Inside Track | Florence and the Machine: ‘You’ve Got The Love’
Technique : Recording / MixingFlying the flag for a distinctively British sound, and scoring some huge hits in the process, Cenzo Townshend is one of the UK's hottest mixers. He gives us the inside track on Florence and the Machine's hit 'You've Got The Love'.
Paul Tingen
Cenzo Townshend in his mixing room at Metropolis. Note the scribble‑strip curtains!
Many of the American mixers featured in SOS's Inside Track series have explained that they pay little attention to rough mixes, preferring to start from a clean slate. Cenzo Townshend, one of the UK's leading mix engineers, takes the opposite approach. "I don't subscribe to the view of not wanting to hear the rough mix. I am here to represent what the band and the producer want from a project, not to completely start from scratch and do my own thing. It is not my record. If people want me to take a song in a different direction and want my input, great, but most of the time my job consists of giving the best possible rendition of the vision that they have. So I spend a lot of time listening to their rough mix, or the monitor mix that they like the best, to hear where they're at.
"Because recording budgets are so small now, a lot of records are recorded in private rooms, often with laptops. In these circumstances people haven't really been able to hear what they're doing and haven't been able to properly balance things; I therefore spend a lot of time cleaning sessions up, organising and editing. But they do have a vision, so my job is to work as closely as possible with the artist and producer to find out what they like about their own mixes, what they have been able to achieve and what not. Many people love their monitor mixes, but there will be things that they felt they never got right.”
Alone With The Music
Townshend's mix room is based around an SSL G‑series console, with KRK 9000 and Yamaha NS10 monitors.
Townshend must be getting a lot of things right, judging from the long string of hit records he has worked on. The list of artists he has helped includes Bloc Party, Interpol, Babyshambles, Editors, Jamie T, Graham Coxon, Friendly Fires, New Order, Klaxons, Kaiser Chiefs, Snow Patrol, U2, Franz Ferdinand and many more. Cenzo Townshend (his first name is an abbreviation of Vincenzo — he has a half‑Italian father — and the 'C' is pronounced 'cz' as in Czech Republic) began his career in the late '80s at Trident Studio in London, working his way up the usual studio greasy pole, starting as a tea boy and then becoming assistant engineer for legendary studio professionals such as Flood, Alan Moulder and Mark 'Spike' Stent.
Townshend subsequently went independent, working for eight years with producer Ian Broudie before joining producer Stephen Street at The Bunker, located at Olympic Studios in London. Having spent the best part of two decades mainly engineering and occasionally producing, the last five years have been devoted 95 percent to mixing. "Obviously, there is a larger need for mixers these days than there is for recording engineers. But it also suits my personality. I like to be left alone to get on with things. Also, mixing is all about attention to detail and balance, you have to have the mind-set to be able to concentrate and go through all the mundane stuff before it comes together. A mixer is basically a balancing engineer with an aptitude for detail and an interest in sound.”
Townshend's achievements were recognised last year when he received the Music Producers Guild's award for Best Mix Engineer, and his prominence means that he mixes "a lot of singles, especially for radio. This is slightly different than mixing for an album. It means that everything has to sound a little bit more exciting and bombastic and has to jump out a little bit more to compete with the other things that are on the radio. Something that I am also aware of, but can't do much about, is the frightening fact that the same song will not only sound different on different radio stations, but sometimes will sound different on different days or even times on the same radio station! XFM went through a period of sounding harsh and brash, but now sounds better than Radio 1. Radio 2, for some reason, sounds fuller and bigger than Radio 1. Meanwhile, Radio 1 seems to sound better at night-time than in the day. In LA the radio always sounds brilliant. I don't know why this is, it's probably to do with the different compressors they are using, but no‑one seems willing to volunteer any information.”
After a long residency at Olympic Studios, Townshend moved to Studio B at Metropolis in February 2009, following Olympic's closure, something that clearly still rankles with him. "It was due to the whims of EMI, and it was very unfortunate, because it was one of the last great recording studios in the world. Amazing records were made there, and I feel that we lost a lot of the heritage of British music. With that, I think we're losing the sound of British music. It's all being watered down, and I'm concerned that we'll be completely overrun by American music and everything will sound the same. So I'm a great advocate of keeping that British sound going.”
