing the record, but the band had decided to go with someone else. He invited me over to his studio in Malibu and we hung out for an hour and talked about music, and then, more than a year later, totally out of the blue, he called me to ask whether I was willing to mix half of a new Jake Bugg album. I was like, 'Yeah!'”
Billy Bush's approach to mixing embodies a rather unusual working method which involves elaborate use of several layers of aux, subgroup and VCA tracks. The original session for Shangri-La's opener, 'There's A Beast And We All Feed It', for example, contained only 19 audio tracks, to which Bush added a whopping 26 non-audio tracks. Similarly, the acoustic guitar ballad 'Pine Trees', which features just Bugg and his acoustic guitar, spread out over four tracks, saw Bush adding nine aux, subgroup and VCA tracks. The fact that Bush is essentially self-taught presumably contributed to him developing this slightly leftfield method.
"I probably learnt 80 percent of what I know from Butch, and the rest from self-study,” notes Bush. "I love learning about gear and techniques, and I am constantly reading about it and researching things and trying out new things. It's one of my favourite pastimes, much to the chagrin of my wife [laughs], who is an artist, and therefore more interested in the result and not so much the process. But for me, the process is where I get to be creative and have fun. I'm always exploring new pieces of gear and trying out different options, and seeing what they can add. During the making of the Garbage albums I got many real crash courses in all the pros and cons of different gear, in particular the analogue and digital worlds. With the current Avid HD interfaces I am the happiest I have ever been in terms of how digital sounds. To my ears things come back sounding exactly the same. My goal has always been to try to get the audio to sound as much as a real-life experience as possible, so I really feel like I'm not compromising at all now by working in the digital domain.
"I was always sceptical about the whole higher-sample-rate thing, but there are times when sessions with real instruments done in 96k do sound more organic. The Jake Bugg record was one of them, because there definitely was something about the sounds of the cymbals and acoustic guitars that translated better in 96k. It was less work to get these things to sound correct. Rick wanted the Jake Bugg album to have an analogue feel and flavour, and this was much easier to achieve for me by working in 96k. Higher sampling rates allow me to get the digital domain to react like the analogue world, and achieve a solid stereo image and great definition and three-dimensional clarity in which it's possible to hear all the reverb tails.
"With the Jake Bugg album they were going for something very real, very visceral, and so they were trying to record things live in the studio and going for whole takes. There were a few songs that I mixed that were comps of two or more takes, but at least a couple were one take all the way through, totally live in the studio. 'Pine Trees' was one of them, and 'Storm Passes Away' another. They were going for the takes that felt best, and I go for the same thing while mixing. If a take, or a mix, has something that makes people react, then it's like: 'That's the one.'
"The rough mixes for Shangri-La had been done in the analogue world, on the API desk there, with outboard treatments, so when I received a session, the rough mix would be separate. I'd listen to the session and I'd listen to the rough mix and I'd then try to get the session to sound like the rough mix. This process involved housekeeping things like getting the kick drum and other details to sound right, and dialling in effects, and so on. Once the session sounded like the rough mix, I'd switch gears and tried to approach the song from an emotional point of view. What is the song making me feel while I listen to it? Does it excite me? I'd ask myself whether there was anything about it that was not making me feel connected enough to the track. I'd listen to any spot where my attention would be wavering from the song, where I would feel easily distracted. If this happened at any point, I'd look at what I could do to make something happen that would draw your attention back to the song. Once I got to the stage where magic was happening I'd go, 'That's the one,' and I'd send it to Rick. He'd come back with his comments, which would also be about the feeling of the track. Like when he wrote, 'This needs to sound like Van Morrison's Astral Weeks,' despite the fact that the song in question sounded nothing like any of the songs on that album. But Rick would be referring to the emotional response he wanted from the track.”
Shangri-La's opening track, 'There's A Beast And We All Feed It', is an uptempo number that wears its late-'50s/early-'60s R&B and British skiffle inspirations on its sleeve, in the feel, instrumentation, recording and particularly the distorted, lo-fi, nasal vocal sound. As mentioned above, from the 19 actual audio tracks, containing drums, claps, bass, electric and acoustic guitars and lead vocal, Bush created a complex mix with several layers of subtle effects using many different non-audio channels.
