Mumford & Sons have dragged acoustic music out of the folk clubs and straight to the top of the charts, with a sound sculpted by producer Markus Dravs and mix engineer Ruadhri Cushnan.
With a number one album and two Grammy nominations, Mumford & Sons have become the most successful of the new breed of mainstream 'indie folk' bands. The chart prowess of their debut disc Sigh No More owes much to the golden touch of producer Markus Dravs, who has also scored big hits with Coldplay and Arcade Fire in recent years. The album was mixed by Ruadhri Cushnan, who has worked with quite a few acoustic acts over the years, ranging from Duncan Sheik to Celtus to Johnny Flynn's Been Listening (2010).
Born in Belfast, Cushnan already had aspirations to work in recording studios as a teenager, and attended an engineering course at the Sound Training Centre in Dublin in the late '80s. He moved to London in 1990 and started to work as a tape‑op at Metropolis Studios in 1991, quickly working his way up to fully fledged engineer. During the second half of the '90s, he worked extensively with producer Rupert Hine (known for his work with Tina Turner, Howard Jones and Rush) as well as with Cameron McVey (Massive Attack, Neneh Cherry, All Saints, Sugababes). In the first years of the new century, Cushnan was predominantly busy as a writer/programmer, working from a programming room at AIR Lyndhurst in London. He co‑wrote, amongst other things, a few cuts on George Michael's album Patience (2004), and continued to work with Cameron McVey on a number of projects, including Virgin Souls, Siobhan Donaghy and Mutya Buena (both ex‑Sugababes).
By 2007, Cushnan says, he "decided to get back to engineering and concentrate more on mixing. I moved out of AIR and was very lucky to find another room, part of the Manatee Audio complex of rooms in North London, just five minutes from home, which I turned into a programming and mixing suite that I called Rhubarb. I had just finished working on the Guillemots' second album and somewhere during the course of that project I picked Rhubarb up as a nickname. When I set up in my new room the walls were painted a very distinct shade of red so I started referring to the room as Rhubarb, and the name seems to have stuck.”
Despite having taken more than half a decade out from engineering and mixing, Cushnan has since established himself among the top flight of mixers in London. During the last three years he's worked with the likes of the Maccabees, Lauren Pritchard, Corinne Bailey Rae, Seth Lakeman and Unkle, and mixed almost all of KT Tunstall's Tiger Suit (2010), and, of course, Sigh No More. In many ways, it's the breadth of skills the Irishman has acquired that has made possible his remarkably fast run up the greasy pole: "By 2001, most of the sessions I was doing had moved over to the digital world, and the lines between engineering and programming were becoming blurred. It was increasingly expected that as an engineer you would also be able to program up some beats, drop a bass line, or cut up and do something interesting with an arrangement. As a result, my programming skills got much better. I was mostly using Logic at the time, I think I got fastest on version 5 and it was arguably better than Pro Tools for that kind of combined programming/recording session that I was doing. Although I have to say that since then I have got more and more frustrated with Logic: every time there is an update, lots of little things change, to the point where you have to re‑learn how to use it, which slows things down. It's like the stool is kicked from underneath you every time there's a new version. This is unlike with Pro Tools, where the layout and the structure of the key commands and such like has been the same since year dot. I still do use Logic for programming, but it just seems slower than it used to be.”
Today Cushnan conducts most of his mix sessions from Rhubarb, occasionally venturing out to a commercial studio to run a mix through a desk, as was the case with Tiger Suit. Mumford & Sons' Sigh No More was mixed entirely 'in the box' at Rhubarb, which doesn't have a main mixing desk as such, only two small consoles, a Midas Venice and a Neve 542, which are mostly used for routing and monitoring. The heart of the two‑room studio is formed by state‑of‑the‑art Pro Tools and Logic systems, with a small selection of choice analogue gear, such as Thermionic Culture's Culture Vulture and Phoenix, Roland 501 Space Echo, Neve 8803 and 1081 EQs, Alan Smart C1 compressor, rackmounted SSL channels, and Summit TLA100A compressor. Other crucial bits of hardware include Cushnan's favoured Apogee Rosetta 200 A‑D converters, and Adam P33a and Yamaha NS10 monitors, plus a Panasonic boom-box.
