Inside Track: Justin Bieber (feat. Ludacris) 'Baby'
Technique : Recording / MixingIn his second career in the music business, Jaycen Joshua has become one of the USA's top mix engineers. With child star Justin Bieber, the brief was to bring urban grit to pop production.
Paul Tingen
Jaycen Joshua at his SSL desk in Larrabee Studios. Photo: Tiffany Lo
"Once you get to the upper echelons of mix engineers,” asserts Jaycen Joshua, "you are being paid for your taste. You're not hired just to make the rough mix sound better. You are the last line of defence, creatively, and you're expected to improve the record. It is your job to change things. And my idea of mixing is to make listening to a record like a movie experience. Every song is like a story, and so I may try to give something pleasing to the ear just for a split second, or open things up a little bit by taking things out, always with the question in mind: how am I going to tell the story from the verse to the chorus? How can I best tell the story of the 'B' section? How am I going to improve the story in the outro?”
Jaycen Joshua is talking from Studio 1 of Larrabee Studios in Los Angeles, a place that has been his home for the last seven and a half years. Larrabee houses three of the word's top mixers, the other two being David Pensado (see SOS January 2007: /sos/jan07/articles/insidetrack_0107.htm) and Manny Marroquin (see SOS December 2007: /sos/dec07/articles/insidetrack_1207.htm), and Joshua estimates that between them, the trio cover about 50 percent of charting American pop mixes.
Unlike the others, however, Joshua's list of published credits is relatively short, and it turns out that this reflects his close connection with Pensado. Joshua worked as the legendary mixer's assistant between 2006 and 2008, and during the latter part of his tenure, the two worked together on equal footing. In this capacity the duo shared a Grammy for Mary J Blige's Growing Pains (2008), but in many places only Pensado's name is mentioned, the assumption presumably being that Joshua was just his assistant.
This is despite a music industry pedigree that goes back, astonishingly, to his early teens, with his main introduction having been provided by Jheryl Busby, the Motown mogul who died in 2008.
Joshua: "My mom and my stepdad were pretty heavy players in the music business. My mom was Busby's assistant and later his general manager, and he thought that youth was very important, so he made me junior A&R when I was 13 years old, and he used me to bounce ideas off of. At age 15, I obtained a more prominent role that included signing bands. I worked as an A&R for Motown, MCA, Sony, Clockwork, and finally Dreamworks, until I was 24, when I decided that I had had enough, and went into advertising. That was great, but it was a nine‑to‑five job, and I had remembered that I had always been very interested in engineering, the little nuances that come with making people's music better, so I decided to take an engineering course at the Los Angeles Recording Workshop. As part of that I managed to arrange an internship at Larrabee. This was in 2003, and I started at the bottom, as a runner, getting food for people that I had dealt with in the past as a record exec! It was humbling, to say the least, but to become great at something, you have to pay your dues. I ended up being a staff engineer for Larrabee Studio 1, until Dave Pensado moved to Larrabee and I became his assistant. Pensado identified my abilities early on, saying I had a special gift in the genres of hip‑hop and rhythmic pop that he had never seen before, and asked me whether I could start mixing certain records with him. This quickly evolved into us having two rooms here at Larrabee and mixing many records together. He later noticed how my talents outgrew the team and encouraged me to spread my wings and use our respective talents for what we're best at under the same umbrella, 'The Penua Project', which is a combination of our last names. So for the past three years I've been mixing alone, on average a record a day.”