"They were sending me the tracks as they were being cut at Shangri-La, via download links,” elaborated Bush. "The sessions would be stripped back to include only the take or the edit that had been approved, plus the rough mix of each song, so that there was no confusion over what needs to be in the song and what not, and what track is the lead vocal, and so on. Occasionally there might have been a part with a message saying, 'See if you can work this in, but if it doesn't work, just leave it out.' My first step after receiving a session for mixing is to go through it track-by-track, part-by-part, and if necessary rearrange the session in a way that makes sense to me. Some guys have an assistant to do that for them, so they can just walk in and start mixing, and I am completely envious of that! But it's become part of my process to spend those two hours going through everything and seeing what's there and what things are working and not working, and to set up all the routing and so on.
"Once I have everything set up the way I want, and have gone through all the individual tracks and have routed things, the mix will be pretty close to the rough mix, and I switch over to real mix mode. Of course, sometimes an artist will tell me that the rough mix is not important and will ask me to take the song somewhere else, but most people I work with have a very specific vision of what they are trying to accomplish, so I need to take that as my starting point, and my job then is to create the best possible version of their vision, and if possible go beyond that. Real mix mode initially involves me working on sections. My D-Control only has eight faders, and this works for me, because I concentrate on a couple of instruments at a time. Most rock and pop music is about the groove, so I tend to start with the drums, and I'll then work in the bass, and then I'll move on to the vocals, and after that I'll work the other instruments, like guitars and keyboards, in around that. So my first focus is to get a great groove, and then to make sure the vocal works with that, and then fit everything else around the vocal.
"The audio of the 'Feed The Beast' session came to me pretty much as you can see it in the screenshots. The only thing that's important to me with regards to colour-coding is that things like instruments, mics and track types have the same colour, so in this case the two overhead tracks are in purple, the three clap tracks in red and the three electric guitar tracks in green, the vocal effects subgroup tracks in dark grey and the instrument VCA tracks in purple, and so on. There's very little in this session, with the drums just being kick, snare, overheads straight and side, two room tracks, and drum talkback mic. In addition there are three clap tracks, four bass tracks — two DI and two amp tracks — three electric guitar tracks, one acoustic guitar track and a lead vocal comp. I added some chamber and plate reverb and tape slap echo, which was all I could get away with. If you take it away, it sounds completely flat and you can't tell where the vocals are coming from. I personally do like reverb, and in this track I used many different effects in very small quantities to create something that gives you a sense of space.”
To understand the mix of 'There's A Beast And We All Feed It' it is necessary to grasp Bush's multi-layered mix approach. The mix engineer left the instruments at the top of the session, in the traditional order — drums, percussion, bass, guitars, vocals — and bussed some related groups of tracks such as basses and electric guitars to single aux tracks. So far, so normal. But there are also seven vocal aux tracks with many different effects, followed by a set of subgroup mix tracks for the drums (stereo), percussion, bass and electric guitar (all mono), acoustic guitar (stereo), and then one single submix track for the vocal and one for the parallel vocal compression. Next up are eight VCA tracks, respectively for all subgroups, drums, percussion, bass, electric guitar, acoustic guitar, vocals and all VCA tracks. Following this are the tracks used for mixdown, with the stereo mix, which was sent out to the analogue domain, coming back in on a 'Mix Bus' track. Different versions of the final mix are printed to yet more audio tracks, and at the end of the session are stem tracks for all the individual instruments.
Bush traces the ins and the outs: "Basically, the way I mix is that I first approach the individual tracks separately. I EQ or compress them via plug-in inserts on the channels, and set levels at the channel fader until they sound cool individually. Then I view the kit, or all the bass tracks, or all the guitars, and so on, as one group and send this to aux tracks to process these group sections as a whole, with plug-ins and/or outboard on the aux tracks' inserts. Each instrument group also goes to a VCA fader, and for example the drum group controls the level of the kit going into the aux processing so I can control how hard I am hitting the processing. I then also have the auxes grouped and controlled by the VCA called 'All subs' that controls the level coming out of the aux groups before it hits the D-A converters and enters the analogue realm. I also blend all the instrument group VCAs together and control them via the 'All VCA' fader, just in case I need to turn down all the tracks before they hit the auxes — for example in the case of a very hot recording that would clip the processing at the aux tracks.