Machine Made
A small selection of Cenzo Townshend's vast array of outboard, including several items used on his mix of 'You've Got The Love'. From top: Manley Vari‑mu compressor, Empirical Labs Distressor compressors (x4) and Fatso tape emulator, Summit EQP 200B equaliser, SSL XLogic compressor, NTI EQ, Smart C2 compressor, Chandler/EMI TG1 limiter and EAR 660 compressor.
One of Townshend's most high‑profile projects recently was Lungs, the debut album from Florence and the Machine, most of which he mixed. It was released in the middle of 2009 and is still a big seller, recently hitting the top spot in the UK charts for the first time. Townshend mixes included the last three singles, 'Rabbit Heart (Raise It Up)', 'Drumming Song' and the top 10 hit 'You've Got The Love'.
The last is unusual for the band, in that it's a cover of a song by The Source (featuring Candi Staton), first released in 1986 and a UK hit in 1991. Florence and the Machine's cover was initially a 'B' side, but following widespread critical praise, Universal Island decided to re‑release it as a single in its own right. For this reason, Townshend ended up mixing the track twice.
"It's a song that she had been singing live for a while, and everybody really liked it. After three successful singles, the record company was looking to release it as a single, and she came in here and did about five or six new vocal takes. She was fantastic. I recorded her with a Neumann valve 47, going through one of my Neve preamps, then into an 1176 and via my Lavry Gold A‑D converter into Pro Tools. We then sat together and decided which bits to use. We didn't do much comping, and only replaced the first two verses and the first chorus. I think from the second chorus onwards we kept the original vocal. I then basically remixed the whole track from scratch, though the only real differences in the single version are the new vocal, and the bass and drums are a bit harder and the bottom end a bit heavier.”
Making Preparations
Townshend's upstairs preparation and post‑production room, featuring the SSL Duality, a Pro Tools HD3 system and several monitors, comes into its own the moment a project arrives for mixing. "The first thing that happens is that Neil Comber, my assistant engineer, will prepare the session for me, cleaning it up and laying it out in the way I like. Sometimes the session will come in as a Logic file, and Neil will have to build a Pro Tools Session from it, which can be quite laborious. We have to check every edit as we transfer things to Pro Tools. Sometimes there are 120 tracks to go through! During mixing later on, I'll have the Logic session on a separate computer somewhere for me to refer to and to see if they used any particular kinds of plug‑ins, panning, balances and so on. Even when we get a Pro Tools file, we often receive just a bunch of files without having any idea what they are. Tracks may just be called 'Audio 1', 'Audio 2' and so on, and we'll have to work out what's what. If sessions have been FTP'ed we also need to check that we have the right amount of tracks, and so on. Sometimes the sessions are so big that they won't play. Pro Tools won't play more than 96 tracks at 96kHz, and to compensate people may have been bouncing and so on, and then we have to deal with that.
"Quite often, Neil will also add five or six snare-drum samples and three to four kick‑drum samples and kick-drum ambiences, and he'll time those to make sure they're accurate for each hit. We'll talk beforehand about what samples to add. Basically, they are there to beef up the existing kick and snare, if necessary. I don't replace kick and snare drums, and probably 50 percent of the time I'll decide that the kick and snare I have are so good that I don't need to add anything. The reason he adds them at this preparatory stage is that I don't have the time to start adding samples when mixing, because we only have one day to mix a song. So Neil will add them in case we need them. All in all, Neil usually has three hours of work to do before I can actually begin the mix.
"The samples I use are usually things I've recorded over the years. But I'm not purist about it. If I hear a good sound somewhere, whether from a library or from something else, I'll borrow it. I have quite a few sample libraries, but you have to audition the sounds, and I don't have the time to listen to hundreds of samples. Neil uses the Drumagog plug-in to trigger the samples he adds, but you still have to go through each hit and move it in time to make sure it's 100 percent correct and the phase is correct. You look at the waveforms, but you also need to use your ears. Time‑alignment software is very time‑saving, but the drawback remains that it's not as good as doing it manually. I also use the Aptrigga2 plug‑in, which is great for things like tambourines, because it's a little random, and you don't want to have tambourines bang on time. Sometimes we spend as much time trying to slightly offset things to make them sound more natural as trying to get them to sound exactly in time.”
