Issue 93


Ableton Live 12
What’s in. What’s out. What to expect.


Mix Masters: Jonathan Castelli on Assembling Downtown

Jonathan Castelli on Assembling Downtown.


15 December 2015

Artist: Macklemore & Ryan Lewis
Song: Downtown

Bruno Mars and Mark Ronson’s Uptown Funk was a very well orchestrated pastiche of ’80s funk. All the key elements were there — keytars, horns, low fisheye lenses and square-shouldered pastel sports blazers. But there was still a collected, serious side to the parody. Whether it was yammering on the phone while getting a perm or reading the paper while their shoes were shined, Mars and Ronson were getting down to the business of being funky.

Macklemore & Ryan Lewis don’t have to parody jack. Their songs and videos are too oddball to attract direct comparisons. Especially so on their latest hit, Downtown, which, on the face of it, could be read as a direct response to Uptown Funk, but is so much more bizarre.

Downtown’s chorus — top-lined by Foxy Shazam singer Eric Nally while riding a chariot pulled by Royal Enfield motorcycles — doesn’t come in until almost the two-minute mark. The verses — one-upping the anti-cool vibe of Thrift Shop by geeking out on mopeds — are made up of completely different sections, featuring not only Macklemore, but also rap pioneers Melle Mel, Kool Moe Dee and Grandmaster Caz. The song has echoes of both the ’70s and the early ’80s funk aped by Uptown Funk, and is almost entirely played on live instruments with prominent sax, trumpet, tuned percussion, and piano.

The song’s unconventional nature apparently also initially foxed its makers. Mixer Jonathan Castelli recalled, “Ryan called me and said they had been working on this song for two years, and they really needed fresh ears and perspective to help finish it. When my assistant, Ryan Nasci, and I listened to it for the first time, it sounded crazy. It was going to take a lot of work to make it feel like one cohesive concept. It sounded like four songs stitched together with transitions that made the sections sound completely different. There were no effects and the voices weren’t cutting through enough. The challenge was to make it sound like one whole piece so America would not be freaked out by it, as well as getting the voices to step out in front, make the whole song jump out of the speakers, and take it into the radio world.” A big job for an already unwieldy song.


Originally from New York, Castelli studied saxophone at the Hartt School of Music on Long Island, and started and ran a studio together with his father in the same area, where he cut his engineering and production teeth. 

In 2013 he signed to Mirrorball Entertainment in Los Angeles as mix engineer and producer, where co-founder and star mixer Tony Maserati took Castelli under his wing. Castelli: “The first thing we mixed together was a Lady Gaga song, then he invited me to move in across the hall from him, where I still have my studio. It’s all in-the-box, my only outboard being a Rascal Audio Two-V preamp and a Kush Audio UBK Fatso compressor, both of which I use over the master bus. I also have two of Tony’s Neve 1066 mic pres, plus the Black Lion Microclock MkIII and Sparrow AD/DA. My monitors are PMC IB2s, with a Bryston 4B amp. I also have Chris Pelonis Mode 42 MkIII speakers with concentric drivers that I use like Auratones. They are awesome.”

Castelli’s credit list already includes mixes for Lady Gaga, Beyoncé, Ariana Grande and Backstreet Boys. As one half of 4FRNT, he also did a remix of Somebody That I used To Know that was endorsed by Gotye. As well as Downtown, Castelli eventually mixed the entire yet-to-be-released Macklemore & Lewis album.


Castelli got the gig through Joshua ‘Budo’ Karp, a frequent collaborator with the Seattle duo he’d met at a writing session a few years ago. 18 months after that first meeting, Budo sent Castelli a text asking if he was up for mixing some Macklemore & Lewis material. He was hardly going to refuse, and got stuck into the mixes at his Mirrorball studio.

The uncertainty around exactly how to approach Downtown surprised Castelli. It was the fourth song he’d mixed for the album and the direction for the first three had been fairly clear. Although the duo weren’t exactly clear about the direction for Downtown, they did recognise that Castelli’s first mix wasn’t what they were after.

Castelli: “For my first mix, they had given me no direction, and when I sent it to them, I was very happy with it. Ben [Macklemore] contacted me and said, ‘Man! How did you get Eric’s voice to cut through in such a loud arrangement?’ But when Ryan [Lewis] called he said ‘Jon! I really like that mix, but it is completely the wrong direction! I like that warm, fat, more tube-like sound with extended bottom end, that you got on the other songs, but this song needs to be a little more natural sounding. It needs to be less EQ-ed in the low end, with more of a natural mid-range — more solid state.’