"The reason I lay out my sessions like this is that the more processing you apply, the more you lose the inherent character of the original track. Particularly with drums and vocals, you start to lose transients, and with that, the ability for things to cut through the mix. For this reason I want to be able to have full control over every bit of processing I apply, and I prefer to have the effects mainly in parallel. The VCA channels combine the original signals of the individual tracks, and with them, I can affect the gain of each track, pre-processing. If I want the drums louder I turn up the VCA, and this will hit the processing harder but not the processing outputs. If I want to turn up the output of the processing, I turn up the drums subgroup. It's difficult to describe what my setup does without being able to listen to it. But if someone is in a room with me, he or she can for example hear a kick drum sounding pretty good, but the moment I add some processing it won't cut through, and so I put some of the original back in, and suddenly all the original transients are there again, meaning the drums now sound cool and cut through the track. The tracks for this session were recorded really hot, so I did in this case use the VCAs to pull the fader outputs back a bit, so I'm not hitting the subgroup aux processing too hard. One thing that the analogue and digital world have in common is that they both have sweet spots, ie. the places where you have optimum level. When you are in that zone, everything sounds great, but if things are too hot they sound small, and if they are too quiet they sound thin. The combination of aux subgroups and VCA tracks allows me to really find and optimise those sweet spots.”
- Drums: SPL Transient Designer, Waves CLA Drums & API 550, McDSP Filterbank F2, FabFilter Pro-Q, Avid EQ III & Expander/Gate III, UA Studer A800, Roger Mayer RM57.
"The kick has the SPL Transient Designer on it, because I felt that the kick was a little soft and woolly-sounding. I took out most of the sustain and added a lot of attack to get it to cut through. It also has the Waves CLA Drums plug-in, which is an EQ to add some bite and low end. The API 550 is adding a ton of 1.5kHz, again trying to get the kick to cut through, and the McDSP Filterbank F2 is simply a high- and a low-pass filter, cutting out everything under 20Hz and above 2.1kHz.
"The snare and two overhead tracks all have the FabFilter Pro-Q EQ on them, and that's all for individual track processing. All drum tracks go to the 'Drum Sub' group, which has an Avid EQ, Universal Audio Studer A800 analogue tape simulator, and Pro-Q EQ. They also go to the 'Drum Smash' aux track, which is the parallel compression. For this track I duplicated the Avid EQ, UAD Studer A800 and Pro-Q, and added my Roger Mayer RM57 outboard compressor and a D3 expander gate. My Roger Mayer is from the 1970s. It's an incredible drum compressor, which makes anything sound immediately like Keith Moon! It really makes the kick and snare explode. Mine is the rack unit, and the VU meters are smashed, because I always totally overload it! The gate comes in after the RM57, because it is a bit noisy.”
- Percussion: UA 1176LN.
"The percussion, aka the three clap tracks, don't have any inserts, and are sent to a percussion submix track, on which I have a Universal Audio 1176LN limiter, just to control the level. The thing I like about the UA compressors is that they manage to model what the outboard model does tonally. I'm often not using the compressors for compression but rather for the tone that they add. I just like the way things sound when you send them through a compressor, even if it only compresses 1 or 2 dB.”
- Bass: Little Labs PCP Distro, Audio Kitchen Big Trees, FabFilter Pro-Q, Avid Expander/Gate III, UA LA2A & Pultec EQP 1A, McDSP ML4000.
"The four bass tracks consisted of two DI tracks and two amp tracks, but I didn't use one of the amp tracks because I didn't like the way it sounded. I didn't do anything on the three tracks individually, but instead pulled them together to a sound that I liked, and then I effected that sound. So I sent the three bass tracks to an aux track right next to them, on which I use a Little Labs PCP Distro to send the signal through an Audio Kitchen Big Trees valve amplifier, and then back in and through another Pro-Q EQ and then the D3 gate. The three bass tracks and the Big Trees bass aux track all go to a 'Bass Sub' track, on which I have the United Audio LA2 compressor and the Pultec EQP 1A EQ, and then another Pro-Q, and finally the McDSP ML4000 multi-band limiter. The EQP 1A tweaks the overall low and high end, while the Pro-Q is more surgical, adjusting some specific frequencies. Similarly, the LA2A does some general, overall compression to make the sound bigger and give some more control over the level, and the ML4000 lets me control the dynamics of individual elements of the tone.”
- Guitars: Peavey spring reverb, Waves CLA Guitars & CLA Unplugged, Avid Euphonix Channel Strip, UA 1176LN.