'You've Got The Love'
Written by Anthony B Stephens
Produced by Charlie Hugall
"'You've Got The Love' came in as a Pro Tools file with some of the effects already present. Her vocals were slightly distorted with a Fairchild plug‑in. It's not a very big Session by modern standards, though it has many rhythm tracks. The first thing I do when I get the Session from Neil is to spread it out over the board, and then I'll do my best to get the song as close as I can to the monitor or rough mix — within reason, I'm not going to obsess over it. I'll reference the band and/or producer's favourite monitor mix to get a balance, and while I'm doing that I'm getting to know all the parts that make up the song and getting a sense of what I'm trying to achieve with the song. In the case of 'You've Got The Love', the tricky aspect of the mix was the interplay between the guitars and the harp. The harp plays a huge part, melodically and rhythmically, and to get the harp and guitars to bounce off each other in the right way needed quite a few plug-ins and limiters.
"Having got to know the song in an hour or so, I start moving things around and begin to find the right compressors and EQs, for individual tracks and also for the stereo mix. I'll then get the vibe of the bass and the drums working, and may fire things through the live room here in Metropolis to record various ambiences for the kick, snare and toms. We have various speaker setups with ribbon mics to pick up the ambience. Among other things, I have an Auratone speaker on top of a snare drum, and I'll send the signal through the Auratone, which hits the drums, and this is picked up by two Coles ribbon mics. The room is also great for re‑amping guitar to get some guitar reverb, so I may spend a little time firing guitars, or the vocal, through the live room as well.
"Because I initially try to get the balance back to where it was originally, I don't tend to solo too much. But I will try to get the drums to sit well with the bass first, so I know what the rhythm is doing, and then add the vocal into that, and then the other instruments. I'm also constantly checking whether everything is in phase. Even if you think everything is OK, you may be halfway through mixing the song, and then you put your bass in and suddenly it's not sitting well with the kick. So you may have to flick the phase on the bass to make sure it has the same polarity as the kick drum. I tend to do that on the desk. There are various Trim plug‑ins on the Session, and they generally are for phase as well. Also, although there are many drum tracks, in the final mix it doesn't sound like you are listening to a huge rhythm track. That's very important.”
Drums: Audio Ease Altiverb, Waves Audiotrack, Massey CT4, Tapehead & EQ, EMI TG1 (hardware) & EQ (plug‑in), Sound Toys Echoboy, Digirack Expander/Gate & EQ, Eventide H3000, desk EQ & compression, Empirical Labs Fatso, Bomb Factory Sansamp PSA1. Sound Toys' Echoboy is one of Townshend's favourite plug‑ins, used here for a short delay on a snare track.
Tambourine: "There are two tambourine tracks at the top of the Session. There originally was one played track, which I copied, so I have two tracks with the same audio and different treatments. One of them I put through some guitar amplifier modelling, the Altiverb plug‑in on a 'Fender Superstring' setting, then a Waves Audiotrack plug-in for compression and EQ, and then a Massey compressor. Massey is one of my favourite plug‑in makers. The other track also goes through an Altiverb set to a room sound called 'Cello Studios', again Waves Audiotrack and then Massey Tapehead. Just below the tambourines is a snare track, put there to make sure the tambourines are in time. Normally, the snare tracks are further down. The 'stone room' track [Townshend is referring to the track naming in the Pro Tools Session, which there is not space to reproduce here!] is the Auratone sitting on a snare drum in my live room.”
Kick: "There are nine kick tracks. Neil added 'Polekick', which is a kick ambience sample, and 'Meat' and 'Limrock', which are kick samples. 'Polekick' is a trailing thing that adds a little bit of space around the kick sound, and you can EQ it to give a feeling of more low end without there being too much attack, so it doesn't flam with the real kick drum. Immediately above the real kick drum are two kick comp tracks, which trigger an Expander/Gate [plug‑in] that is advanced in time. By advancing the track and sending it to a side‑chain of the gate, it opens faster and without any strange artifacts in the sound. There are two more kick mics, called 'BD23' and 'NS10', and [tracks] two master faders for all the kick tracks. All the kick tracks go up and down in level in different places, so the balance between them varies. I do this in the computer. On the desk I'll have two kick channels: the real kick on channel one, and the samples on channel two. In terms of treatments, the real kick and the 'Punchy Room Kick' both have a Trim and a Massey three‑band EQ. On the desk I'd have EQ'ed the kicks, and they would have been bussed to an EMI TG1 compressor.”
