TRACKS UNTOLD: ONEOHTRIX POINT NEVER’S ‘LONG ROAD HOME’ : with engineer and mixer Matt Cohn

In this Lux Cache track breakdown series, we ask artists, producers, engineers and songwriters to uncover the creative process of their work in their own words. To debut our series, maestro audio engineer Matt Cohn shares the techniques and process of his work with artist Daniel Lopatin - also known as Oneohtrix Point Never. ‘Long Road Home’ is the leading single off OPN’s 2020 album ‘Magic Oneohtrix Point Never’. The track is performed, written and produced by Lopatin and features additional vocals from Caroline Polacheck.

This article is available as both a Patreon text post and .pdf document format. We ask you kindly to not share Lux Cache content outside of the Patreon, our contributors rely on your donations.

~

Cover art for Oneohtrix Point Never’s ‘Magic Oneohtrix Point Never’, by Robert Beatty - released by Warp Records

LC: Could you talk about your workflow with Daniel and your shared studio space Gaia?

MATT COHN:

Dan and I work out of Gaia, a studio I built for him about 2 years ago in Brooklyn. We wanted to create a space that was really tailored to the way we work, so there’s some neat features in the room to make our lives easier. All of the synths are mounted on the wall and come up as points on the patchbay for easy routing. We have a pedal send and return on the patchbay as well to route synths through, or to process DAW audio. The frequent flyers in our room are definitely the Moog One and the Juno 60. I absolutely love our Hedd type 20 monitors, and I can’t mix (or live, generally) without my ProAcs either. Other fan favorites include our vintage Wurli, RE-201 Space Echo, Flea U47, and the 2-in-1 VCR/TV combo for Dan’s massive VHS collection housed on a small library worth of shelves above the couch.  All of these things and more are incorporated into a massive Ableton template with external instruments and print tracks, armed audio and midi tracks, etc so when we sit down to begin recording an idea we have all of these resources ready to go instantly. In the case of this record we moved some of our studio gear to a spot in western Massachusetts where Dan stayed for about a month and recorded himself before passing it off to me for the mixes. I mix in Pro Tools, so before I began mixing I sifted through the Ableton sessions of each song to print wet/dry tracks as needed and bypassed plugins here and there while simultaneously familiarizing myself with the songs as I printed the stems. I spent a day doing this for all of the tracks on the record, which included importing them into Pro Tools and prepping the mixes (doing things such as color coding, organizing, cutting dead space,integrating each track into my mix template routing, etc).

OPN and Cohn’s shared studio space Gaia

LC: What was different in engineering Magic Oneohtrix Point Never versus Daniel’s Uncut Gems OST?

MATT COHN:

There were a few main differences between mixing MOPN and the Uncut Gems score/soundtrack. With the score I had to be conscious of the dialog at all times, constantly trying to find creative ways to make the music feel impactful while still giving room to hear what the characters are saying in the movie. If you’ve seen the movie then you’re aware of how chaotic the dialog is, and combining that with the sometimes chaotic nature of the score was certainly a trial and error process at first. WIth MOPN the mix was much more straightforward as there was nothing for the music to compete with, so every sound that deserved a moment had one. The other main difference was the client relationship, which was far more complex while working on the Uncut Gems score. In addition to Dan being in the room every day, Josh, the co-director was there as well and was heavily involved in the score to the point of asking for thing like the occasional small volume or panning adjustment during the mix. Sebo and Ronnie (from Elara) were there almost every day too, and the producers of the film Scott and Eli would also chime in with their notes. The notes from all of these people combined ranged from simple tasks like Josh’s aforementioned volume or panning adjustments, to things like replacing synthesized choir pads with a full real choir just days before the film’s first premier. With the MOPN mix there weren't half a dozen actively participating clients (sometimes with differences of opinion) to please, which of course made things a lot easier. Dan has such an incredible singular vision in terms of the songwriting and production of his records, and at this point we’ve spent literally thousands of hours in the studio together so when it’s just the two of us working we’re able to communicate very efficiently, and oftentimes don’t even have to communicate at all with how familiar we are with each other’s tastes and tendencies.

‘Uncut Gems’ Original Motion Picture Soundtrack - Composed by Daniel Lopatin, engineered by Matt Cohn

released by Warp Records and A24.

BREAKDOWN

LC: How did you treat each sound in Long Road Home?

MATT COHN:

My first step after doing the initial mix prep mentioned above was to get a balance as quickly as possible without worrying too much about eq, dynamics, fx or automation yet. If there’s a rough mix and the audio in the session doesn’t match up with it then I’ll usually spend some extra time beforehand making sure that my session is matched identically to how the rough mix sounds before moving on to my own balance. In this case the stems matched closely and only took a few adjustments to get them sounding the same as the rough.