"I had nothing on the three individual electric guitar tracks. Instead they all go to an adjacent aux track which has an outboard Peavey valve spring reverb on it. The three tracks and the aux track then go to the electric guitar sub track, on which I have the Waves CLA Guitars plug-in. I like the way the EQ and compressors sound with that plug-in. They did a really good job modelling them and I often throw this plug-in on to see whether or not it will work and that one plug-in can do everything, or whether I need two plug-ins. The acoustic guitar track has nothing on it, and is sent to two acoustic guitar sub tracks, because I was applying parallel compression to the acoustic guitar as well. Both sub tracks have the Avid Channel Strip for some compression and EQ, and one of them has the Waves CLA Unplugged and the parallel compression track has the UA 1176A set to 'stun'. But I ended up not using the track with the CLA on it, and kept only the 1176 track, which helps to get the acoustic guitar to cut through the mix better.”
- Vocals: UA 1176A, Plug-in Alliance Maag EQ, FabFilter Pro-DS, Waves L1, Doubler, SSL EQ, S1, De-Esser & Abbey Road J37, Bricasti M7, Eventide H8000, Sound Toys Echo Boy & Decapitator, Little Labs PCP Distro, Audio Kitchen spring reverb, Avid Expander/Gate III, Mod Delay & EQ III.
"Jake has a very unique vocal sound, and as with any singer, you find the aspects of the voice that make the singer like him or her and you feature them. The original vocal sound was a little distorted because they were going for a certain vibe, and I liked that, because it had a real energy to it. Some of my processing will have accentuated that distortion, like the Decapitator, compression, and a spring reverb that has some graininess. Other than that, working on his vocal was for the most part a matter of finding the spot where it sat well in the track. The lead vocal track itself has quite a few plug-ins on the insert: the UA 1176A, the Plug-in Alliance Maag Parametric EQ, the FabFilter de-esser, and the Waves L1 limiter. I like the Maag because it reminds me of the Sontec mastering EQ: it has a 20kHz shelf that allows you to add a lot of air to the vocal. The L1 is there to keep the level under control.
"The lead vocal also goes to six different aux tracks. It first goes into the 'Vox Comp' track, which has the L1 limiter on the insert and to two outboard pieces of gear on the sends, the Bricasti M7 and the Eventide H8000. There are two presets that I like a lot on the H8000, one of them being a plate, the other a micro pitch-shift and slap delay, but neither was appropriate for this song, so in the end I did not use it. The next vocal send goes to the 'Doubler' track, which has the Waves Doubler on it, doing what an Eventide H3000 micro pitch-shifter would do, and it then goes through an SSL EQ and a Waves S1 Stereo Imager to spread things out a little more. The send below that goes to the Spring aux, which has loads of plug-ins, starting with the Waves De-Esser, to take out as many esses as possible, because a spring reverb really accentuates them, then the Echo Boy delaying a 16th note, which acts like a pre-delay for the spring reverb, then an EQ to narrow the bandwidth of the vocal, and then it goes into the PCP Distro which connects it with an Audio Kitchen spring reverb that was custom-made for me, and on the way in that then goes through another gate to get rid of any noises. As I said, Rick does not like any reverb that he can hear, he likes reverb that he kind of feels, and the cool thing about a spring reverb is that it does not really stick out and does not have a long trail that you hear afterwards, like a plate does or other reverbs, yet it naturally creates a sense of space.
"The fourth vocal send goes to the 'Tape Slap' aux track, which has the Waves version of the Studer J37 on it, the Abbey Road tape recorder, and then the next send routes to the Bricasti aux track, which first has the Avid Mod Delay acting like a pre-delay to the Bricasti. It then goes through an Avid EQ3, which is set as a high-pass filter, taking out some of the high and low end, because I wanted the Bricasti reverb to be dark and unobtrusive-sounding. Then there's the insert to my Bricasti M7 outboard reverb, set on a 'Sunset Chamber' setting. The final vocal send is to the Decapitator aux track, which has the Sound Toys Decapitator to add some bite and aggression. The Vocal Compression and Decapitator aux tracks are the ones that I used to push the vocal out in front. At moments when the vocal was swamped by something else in the track, I'd bring those faders in. It may seem like there are many vocal effects, but what each of them does is very minimal. It's when you put them all together that they create a certain vibe. It was the only way I could get a sense of space in the track while keeping it feeling natural and true to the original recording. All these effects didn't stop the song from being really straightforward, because everything is so subtle. If I would turn everything off, it would still sound 90 percent the way it does. But this is what gives you that extra 10 percent.”
- Mix bus & stems: Shadow Hills Equinox & Mastering EQ, Manley Massive Passive, Cranesong HEDD, UA ATR102, Pultec EQP-1A, Slate Virtual Bus Compressor.