Snare: "There are seven snare tracks, with a sample that's added by Neil, and some snare samples that remained from the Session as I received it. I didn't use the latter. Again, I created a balance of snare sounds that changes over the course of the track. In terms of plug‑ins, a number of the snare tracks, again, have a Trim plug-in, and on the 'snare comp' track there's also an Echoboy with 90ms delay and a seven‑band Digidesign EQ, pulling out 18dB at 500Hz. There was a peak there that I didn't like. Like the kicks, the snares would come up on two faders on the desk, on which I use SSL channel compression and, again, the EMI TG1. For ambience, I would have added the stone room, via the Auratone into the Coles, coming back into Pro Tools via a pair of Neve mic preamps. There's also some Eventide H3000 ambience on the snare, set to a room with very short reflections.”
The rest: "There's nothing on the hi‑hats, except desk compression. The toms have an EMI EQ plug‑in over the master fader. The overheads go through an Empirical Labs Fatso that's inserted on a channel on the desk, no plug‑ins. I have a Massey plug‑in compressor on the cymbals. After that is a room track that I duplicated, with one side being sent through a Sansamp PSA1 plug‑in and an Expander/Gate. I probably panned these two room tracks hard left and right.” Massey's CT4 compressor was used on the cymbal track.
Bass: Waves Q4, Digidesign Recti‑Fi & Lo‑Fi, re‑amping, EAR 660, Pultec EQ.Digidesign's Recti‑Fi plug‑in helped to shape the bass sound.
"There's only one bass track, which has a Waves Q4 EQ taking out a hump at 209Hz, because the bass was boomy, and the Lo‑Fi to add distortion, plus Recti‑Fi to twist the sound a bit more. If you have something incredibly noisy you can take the top off and add some distortion to the mids with the Lo‑Fi, which works great. The Lo‑Fi is a great plug‑in, and it's not just for 'lo‑fying' stuff. The Recti‑fi is a stranger beast, with which you can do some ridiculous things, like robotic distortion and ring modulator‑type things. The bass came up on two channels on the desk, and one of them I sent to my Hiwatt amp, for a warmer sound and some distortion. I have a Little Labs phase tool that interfaces the Hiwatt with the Sequis speaker simulator, which is amazing. I've gone through dozens of speaker simulators, and the Sequis is the best. The other bass channel went through an EAR 660 compressor and a Pultec EQ. So I had two completely different bass sounds on the desk, which I blended throughout the song.”
Guitars: Digidesign Lo‑Fi, Waves Q4, Neve 33609, Roland Dimension D, Manley Vari‑Mu, Empirical Labs Distressor, Eventide H3000.Townshend used Digidesign's Lo‑Fi to darken the acoustic guitar sound, and Waves' Q4 to notch out problem frequencies.
"There's one acoustic guitar track, on which I had the Lo‑Fi to make it less shiny. I'm not too keen on overly sparkly acoustic guitars in a mix, I prefer a more Kinks‑like, old‑fashioned sound. There's also a Waves Q4 notching something out. Outboard was the Neve 33609 and Dimension D chorus.
"The two electric guitar tracks are the same part recorded with two mics, and had no plug‑ins. On the desk, I sent them to a Manley Vari‑Mu, and I would have set up some parallel compression channels with two Distressors. I also had the same H3000 ambience on the guitars that I had on the snare. I have no problems sending five different things to the same ambience. You get more of a unity of sound that way. I don't like using too many different delays and reverbs, because things can end up sounding very separate and lacking in cohesion.”Harp & piano: Waves Renaissance Axxe, Renaissance Bass, SSL G‑series compressor & Q4, Sound Toys Echoboy, Massey 2007, Neve 33609, Bricasti M7.
"The harp had the Renaissance Axxe compressor, an Echoboy doing 16th‑note delays, and the 2007 Massey limiter. Outboard was the Neve 33609 and a Bricasti M7 on a plate setting. The Bricasti is £2000 worth of reverb, and while people may think that's mad, it does a few things incredibly well. The processing power in one of these boxes is 10 times that of a Mac and the calculations it can make are frightening. It does sound like an analogue reverb and it's one of the best reverb boxes you can get.