Pro Tools project session of ‘Long Road Home’ stems

FIRST PASS ADJUSTMENTS

Once I had my rough balance ready I started listening for obvious things that needed broad EQ help such as filtering out of excess highs or lows, notch filters for unwanted frequencies, broad stroke bell curve boosts for frequencies I wanted more of, or shelving boosts/cuts depending on what was needed. Lately I’ve been using Fabfilter ProQ 3 often for this step but you can use any flexible multiband EQ such as the Ableton stock EQ or the Pro Tools stock EQ-7 (which I’ll still use sometimes for this as well). There’s some other EQs I’ll reach for early on in a mix too like the UAD 1073, Helios EQ, and Maag EQ - they’re all great for different things and sound amazing.

Initial lead vocal pass with Fabfilter Pro-Q 3

Dan’s vocals in this song and throughout the rest of the album utilize a vocoder effect that was printed in Ableton before moving to the mix. The challenge in general with his vocals was to maintain some of the character and mangled qualities that the vocoder provided while also taming some of the harsh resonances it was creating. Here is an example of my initial EQ pass on the lead vocal. You can first see that we have a generous high pass filter at 175Hz to cut out any unnecessary info in the low end. Note that this is a bit extreme and you’ll likely need to roll off far less for your applications but personally I didn’t like what was happening in the low end of the vocal and felt it was extraneous so I rolled off more than usual in this case. You can see that there’s also a dip of -2dB at 340Hz to cut out some of the mud, a light boost of 1dB at 1600Hz to establish a bit more presence in the vocal, and a lowpass at 13.5Khz to get rid of some of the harsh artifacts produced by the vocoder.

In this same stage of the mix I’ll also use saturation interchangeably for EQ (or in conjunction) if it feels like more of an appropriate tool to use instead.

Bass group parallel compression with Black Box Analog Design HG-2

Here is an example of some saturation applied in parallel to the bass group in order to achieve some more definition within the mix in the upper and mid register, especially on the initial attack of each note, but without having to raise the overall level of the bass track in the mix.

Snare drum saturation with Thermionic Culture Vulture by Universal Audio

Snare drum EQ processing after saturation

Similarly here is an example of some saturation on the snare track to get more snap from the attack and a bit more air and sustain on the actual “snares” of the snare. This created some unwanted weight in the midrange frequencies which was then addressed with this eq following the saturation.

STRING LOOP TREATMENT

The main string loop is one of the most important pieces of the song, but I originally felt it sounded too much like a sample and wanted to liven it up for the mix. For that reason I tried to get away from the digital realm and utilized the API 5500 and UBK Fatso we have here at Gaia. I made a 6dB shelf at 12.5K and cut -4dB at 400Hz. The Fatso was set to “spank” on the 2nd warmth setting with the transformer in circuit, and was coloring the signal from driving the input and attenuating the output to match the original level at which the signal was at before the compressor. After returning to Pro Tools I made some extra EQ cuts of -3dB at 1.8K, -2dB at 540Hz with a fairly tight Q, a broad Q boost of 1dB at 5K,  and mono’d the bass below 210Hz using the Brainworx digital V3.

DRUM PROCESSING

The drums in this song came from good sounding samples but needed some help getting the transients to hit harder and needed some more weight to them as well to fit into the track. Here (i) is the EQ on the kick followed by a transient shaper to exaggerate the attack and tighten up the sustain. The main snare is being saturated rather than EQ’d with an instance of Decapitator pictured here (ii), and the rest of the drums have either light EQ adjustments or nothing at all beyond volume and pan automation before hitting the drum bus. The drum bus was where the bulk of the work happened for this mix and was a combo of EQ, serial compression, and saturation. Here (iii) is the chain- The high shelf on the EQ is mostly to attenuate the hats but also helps to remove some of the sizzle from the snares. The Fairchild is smoothing out the dynamics but while still retaining the punch of the kick transients and also simultaneously helping the groove of the drums sit better. The Culture Vulture is what’s creating the heaviness and grit, but with the mix knob dialed back to let some of the unaffected raw transients through. The SSL compressor is barely compressing the signal but is helping more with both the transients and dynamics to take some of the load off of the Fairchild and also to add one last stage of dynamic control after the addition of the saturation beforehand.