“I immediately knew what he meant. I love tube gear and tend to go for big fat bottom with subharmonic stuff and saturation. That worked on the first songs I mixed, but for this song he said, ‘do yourself a favour and go listen to Grandmaster Flash’. Finally he gave me some of the vision. If he’d given it to me earlier, I could have saved myself three days’ work! I had worked really hard on it, because I knew it was earmarked to be the first single from the album. Usually you can backtrack in a mix, but not with this song. I decided to start again. Downtown eventually turned into a two-week mix!”


The Pro Tools session for Downtown is an impressive 112 tracks, including group and effect tracks. From top to bottom it consists of five mix tracks in purple, including the rough mix right at the top, eight group tracks in bright green (drums, guitars, keys, synths, strings, horns, vocals and effects), 34 drum and percussion tracks (mostly red and yellow), three effect tracks, five bass tracks (mostly orange), one piano track, seven guitar tracks (dark green), one Oberheim synth track, and five horn tracks (yellow-green). The bottom of the session is taken up with a massive 45 vocal tracks, including 14 for Ben Macklemore (though the three tracks labelled Mod are of Macklemore demoing guest rap vocals), 18 tracks for the three guest rappers, Eric Nally’s, the choir, and other incidental vocals in between. 

While 112 tracks is indeed large, Castelli’s mix session actually comprises the stems Ryan Lewis’ engineer Stephen Hogan supplied. Hogan had bounced them out of an original session that added up to well over 200 tracks.


Castelli: “Ben’s main vocal track sounded really good, and didn’t need too much work. I have five plug-ins on it. The FabFilter Pro-Q2 is a high-pass at 100Hz and also takes out some 700-900Hz — a frequency range that I tend to dislike in vocals. The UAD API Vision Channel Strip adds a nice aggressive sheen. You can make it sound hi-fi if you want, but it’s best suited to that solid-state aggression with a rock edge that I wanted here. I boosted high end and took out some 240Hz with it, and also used the compressor, set to a ratio of 3:1. The UAD Pultec Pro is boosting 200Hz, adding some broad warmth, plus I also had the FabFilter Pro-DS. Ben’s vocal has no reverb. They don’t like reverb on his voice; I only used it on him in one song.”


Castelli: “When I was nearly finished with the mix, I went to Ben and Ryan’s studio in Seattle to work the three rappers they’d recorded into the session. They asked me to put the mix through a console, so I rented EastWest Studio B to use the vintage Neve for a more classic 1970s sound. It didn’t work, it didn’t sound as good as my in-the-box mix that had gone through my Rascal Two-V and UBK Fatso. We spent $2000 for a day, and it didn’t sound any better!

“I mixed a song for Robbie Robertson on his Neve at the Village in LA, as well as in the box, and via a Dangerous summing mixer. In that case we also ended up using the in-the-box mix, because it sounded better. The Neve and summing rig made it sound mushy; the mix lost its edge. 

“That said, I recently mixed a couple of songs at a studio in Burbank on Chris Lord-Alge’s old SSL. I don’t think I could have made it sound the same in-the-box. I also went to Blackbird in Nashville a month ago and summed a mix through a console, and again it sounded great. I’m open to either way of working. In-the-box is definitely easier for recalls, but as far as the sound is concerned, you can mix in-the-box or on a console. It depends more on the room, the monitors, and most of all, the song.”


Castelli: “Eric’s main vocal track has eight plug-ins on the inserts and a send; though all the plug-ins are making subtle changes. I started his vocal chain by hitting the Waves API 2500 compressor pretty hard, then I boosted 100Hz and took out 500Hz on the Waves API 550B. He has a quiet voice, so I needed to give him lots of body. As with Ben’s voice I used the FabFilter Pro-Q2 for its hi-pass filter and ducking around 700-900Hz. The FabFilter Pro-Mb is controlling some of the 800-2000Hz range that comes from the mic. The Softube TubeTech PE1C EQ is attenuating some 10kHz, and the Plugin Alliance Maag is boosting 160Hz, 650Hz and 2.5kHz. Finally, the Softube TubeTech CL1B is doing some mild compression. The send is a UAD EMT140 plate, set to a short, bright reverb, with a long pre-delay, to give his vocal a short but big glow, so he sounded huge. His vocals are also parallel compressed using a Universal Audio LA 610 and a Bomb Factory BF76.”