"There are two piano tracks, low and high, and on the low one I had the Renaissance Bass plug‑in to accentuate the low end, like a sub‑harmonic synth, and the Waves SSL G‑series master bus compressor plug‑in. On the high piano I have the Q4 to notch out at 750Hz and, again, the SSL compressor. Pop pianos require quite a bit of work to sit well in a track and for people to be able to hear them.”
Vocals: Massenburg EQ, Waves C4, Q4 & De‑esser, Bomb Factory Fairchild 660, Teletronix LA2A, Pye compressor, Dbx 902, Pultec EQ, Bricasti M7, Ibanez delay, Empirical Labs Distressor, Sound Toys Pitchblender & Echoboy, Massey Tapehead.Among the many plug‑ins used on Florence Welch's vocals were Massey's Tapehead tape simulator and Waves' Q4 equaliser. Sound Toys' Pitchblender was used as a vocal effect.
"The main lead vocal track is called 'LV Comp', with a track with some chorus doubling just below it. I treated both tracks the same way. I have a Massenburg Designworks EQ plug‑in on them, taking out 6dB at 1.5k and boosting at 5k. Then there's a Waves C4 multi‑band compressor, and a Fairchild plug‑in on the master fader [of the vocal group] which adds a bit of distortion, to make it sound more crunchy. On the desk I will have used an LA2A or a Pye compressor, a 902 de‑esser, and Pultec EQ. I added some reverb with the Bricasti M7, on the same plate setting as the harp. I also added various analogue delays using my Ibanez rackmounted guitar delay. In addition, I will have set up some parallel compression channels for the vocals with my Distressors. Under the master track are two vocal effect channels: one had the Pitchblender, which is an automated harmonising delay, and the other is an Echoboy with an automated eighth‑beat triplet delay. The next three tracks are all backing vocals, with just a Q4 EQ on the master. Across the board I would have had two Distressors on them and they would have gone to the same effects as the main vocal. Finally, there are two end backing vocal tracks, which have a Waves de‑esser and a Tapehead. I used the Waves de‑esser because I didn't have any more 902s. The Waves de‑esser is fairly broad and works fine for backing vocals.”
Strings: Waves C4, NTI EQ, Bricasti M7, TC Electronic D•Two.Waves' C4 multi‑band compressor was used to darken the string-machine sound.
"There are four stereo string tracks from a string machine. I have a C4 across the master fader to get some of the high end out. Outboard was the NTI EQ, which has this great thing called Air band that adds space in the high frequencies. The strings would also have gone to the Bricasti plate, and I would have used my TC Electronic D•Two stereo spatial delay, for a little bit of swirl. You'll find that the more acoustic instruments go to plates and lusher reverbs, while the more electric‑sounding stuff, like electric guitars and drums, tend to go to something a little harder and shorter.”
Final Mixdown
"The session was in 96k, and I recorded the stereo mix on another Pro Tools rig, again at 96k, going via my Lavry Gold A‑D converter, and through Analogue Tube AT101, which is a copy of the classic Fairchild 670 stereo limiter/compressor. It's made by Simon Saywood here at Metropolis, costs close to £14,000, and it's amazing [see www.analoguetube.com]. It's huge‑sounding and has valves at every stage. It's taken him four years to develop, and it sounds fantastic. I had it on the inserts of the master fader on the desk. I've used it on all sorts of records — U2, Editors, Detroit Social Club, all of Florence, and so on. Do I take the MP3 and other user formats into account when mixing? Well, I do make MP3s of my mixes, to send to the artists and producers, so I make sure it sounds good in MP3, but I don't mix for MP3. Otherwise it'd be useless to have a rig running at 96k using ridiculously expensive converters and clock sources. I'll always mix in whatever the highest possible quality is.” .
The British Sound & Radio Mixes
In last month's Inside Track feature, Cenzo Townshend's former mentor 'Spike' Stent noted differences between American and British mixes, in his opinion mostly to do with the way the low end is handled. Townshend picks up on other contrasts: "I think British music is edgier and more natural. I have nothing against American music, some of it sounds amazing, but a lot of it does get over‑polished and sanitised. I think English music as a whole has more aggression in the mid‑range, especially in the guitars. The guitars tend to be softer and more perfect and polished in American music. With a few exceptions, American rock guitars tend to sound very similar. Their bass sounds are fuller and rounder and cleaner, whereas I like bass sounds to be quite dirty; that gives more attitude, depending on the track, of course.