Cushnan comments, "I have these pieces of outboard because I can't get the equivalent of them in the box. I don't own racks and racks of outboard, but I get very good use out of everything that I have. When I see these pictures of big rooms with racks and racks of gear, in the end I think they'd be more of a distraction to me, because you can get lost in trying to decide what the next piece of kit is that you're going to use. I'm not really a big fan of digital EQs, other than the Sony Oxford [now Sonnox] plug‑in, and I suppose the Massey VT3 and Joe Meek are also quite good, and useful for making something poke through the mix. I use my outboard EQs, Thermionic stuff and other outboard to add character and colour. The Neve 542 has a very nice crispy top end, and I often use the Summit on lead vocals. But ultimately everything ends up as two channels coming out of Pro Tools.
"The Neve 542 is in the other room at the moment, I use it mainly to pass audio through, like the Midas. I'm setting up a Logic system next door, which is designed to be more of a programming and occasional recording room. The Midas is a fully balanced little mixer, and it's really clean, with a low signal‑to‑noise ratio. The mic amps are good, and so is the EQ. It mostly just passes audio and is a really trustworthy and compact console. With regards to the monitors, the NS10s are what they are. I've used them for so long, I know exactly what they do. I really like the Adams. I tried them for mixing the Mumford & Sons album, and that worked so well that I've stuck with them. I usually have the ghettoblaster off to my left. It's all about having a change of perspective in your room. After I've been mixing a track for two days, I want to have things come at me from a different angle. It may allow me to hear things that I have not paid attention to before.
"With regards to mixing in or out of the box, A&R people are aware of the difference, but ultimately they don't care, provided you can deliver the mix they want. The only thing that makes a difference for them is the price. Mixing in the box can be very powerful, and I certainly enjoy it, but it is also more difficult and time‑consuming. For me it's easier to mix expressively and quickly on the board, but hiring a studio tends to work out more expensive than me working from my own room. So I usually present the record company with two options: I can mix a song in the box in my place for amount A and take two days, or I can do it for a larger amount on a desk in a professional studio, and only take one day. I mixed the KT Tunstall album in about three weeks, but it would have taken me six weeks in the box. Of course, there are other considerations. Working in the box makes it possible to move very quickly from one mix to another, which is a great way to keep perspective. I love working on two mixes at the same time. And record companies expect you to be able to tweak a mix a week after you've delivered it. If you mixed on the board, you're stuck, unless you printed an exhaustive and thorough array of stems, as I did with the KT Tunstall album.”
According to Cushnan, all the sessions he receives these days are digital, with "perhaps eight out of ten sessions coming in on Pro Tools. The rest is Logic. I wouldn't want to mix in Logic, so I always ask for the Logic session to contain audio that I can then import into Pro Tools.” The mixer adds that organising his Pro Tools sessions is key. "I first make sure all tracks are labelled and put in the order that I like to have them. I'll then organise my subgroups and effect tracks. Once I've organised the Pro Tools session, if I'm working on a desk I'll assign outputs and lay things out on the desk, and I'll write my board strip. If I'm mixing at Rhubarb, I'll make sure I have the outboard gear routed the way I want it.
"The way I like to lay out my Pro Tools session harks back to my days as an engineer working on a console. I like to have the drums at the top, then the bass, then the guitars, then the vocal, which I'd have at the centre of the console, and then keyboards and backing vocals, and other things, and finally effect tracks. However, these days I end up having my vocals at the bottom of the session, just above the effect returns. I also use a lot of subgroups, and in many cases mix from a Pro Tools layout page on which you can just see the subgroups and I'm only dealing with eight to ten faders at the end. I love getting well‑organised sessions, because otherwise it can cost me half a day to work out what's going on and label and layout things properly. Rather than having buses 31‑32 or 110‑112, I like to have names on them, because that really helps to stay on track with what's happening. Otherwise, if you come back to a session a week later, you'll see this ridiculous list of numbers that won't mean anything to you. Also, Pro Tools sessions move from studio to studio, and it's important that your track makes sense when someone else looks at it. It's like in the days when tape was sent around from studio to studio, and you had to make sure that the track sheet and track notes were clear. It's no more than being professional to make sure that your track is self‑explanatory.
"My mix preparation also usually includes switching off most of the plug-ins that may come with the session. Most sessions done at professional studios are plug‑in free anyway, because the tracking engineer would have run the session through his desk, and would have simply run off monitor mixes from the board, rather than done rough mixes in Pro Tools. In the case of Mumford & Sons, the sessions at Eastcote were recorded via an MCI board. Sessions that come from private setups tend to have plug‑ins on them, but that's not really an issue, and the tracking engineer will have cleared out most stuff. Of course, plug‑ins that give a special effect or character to a part will usually remain. Finally, in the process of running through the mix and organising it, I will have been listening to the song the whole time, I'll have familiarised myself with it and will have built a picture in my head of what I'm dealing with and how I want to tackle the mix.”