"The subgroups fed the Shadow Hills Equinox, which summed them together to stereo and there was some analogue two-bus processing from the Manley Massive Passive EQ and Shadow Hills Mastering EQ, before it came back into the session via the Cranesong HEDD. The two-mix came back into the session via the 'Mix Bus' track, on which I had additional processing from the UAD ATR 102 tape emulation plug-in, to give it a slightly more analogue feel. This really added an extra level of polish. Then there was again the UAD Pultec EQP1 and finally the Slate Digital 'Red' Virtual Buss Compressor. The Focusrite [this plug-in emulates the Focusrite Red Compressor] added an extra level of maximisation that made it sound competitive when I sent the mix to Rick and the record company, but I took it off when I sent the final mix to mastering. I left the tape emulation and Pultec on, because they were part of the sound.”
Billy Bush's own long and windy road to Shangri-La began in 1988, when he was fresh out of the College of Music at the University of North Texas and gearing up for a career as a musician. "I had a car crash in that year that cut short my guitar-playing career. In order to pay off my college and medical debts I started going on tour with bands as a guitar and keyboard technician. I'd always had a love of gear: I had a four-track when I was young, and I was always recording the bands I played in. After six years as a guitar and keyboard tech on tour with various bands I was invited to help Garbage figure out a way of recreating live what they had done in the studio. This was in the beginning of 1996. While we were on tour, Butch [Vig] wanted to explore the world of digital recording more closely. When he'd worked on Nirvana's Nevermind (1991) as an engineer and producer he'd used digital tape for the first time, and he was looking for a way of doing the editing in the digital world. He asked me to find the best digital recording mechanism, buy it, learn it, and show him how to use it. That was the beginning of my career as a studio rat!”
For a decade, Bush's studio-rat career involved working predominantly as a live and a studio engineer for Garbage, and also as a mixer and additional producer on the band's most recent album, Not Your Kind Of People (2012). Several of the band's albums were recorded using both analogue tape and Pro Tools, in Butch Vig's Smart Studio in Wisconsin. Bush has also engineered several of Vig's non-Garbage-related productions. During the last near-decade Bush has branched out into acting as a producer, engineer and mixer for other people, and in so doing he has earned credits including Eric Avery, Tom Gabel, Muse, Fink, Foster The People, the Offspring, Beck, the Subways and Natasha Bedingfield. Five years ago Bush moved to Los Angeles, where he currently lives with his wife, Garbage singer Shirley Manson, and where he has his own digital-meets-analogue studio called Red Razor Sounds. (Vig, who packed up Smart Studios a year ago, lives a five-minute drive away.)
"Engineering, producing and mixing each use different parts of my brain,” reveals Bush. "I love doing all three together, and seeing an entire project through from conception to fruition. It's a lot of pressure, but it also has its own rewards. When you are producing the relationship you have with the artist is really important. I really enjoy getting to know the artist and helping them accomplish their vision. I love engineering because it allows me to be sonically creative and to come up with sounds that people have not heard before. Mixing uses yet another different set of creative tools. I often mix remotely, so it's just me in my studio, which is not in my house but in a facility that contains a whole bunch of other studios. Mixing is pure creative fun for me. It is not about the relationship with the artist, but a matter of me camping in my underground bunker until I get something that hopefully is more than what the client is looking for. I tweak things until I think the mix sounds cool, and I then send it off and see how people react.
"My studio is built around an eight-channel Avid D-Control and an HDX Pro Tools rig with the new Avid HD interfaces, and the Antelope Audio Atomic Clock and Trinity Clocking System. I have quite a bit of analogue outboard, including a 30 I/O Shadow Hills Equinox summing mixer, the Cranesong Phoenix HEDD, Audio Kitchen Big Trees, Roger Mayer RM57 compressor, Bricasti M7, Manley Massive Passive EQ and so on. Most of my outboard is EQs and compressors, because I feel that those are the two things I miss the most in the digital world. There's something that happens in the harmonics when you hit the transformers in analogue EQs and compressors that you don't get in digital. But I don't really get hung up on the tools that I use. Instead I always go for a specific thing that I hear in my head, and I don't care how I get there, whether by using analogue or digital equipment. There are pluses and minuses to both media. When we were recording at Smart we did not necessarily enjoy what tape was doing. If we really wanted the transients in there, we'd use digital. The thing about my studio is that it gives me the flexibility to go in either direction. Sometimes I'll mix completely in the box, sometimes I'll split tracks out to the Equinox on an Iron setting, and I'll compare the two, and I'll send them out so other people can compare them, and from there a decision will be made whether to go the digital or the hybrid analogue-digital route.”