Castelli: “The horns are a big part of the song. You don’t get many horns in pop music these days, so I wanted to make sure they sounded special. I wanted a grimier, darker horn sound, rather than the ’70s funk sound the track already had. I opted for a dark tape machine vibe, with very little compression. It was more about filtering the top end and finding resonant peaks so it cut through the track without sounding too fancy. The horn tracks are all sent to a bus on which I placed the UAD Studer A800 tape machine emulation, the Waves SSL E-Channel to try to make it sound more classic, the Softube TubeTech ME1B EQ dipping around 2kHz and boosting at 200Hz and 5kHz, the Pro-MB, and the Pro- Q2 which is a hi-pass. Mostly it’s about the A800, which is set to 15ips, aiming to make it sound darker.”


Castelli: “Towards the bottom of the session are the rest of the vocals, including the big choir with about 60 voices compiled in stems and the three rappers. All of them have no plug-ins at all. They were more about balances and rides. Many engineers feel they have to do something on each track. I may have used a lot of plug-ins on Ben and Eric’s vocal tracks, but in general I subscribe to Tony’s school of thought, which is that you don’t have to do much, and you only do things when really necessary. If it’s already good, don’t touch it!”


Castelli: “The drum tracks fall roughly into two categories. The more obviously programmed drums are sent to a bus with the Studer A800 and parallel compressed using a dbx160. The rest of the drum tracks comprise drum samples that aim to sound live. The drums already sounded great, so I just used EQ to make them sound a bit harder and added a sample to reinforce the kick sound. The bass also didn’t need much. I put on a bass amp plug-in and Culture Vulture parallel compression to make it gel with the drums. It was all about the glue.”


Two weeks for a mix sounds exorbitant, but Castelli said his mix for Downtown required a lot of fine-tuning because, “Ben and Ryan wanted to come out with a song that was unlike anything that had been released in the last 10 years. They’d had three No. 1s, and although they want to be successful, they really wanted to ‘wow’ people. High chart positions were less important. During one phonecall I said to Ryan that I saw the song as their response to Uptown Funk. Those guys are posh and living in uptown, whereas Downtown is more down-to-earth. He really liked that perspective.

“Specifically, instead of going for a big bass, R&B sound as with the first three songs, my challenge was for Downtown to sound more like a throwback to the seventies; more Neve and API than tube. Of course, it still needed enough bottom end and punch to sound current. The problem with my direction at that stage was more sonic, as I had already to some degree resolved the issues of Downtown — with rap verses and a clean chorus that transfers into something akin to classic rock — sounding very segmented. And, as Ben had indicated, I had managed to get the vocals to cut through. Bringing all that together was a big challenge. But I love being challenged!”

With so many pieces to assemble, more than ever, Castelli just took the time to listen. He “just listened to it for a good hour, soaking it in. For me mixing is about the journey and emotional connection you get with a song, not about what plug-ins I use. With other songs I might dive in right away, perhaps starting with getting the drums sounding good, but with a proper production and a great song like this I first listen to it a lot. When I have a connection with a song I tend to do less nitpicking and mix more from an overall picture. I do listen in solo mode at times, of course, but you have to be careful not to go down a rabbit hole, as your mix may start to sound disjointed. Everything has to fit, so I mix a lot with everything in.

“Like most people I generally start with the rhythm section, but in the case of Downtown I began the mix by focusing on Ben’s vocals. The drums were already in a good place, though perhaps not quite hitting as hard as they are now, so I focused on getting the vocals to be in charge, and then built everything around them as support. After Ben’s vocals I worked on Eric’s vocals in the chorus, and then the rest of the vocals, so they told me the story of what needed to happen between each section. After that I worked on the horns, followed by the drums and bass, really making sure the groove was solid; and then I worked the other instruments in.”

Somehow, Castelli managed to make everything work together. Over the course of five and a half minutes, the song takes you on a journey that is partly a throwback, definitely inventive, and, most importantly, fun. Pop songs aren’t typically this diverse, and Castelli’s mix makes those transitions effortless so you can just get down without a thought for the business of how it all came together.


Leave a Reply

Your email address will not be published. Required fields are marked *

More for you

Filter by
Post Page
Feature Recording Mix Masters Issue 86 News Quick Mix Solid State Logic (SSL) Microphones Ribbon Microphones Royer Software + Plug-ins Waves Audio interface Focusrite Regulars Issue 89 Fresh Produce Live Sound Presonus
Sort by
Issue 93


Ableton Live 12
What’s in. What’s out. What to expect.