Sigh No More came in as a collection of Pro Tools sessions, recorded by engineer François Chevallier at Eastcote Studios in West London. Cushnan got the job via his connection with Markus Dravs, which goes back to their days together at Metropolis. In 2009, Dravs and Cushnan had respectively produced and mixed the Maccabees' Wall Of Arms album, which was a Mumford & Sons favourite. The band had also chosen Dravs because they're big fans of Arcade Fire.
"The Mumford & Sons sessions were well‑organised and clearly laid out. Mumford & Sons are a very straightforward, very organic band, and the album is very truthful. It's not over‑produced, it was really a case of Markus and François capturing the band's live energy on record. Personally I think that the band's folk connection is overstated. Yes, they have influences in folk, but I'm not sure I'd call them a folk band at all. It's the banjo that makes people pigeonhole them as folk, but they sound very contemporary to me. I don't speak on behalf of the band, and so Marcus Mumford may disagree with that, but they won a Grammy nomination for Best Rock Single, and we never talked about the whole folk‑versus‑rock thing during mixing. They are the way they are and they play the way they play, so let's just try to get that across and make sure that the excitement of their live performances translates into the recording. That was the most important thing. They managed to do that without many tricks and were confident enough to simply let the band's natural sound through.
"I think the album was recorded over two or three weeks, with the band playing live in the studio, and then later adding overdubs, including the drums. The drums are mostly a four‑to‑the‑floor bass drum and floor tom, with the occasional snare overdub. There's a stand‑up bass, the banjo, a piano and some other keyboards, and Marcus Mumford's vocals and acoustic guitar, plus loads of backing vocals, which are very important. One of the big talking points during the mix was the banjo, and how much it should feature. The band were adamant that the banjo is very critical and important to their sound, and trying to make it cut through and giving it the dynamic punch that it needed, without it completely overpowering everything and it becoming a banjo record, was a delicate balance. The other thing was that most Mumford & Sons songs have a very rich, gradual dynamic lift, and I needed to make sure that I kept some of that in reserve, so that we always had somewhere else to go right up until the end of each song and by the time I got to the final peak, I could really let rip. So being careful with the dynamics was very important.”
- Written by Marcus Mumford, Winston Marshall, Ben Lovett, Ted Dwane
- Produced by Markus Dravs
Ruadhri Cushnan: "As I mentioned before, this was a very well‑organised session, and one thing that was particularly helpful was that I had several microphone options for each instrument. There were two banjos, and the first banjo, for example, was recorded with three mics — a Royer ribbon, Telefunken Elam and a Neumann KM84 — and then DI'd, plus there was a mic for the amp room. So I had five tracks for that banjo, and four for the second banjo, with similar mics, and all these went to a banjo subgroup track. A good producer and a good engineer who have recorded in a good studio will do this, and give you a lot of options as a mixer. Of course, I don't expect to comp anything when the session comes to me, I expect that the source material that I'll be using in the mix is set. But I like having the options of choosing different sonics for each instrument, coming from a variety of mics.
"In terms of approaching a mix, in the case of Mumford & Sons, and also with the KT Tunstall album, I began with the lead vocals and acoustic guitar, because with both artists these two things really dictated the energy and dynamic of the song. It's a bit unusual, as I'll often start a mix with the drums, but particularly in the case of Mumford & Sons, the drums are secondary. After working on the lead vocal and acoustic guitar, I would have brought in the piano, then the banjo, and after that the double bass. I would have worked on these for a bit, getting them all to sound quite good by themselves, and then trying to find a balance between them, staying aware of the emotional heart of the song. I first focused on the emotion and atmosphere, and only after that would I have mixed in the drums, which add a lot of drive and energy and dynamic curve to each song. After that I'd add in other bits and pieces, in the case of 'Little Lion Man' a few keyboards, and the backing vocals would have come in last. One reason for this was that I started mixing when the backing vocals weren't recorded yet. Because of working in the box, I could work on getting four or five songs up to 85 percent of how I wanted them, and then I imported the backing vocals as they came in.”