'Pine Trees' is a single take of Jake Bugg singing a short ballad, with his vocal and acoustic guitar recorded together. Strings and an acoustic guitar overdub were added while Billy Bush mixed, but producer Rick Rubin eventually decided that the single acoustic guitar/vocal-only version was superior. The original acoustic guitar was recorded using a DI and a microphone, and Bugg's vocal was recorded with just one microphone, but Bush separated the verse and chorus vocals out to two different tracks, because he wanted to treat each differently, meaning that the session contains four audio tracks in total. The mixer described how and why he added five vocal effects tracks, an acoustic guitar sub-track, two vocal sub-tracks, and a VCA track, expanding the session to 13 tracks before the mix was printed.
"This was one of those tracks that featured an amazing performance, but there were some challenges in the way it was captured because when a player moves during the recording, microphones that were perfectly aligned are suddenly out of place. There also was a lot of bleed between the vocals and the acoustic guitar, and all this meant that there were phase issues with the guitar the moment the vocals came in. This often is the case when you're recording a vocal and an acoustic guitar at the same time. The guitar will sound one way when there's no vocal and completely different when the vocals come in, which is not something that happens when you're simply in the room listening. In the case of this song, the guitar sound really changed depending on whether Jake sang or not, and also depending on how loud he was singing. So a lot of my mixing work involved working around the issues that were inherent in this session. Mostly I tried to make sure that the guitar sound changes as little as possible during the course of the song.
"I grouped the two acoustic guitar tracks to one track and then warmed the sound up by using the Waves Abbey Road REDD 37 analogue emulation plug-in, adding a bit of harmonic detail in the top end. I then addressed the issue with the vocal affecting the acoustic guitar sound by creating a side-chain between the guitar group track and the first 'Vox' sub-track, through which I sent the dry vocals. The Waves C6 multi-band compressor is on the send of the 'Vox' sub-track and side-chained to the insert of the acoustic guitar sub-track, and so whenever the vocal kicks in, the C6 compresses the frequencies on the acoustic that are most responsible for the phasing. So the acoustic sounds full and rich when it's played by itself, but have certain frequencies ducked the moment the vocal comes in. At the same time I tried to make sure that the acoustic guitar is as consistent as possible throughout the track. The other effect on the guitar is the same Bricasti reverb that I also used on the vocals.
"Both the verse and the chorus vocals had the UAD LA2A on the insert, mostly for tone. The LA2A does a great job of making whatever you put through it bigger and more important, while something like the 1176 will make things sound more aggressive. I then also had the McDSP AE400 on the insert of the chorus vocal, because he was singing louder and there were some honking frequencies that were mainly responsible for the phasing problems with the guitar. I take some of these frequencies out on the chorus vocal with the AE400, and again using the AE400 on the 'Vox' dry vocals group channel. I like the AE400 because it's active, it responds to how prominent certain frequencies are. Both verse and chorus vocal tracks have sends to a similar set of vocal effect tracks as I used on 'There's A Beast', consisting of a slap delay from the UAD ATR102, a plate reverb from the UAD EMT140, the doubler effect with the Waves Doubler, SSL EQ and Waves S1 Imager, the spring reverb track with the Waves De-Esser, Sound Toys Echo Boy delay, Avid EQ, PCP Distro going to my Audio Kitchen spring reverb and the D3 gate, and finally the Bricasti M7. I often use similar effects on the vocals of each song when mixing a record that's meant to work as a whole and that needs a consistent sound.
"The five vocal effects tracks go to the 'Vox FX' sub-group track, and the guitar sub-track, the 'Vox' track with the two dry vocal tracks combined and the 'Vox FX' track all go to the VCA Master fader. For the final mix, the guitar and two vocal sub-tracks went to my Shadow Hills Equinox summing mixer, and then again through the Manley Massive Passive EQ, the Shadow Hills Mastering EQ and the Cranesong HEDD, and came back in on the 'Mix Print' track, on which I had the UAD HLF 3C, which is a Pultec copy and with which I applied a high- and low-pass filter to get the track to sound more vintage. After that there's a Slate Digital Virtual Tape Machine plug-in, again to get more of that old-school sound. There's something about the sound of a half-inch tape machine, or about an emulation of it, that glues everything together and really makes things sound finished.”