(i) Kick drum EQ

(ii) Snare drum saturation using Decapitator by Soundtoys

(iii) Drum bus processing chain

CAROLINE’S VOCALS

There’s a few things happening with Caroline’s vocal chain to create that vacuum effect. First off there’s a lot of different panning automation happening in tandem between the tracks to make it feel like they’re crossing in space. The vocals then go through a series of short delays to give them some movement in the 3D plane and exaggerate the feeling of the panning even more. They’re also being fed to a parallel bus with some saturation and automated width as well. In the middle section they’re going through the Soundtoys Crystallizer with a pitched reverse effect, and the “sucked-through another dimension” thing that you hear is a mixture of automating some of the parameters on the Crystallizer in tandem with a lot of volume automation.

LC:

Every layer of the track has a homogenous feeling with one another, was there any treatment of the stereo bus to help achieve this?

MATT COHN:

The stereo bus for LRH was actually one of the more simple chains on the record! It was just the UAD Shadow Hills mastering compressor, shaving off about 2/3 dB at most on the opto compressor and about 1dB or so on the discrete side, with the ratio at 3, slowest attack, and fastest release.

LC:

Is there a certain approach you take to the treatment the more left-field and abstract sounds in the mix?

MATT COHN:

I think the approach I take in treating abstract sounds versus the more recognizable ones in a mix depends a lot on the context of the sound within the composition. If it’s more of a texture like the radio sounds in LRH then I’ll try to incorporate it into the bed of the track so the listener really feels that texture surrounding them and is aware of it, but as more of a feeling and not a main element that you can pick out at any given moment. In this case that meant a heavy high pass filter and a dip around the vocal range in the mids using a mid/side eq. Conversely there are many times that a “weirder” sound deserves the spotlight, and in which case I like to get creative and exaggerate the sound into an otherworldly place. The best example from this song might be the metallic sounds in the outro from around the 2:45 mark. By exaggerating the high end and applying some multiband compression to the top bands (as well as some clever volume automation at moments) there becomes that feeling of tension like cinematic swords scraping together, which is the main thing providing excitement/sharp transient energy and grounding you to this outro when everything else in the mix is floating around in this cloud between the movement of the metallic scrapes.

LC:

Your discography covers a wide variety of artists, styles and genres. Is there a common thread you bring to every project?

MATT COHN:

I try to avoid imparting “my sound” or whatever that may mean when I’m working on a record whether it's in a recording capacity or as a mixer. As far as something I bring to every project, I think maybe my greatest skill set would be the perspective I’ve gathered from working on such a diverse range of music in my career so far. I was lucky enough to have come up in a handful of different environments with specific types of workflows and mentors which certainly has affected the way I approach things in a unique way. I worked at a studio in high school that pretty much only booked rap sessions and they threw me in as an engineer on sessions within the first week which prepared me for a lot at a young age. My career in NYC started (and continued for years) at the XL Recordings in-house studio under the wing of their producer and engineer Alex Epton who taught me a tremendous amount and exposed me to working on types of music I’d never been familiar with. I also worked at Downtown Music Studios where I got the experience of the more traditional “large format Manhattan studio with pop clients''. Additionally I had the pleasure of working under Ben Kane, who makes records with D’Angelo, PJ Morton, Emily King, and many other artists in that scene which provided yet another perspective on recording and mixing techniques and philosophies. I also helped build Ben’s studio Electric Garden, the XL Recordings NYC in-house studio, and the B room for a studio in Philly called Headroom before building Gaia for Dan and I. Having all of these experiences rather than having only worked at one studio specializing in one thing is what has shaped the way I work, and I’m incredibly grateful for the well rounded education I received from the combination of all of these studios and my mentors that operated them.

LC:

What would your advice be to the kids and beginners having to mix their music in the confines of their own bedroom?

MATT COHN:

My advice for the kids having to mix their music in their bedrooms would be to take one of two routes. If you’re not necessarily interested in learning the craft of engineering/mixing but need to do so for your own music I would suggest finding a friend who’s passionate about a career in the field (or searching online for someone if you don’t know anyone locally) and letting them mix your records for practice. They’ll be happy to have some material to practice with, you won’t have to worry yourself with learning things only out of necessity, and can focus on writing and producing music instead while your friend hones their mixing skills. If you’d rather learn and take on the task yourself then my biggest advice in a home setup would be to reference other mixes often. Listen to what your favorite records sound like on your speakers and in your headphones. Pay attention to how the bass sounds in your headphones especially, as the low end may be the most unreliable point in your untreated/ lightly treated home studio. Try to reference and mix at a consistent volume level too, and volume match your reference tracks to the track you’re mixing. Mixing records in an untreated room is an uphill battle but definitely not an impossible one if you reference enough and really know the sound of your speakers and headphones.

~

Matt Cohn is a NY based engineer and mixer who has worked with acts such as Oneohtrix Point Never, David Byrne, Neon Indian and more.

You can visit his website and follow them on Instagram @complexmagazine