Lead vocals: Sony Oxford EQ and compressor, Summit TLA100, Neve 1081, Roland RE501, Avid Reel Tape Saturation, Audio Ease Altiverb, Sound Toys Echoboy.
"There's only one main lead vocal track ['RldT'/'RleadcompT'], and we did a vocal fix at the end of the track during mixing. The lead vocal has the Sony Oxford EQ plug‑in and then the Sony Oxford compressor on it. The lead vocal also went outside of Pro Tools, where I sent it through my Summit TLA100 compressor and some Neve 1081 EQ. These things gave it some warmth and colour, and I printed that back in underneath the main vocal track [as 'LVprint']. I also sent the vocal to my Roland 501 for some spring reverb, and printed that underneath the Summit track ['LV spring']. The main vocal track and the Summit track both had a 'Tape' send, which went to an effect track with the Digidesign Reel Tape Saturation, below the Roland 501 track. I would have driven the Tape Saturation quite hard, with the simulated tape speed down to 7.5ips, to get a good degree of crunch, and add body, weight and a little grit. It fills the vocal out and helps it sit and command the track.
"These vocal tracks were then sent to the lead vocal master ['LVM'] subgroup track, on which I again have the Sony Oxford EQ and compression. The EQ and compression on the original track would be corrective, on the subgroup track it would be for bite and cut. It's an unusual thing [for me] to go back to a mix and analyse it, but in doing so I see that I use a lot of EQ and compression! The 'LVM' track also has three sends, called 'Club', 'Delay' and 'Plate'. 'Club' is a small room Lexicon 480 preset from Altiverb, 'Plate' is an Altiverb plate, and 'Delay' goes to the Echoboy effect track below. I normally set up at least a couple of reverbs when I mix a track, pretty straightforward stuff, really, with one reverb being a gentle, small room with maybe 1.2‑2s of reverb, the other a plate with some more length and a little bit more sparkle to it.
"In this case, both reverbs came from Altiverb, but sometimes I'll use [Avid's] D‑Verb, which can give a little bit more edge. Altiverb can be a bit soft, but that worked in the case of Mumford & Sons, because it was more a matter of putting them in a rich, natural space. You can really drive Altiverb hard, and you never get that ugly kind of digital top end. By the way, once again, you'll notice that I named the Tape send, rather than just having 'bus 31‑32' or something. You can also see that there are a lot of inserts on the vocals, and the session in general. Inserts would usually be EQ and/or compression, sends are more likely to be delays, reverbs or other additional effects.”
Acoustic guitar: Sony Oxford EQ and compressor, Bomb Factory 1176, Audio Ease Altiverb.
"The acoustic guitar was pretty straightforward. There was not a lot to do. The acoustic guitar was recorded with two mics, so there were two tracks, which went to a subgroup. The Sony Oxford EQ was pulling out some lower mids and enhancing the top end, and then there was a Sony Oxford compressor, which I really like at the end of a chain, because of the control it has. The 'Big Mic' track was recorded with a slightly more distant mic, and sounded a bit richer and fatter, and also had the 1176 and the Sony Oxford EQ. Finally I had some Sony Oxford compression on the subgroup track to cap it off and help the acoustic to sit. 'Cl' in the sends is the club reverb and 'P' stands for the plate.”
Piano: Bomb Factory 1176, Sony Oxford EQ, Avid Joemeek compressor, VC5 and Trim.
"There were several microphone options on the piano — an [AKG] C12, a pair of AKG 414s and a room mic. I ended up only using the pair of 414s, which sounded nice and bright and made the piano cut through. There are a whole bunch of plug‑ins on the piano track: '2' is a Digidesign 1176, 'O' the Sony Oxford EQ, the first 'J' is a Digidesign Joemeek compressor to give the piano some character and life, and then there is a Joemeek EQ, which is really good for getting parts to cut through, and 'T' is just a level trim. These plug‑ins didn't necessarily work all the time, I tend to automate my plug‑ins, and I might, for example, add a little 10k only in the final chorus to give it an extra nudge.”