Blow Them Away
Early this year, Jaycen Joshua also earned Grammy nominations as part of the production team of Jamie Foxx's Intuition and Beyoncé's I Am... Sasha Fierce, mixing the most ear‑catching single from the album, the three‑time Grammy-winning megahit 'Single Ladies (Put A Ring On It)'. Another high‑profile hit mixed by Joshua is Justin Bieber's 'Baby'. Joshua and Pensado mixed the majority of the Canadian teenager's debut EP My World (2009), and Joshua shared mixing credits on this year's follow‑up My World 2.0 with Manny Marroquin and Serban Ghenea. My World 2.0 is Bieber's first full‑length album, and even more than his debut EP, represents a deliberate attempt to get the white teenage sensation to bridge the gap between clean bubblegum and gritty urban music. To this end, urban producers like The‑Dream, Tricky Stewart, The Messengers and Bryan‑Michael Cox were contracted, with everything coming together in the lead single 'Baby', written and produced by The‑Dream and Stewart, and also featuring rapper Ludacris. Both single and album ended up being massive international hits.
Joshua: "Yeah, the vision of Tricky, and Dream and LA Reid [of Island Records] and Usher [of the RBMG label] was that they wanted Bieber to straddle the line between American hip‑hop and pop. They did not want him to be a pop artist, they wanted him to be the cool white kid. I think the reason they came to me was that I'm very aggressive in my approach to mixing these kinds of tracks. I like to hit them really hard, so that you can really feel the drums, while everything else sounds expensive, so that it turns into a great hybrid. I really attacked the drums on this track and put my Class‑A Neve 1073 or 1084 preamps on every percussive instrument, to get roundness and punch. I ran the outputs from Pro Tools directly into the Neves, and they then went to the insert returns of my SSL, bypassing the SSL pres. It changes with each mix but I'm big‑time into my Neves at the moment, and I have 18 of them — 14 1073s and four 1084s.”
Organisation
Unlike many other successful mix engineers, Jaycen Joshua puts great emphasis on hearing the rough mix. "You really have to engulf yourself into these rough mixes nowadays, because these kids spend so much time and so much creative energy on these roughs that if you don't really understand the track and their ideas, you are going to get killed. Many older engineers will knock these kids, saying that they're only using plug‑ins, and that the older stuff sounds so much better, and everything is getting worse and worse and worse. And I am saying: your idea of worse is not everybody's idea of worse. Pro Tools, or other DAWs, and plug‑ins might not sound like outboard gear, but they are giving kids who may have better creative ideas than you and me the opportunity to do the same thing as we are doing. What they may be losing in sonic quality they gain in creative possibilities. In terms of the war that is going on between engineers nowadays and all these young engineers coming up on Pro Tools, nobody can tell me that engineers know more about hip‑hop than an 18‑year‑old African‑American kid who grew up listening to it. That is impossible. We are stuck in our studios every day, so how are we going to know?
"Sixty to 70 percent of the mixes I get in are done by the producer and engineers in their own studio setups. They most likely will have spent two to three weeks getting their rough mix to sound incredible for the label, and they will have lived with these mixes and the delays and the drops and the other effects and gotten used to them. Once your ear gets accustomed to things, it's what you tend to like. So I need to hear their rough mix before I do anything, otherwise I'll be chasing that rough mix for the rest of the process. Nonetheless, in 50 percent of cases, people won't send me the rough mix. So in the standard email that we send out to our clients I say that I don't work on records for which I don't have all the files, including the rough mix, at least three days in advance. I also ask them to send me everything on time, and leave every plug‑in, every effect, every automation move. If they used outboard, I ask them to print the track with and without the effect. Ninety percent of the time I may not use what's there, but it works much better for me to hear their ideas, and be inspired by them and respond to them, so I can either expand on them or suggest trying something different.
"Receiving the files three days in advance gives my assistants the chance to prepare the Session, organising everything in the way I like, colour‑coding and labelling all tracks and plug‑ins and so on, so all I have to do is come in and push up the faders. As you can see from the 'Baby' screenshots, I'll have drums and percussion in yellow, the high end in orange, the bass in red, lead vocals in red, my whole slew of effect tracks in dark green and the original effects of the client in light green, and so on. I also ask my assistants to put the contents of the tracks in the Comments window: we don't change the original track names in case we need to re‑import the Session and match it up. For someone who is mixing every single day of their lives, and particularly because people with home studios organise their tracks in any number of ways, organisation is a number one priority. In general I only have a day to do a mix, which is why my assistants do all the tedious organising, which can take two hours or more. For me those are two hours of potential creativity that I can never get back!Jaycen Joshua attaches a lot of importance to organisation in his Pro Tools Sessions, and has them meticulously prepared by his assistants: note the colour‑coding and the comment field listing the contents of each track. This screen shows a small section of the 'Baby' Session, with the prominent 'Synth Keys 1' track at top.