"In general, American musicians also tend to be accomplished players, often playing in more than one band to craft what they do over a long period. Whereas in Britain, being in a band usually is about four or five people making a great noise, the whole being greater than the sum of the parts. Sadly, commercial music is no longer about experimentation or even approximating the experience of hearing a band in a room play. Instead, commercial music is about trying to make a song sound as loud as possible on a radio, and people then downloading it onto their MP3 players. So why do I choose to focus on mixing for radio? Well, I love the radio. I have always listened to it and used be a DJ before I started working in studios. I like pop, and I like hearing it on the radio. I still listen to radio all the time; it's not just for work.”
The Best Of Both Worlds: Cenzo Townshend's Mix Room
The vocal overdubs and mix for 'You've Got The Love' took place in Cenzo Townshend's room at Metropolis Studio B, which features a 72‑channel 4000 G‑series SSL. It is apparently, one of the few piece of kit there that are not the mixer's. According to Townshend's web site, he's known for "his love of vintage recording equipment, and his extensive knowledge and collection of classic valve amps, effects and pedals”. He's also got several Pro Tools H3 rigs loaded to the brim with plug‑ins. Monitoring is provided by KRK 9000B and Yamaha NS10 nearfields with Bryston amplifiers, but also by a Pure portable digital radio.
"I like the SSL for many reasons. I like the general sound of it, I like the EQ, and I like the sound it creates when it busses everything together. I also have an SSL Duality in the prep and post‑production room upstairs, which is fantastic — a very interesting step forward from the very large consoles to something more manageable. I'm in discussions with SSL on how to improve that desk. For now I'm still using the G‑series, which is my favourite SSL, and which obviously also functions as an interface for all my outboard. Concerning monitoring, I'm without a doubt strongly influenced in my choice by Spike. I used KRK E8s for a while, but the reliability wasn't good, and since I've put the 9000s up again, there's no going back. I tend to mix very quietly on smaller speakers, and I use my Pure Evoke 3 portable digital radio a lot, which works very well for editing and balancing. I've also recently been listening to Focal Twins, which are very good. I can easily switch between 15 different speakers. I couldn't just mix on NS10s or KRK 9000s; it would be too tiring and you get used to them too much.
"With regard to outboard, I began my career in the late '80s, and while I could have been totally into digital, my first engineering experiences were on old vintage Neve desks and things, and I also had great experiences with guitar pedals, which I love to this day. I still find analogue more pleasing to the ear. I use digital for a number of things, but I just love analogue. With regard to plug‑ins versus outboard, I don't think there's a contest. They are two different things doing different jobs. It's fantastic that I have the opportunity to use each for what it's good at. There are some Waves EQs that are amazing, and that can be far more surgical than analogue equalisers for removing frequencies that are getting in the way in a mix. For instance, when mixing Snow Patrol, I may have 12 of the same guitars that all have the same ringing frequencies, and I'd never be able to get rid of them using analogue EQs.
"On the hardware side, the EAR 660 is a fantastic compressor, which is particularly great on bass. I also like the Pultec and the Summit [EQ]s a lot, and the NTI has an upper range that creates wonderfully smooth air. I like using parallel compression, for which I use my [Empirical Labs] Distressors, I have seven of them! I also like the old Neve 33609 stereo compressor, and the SSL and [Smart] C2 compressors, which I prefer to use on acoustic guitars. The other thing that outboard is good for is to create more of an out‑of‑the‑box sound. Many projects are recorded at home in people's laptops, and they probably only have two or three good mics and one mic pre, and it all goes into the same box. So when you get 120 tracks that have been recorded like that, it's nice to give the different tracks different footprints by putting them through different transformers and valves. It adds a bit more depth to the sound.
"By contrast, plug‑ins allow me to give more width to the mix. The width you can get now with Waves and [Sound Toys] Echoboy is fantastic, and it really opens up and pans out the stereo soundfield. Moreover, there are now plug‑ins like the EMI Abbey Road Brilliant EQ, which have a very musical sound and work like good outboard. Whatever instrument you put through the EMI, somehow it manages to highlight some nice frequencies, which is amazing, because it only has three settings.” Published in SOS March 2010
No comments:
Post a Comment