"There were two banjo parts, and I mentioned earlier that I had quite a few different tracks to play with, recorded with mics like a Royer ribbon mic, a Telefunken Elam and a Neumann KM84, and then a DI from the banjo amp, plus the amp room mic. I used them all. By turning up or down individual tracks, I could affect the EQ, and for that reason all processing is done on the subgroup track. There are no plug‑ins on the individual tracks. On the subgroup, I have as inserts the Joemeek Meequalizer, then an 1176, then the Reel Tape Saturation plug‑in (R) and finally the D‑Verb for a little bit of medium‑sized room. The Joemeek EQ adds 2dB around 2k and the Tape Saturation helps the track to sit. The sends are two plate reverbs and a Digirack EQ that adds a lot of top end to get some more bite. As always, there are lots of volume rides. I do them both with the mouse and by drawing directly in the volume track. At the time I was using the Frontier Design Alphatrack stand‑alone single fader in order to get away from using the mouse all day long. It helps to do more intuitive fader rides, but I invariably get a lot more analytical [afterwards] and go in with the mouse.”
Double bass: Avid Digirack EQ III, Joemeek VC5 and Trim, Bomb Factory 1176 and Sansamp PSA1, Sony Oxford compressor and EQ.
"There's a great energy coming from the double bass. When I listened to it in solo, I thought that it would be tricky to make it sit, because it was noisy, rattly and boomy, but in the track it seemed to have the right effect. You have to put that down to the whole band being very sympathetic to each other while playing and Ted being able to play with the overall sound, rather than just get his head down and play things unrelated to what's happening around him. Again, I had choices of a few mics, including a [Neumann] FET47, which had more body, and a Telefunken for the top end, and I used both. I had the Digirack EQ on the Telefunken and an 1176. The 47 had the Sony Oxford compression, then the Sansamp to give it some more bite, and then Sony Oxford EQ. Again, the EQ and compression on the channels is corrective. I then had four plug‑ins on the subgroup channel, which were all about enhancing the sound and making the bass sit in the track. I had the Joemeek EQ, probably with a bit of bottom end rolled in, an 1176, another Sony Oxford compressor, and a level trim.”
"I seem to remember that all the drums were overdubbed in a day or two at AIR. For most of the time they consist of driving four‑to‑the‑floor bass drum and floor tom parts, with snare and cymbal thrown in later on in the track. In the beginning of the track I'd thin the drums out, and then add them in, so they build, and by the last chorus you get everything, plus maybe an added triggered drum. I had 12 individual drum tracks, plus a tom sub[group] and a snare sub, and a general drum master channel. The 'kikprint' [track] would have gone through outboard EQ and compression, probably the Neve 542. 'FloorTom1B' and 'FloorTim1B' are mics on the bottom and top of the first floor tom. The 'GogTom2' [track] is triggered by Drumagog. I added that to enhance the first tom and to get it to cut through. What often happens with toms is that when you whack them too hard, they start to thin out, and Marcus would have given it plenty of welly. There's some Digirack EQ and 1176 compression on the individual track and on the Drumagog there's a little bit of Reel Tape Saturation. On the tom bus I had some Joemeek EQ, which has really nice top end. Below the drum subgroup, there's an 'Industrial' kick overdub, which I think was a sample.
"There were a few keyboards — a harmonium and a Solina organ — going to a key sub[group] track. They probably had Markus' [Dravs] hand all over them and they glued the verses together. There's also a bow guitar track, I think bowed with a double bass or cello bow, and it sits in the mix purely for texture.”
Backing vocals: Audio Ease Altiverb.
"The backing vocals are a little rickety and wrong in places but that adds to the overall character and identity of the band. Not all the backing vocals are shown in this screenshot, nor is the sub[group]. I used two Altiverb patches: one was the 480 club, which is a favourite, and the other was a large-hall convolution reverb called 'Paradiso'. The band cited Fleet Foxes as a reference for the backing vocal sound. Quite often being able to reference another track is more effective than somebody waxing lyrical about what sound they want.”
"The session was at 24/48. I ran the stereo mix out of the box, via my Alan Smart C1 and the Neve 8803 EQ on the mix bus, and everything would have been printed back into the Pro Tools session via my Apogee A‑D converter. One thing that's interesting about this mix is that I used a lot of Digidesign stuff. There are many great third party plug‑ins, from Waves and Abbey Road and so on, and there is perhaps an expectation that you use them. But a lot of software that's bundled with Pro Tools is of a really high standard and you don't necessarily have to go and spend a fortune on plug‑ins. In the end it's about how good your source material is. Some people think everything can be fixed with a good mix, but that's not strictly true. A band playing well, recorded in a good room by a good engineer and a good producer, makes the mixing process a lot easier. If the source material is not very good, you're going to have to jump a lot of hurdles during the mix. In the case of Mumford & Sons, I was given really well recorded source material with plenty of options, so mixing was fairly easy and the results came out great.”