"I don't listen to the rough until the day of the mix. The first thing I do is get a cup of coffee, and I'll go on the Internet to check my email. That takes about half an hour, and while I'm doing that I'm listening to the record over and over again. When I hear something special, I'll solo it or have a look in Pro Tools to see what's going on. Of course, if the rough doesn't sound good, I'll try something very different, but if I'm hearing something that's incredible, I'll start from there. Advanced producers like the Max Martins, Tricky Stewarts, or Polow Da Dons, they also basically mix their own records. These guys understand sound, they understand how to represent the whole frequency spectrum. My job is to take these rough mixes to another level. I'm paid for my taste, and at the moment my taste is pretty respected, so I do what I feel is great. So when I arrive in the morning and listen to a mix I did the day before, I usually am pretty happy with it. However, in the cases that the client will ask for minor tweaks, like the vocals or the kick a bit louder, I usually will hear something else I'd like to change, and so I will stem things out and make these changes. If a client comes with requests for changes much further down the road, they are usually made by my assistants.”
'Baby'
Written by Justin Bieber, The‑Dream, Tricky Stewart, Christina Milian, Ludacris.
Produced by The‑Dream, Stewart, Milian.
Jaycen Joshua: "When I begin a mix, I'll basically start with attacking what I think is the most important part of the record. If I think the rhythm section is driving it, then I will start with that. If I think it is the vocal, or a main melody played on strings or piano, I'll begin with that. In the case of 'Baby', it was all about the beat and the vocals, and I began with the drums, which I wanted to sound very aggressive, with lots of punch. I also tried to open the track up a little by taking things out, mostly in the lead‑up to the hooks. All the greyed‑out bits were done by me. Another aspect of the way I organise my mix is that I split the whole track up in three groups: rhythm section, music (mostly keyboards and synths), and the vocals. So after having gone through the SSL, the rhythm section, the melody tracks, and the vocals, each are sent to sub busses on the SSL, which are then each returned to two faders. So I have a six‑channel submix before finally summing to stereo.
"One thing that's unusual about this Session is that all tracks are in stereo. Basically they are all rough mix stems that Tricky [Stewart] gave to me. He knows what he wants, so once he gets there, he prints it. For me that's great. The old guys will talk about how they used to push up faders and the drums would sound incredible, because the engineer would have spent hours getting the right mics and mic positions, and so on. Nowadays, in the music that we're talking about, that's rarely the case. People don't really spend much time dialling in the best sounds. But Tricky does, and for me that's great. The fact that everything is stereo is no problem for me. You don't see any panning in the screen shots because I pan on the board. I never pan in the box. I also ascribe to what I believe is one of Spike Stent's rules of mixing, which is that I'll null all the faders on the board when I start. I'll leave my Pro Tools trims as they are, and I'll do my overall trims on the board. I can then see when I've raised or lowered things. If a client comes in and says that the buzz synth is a bit too low, I can immediately see on the board how much I took off.”
Drums & bass: Neve preamps & EQ, Studer A827, Dbx 160, Pultec EQP1A, API 550A, McDSP Filterbank, Digidesign Trim Adjuster, Medium Delay, Empirical Labs Fatso, Moog EQ.
"There are eight percussion tracks. The 808 is really a hi‑hat, then there are two kicks, clap (with clap reverb), clap 2, hi‑hat, crash and drum roll. Everything was programmed on this track. Apart from the hi‑hat and the crash, which went through the SSL, I put everything through the Neve 1073 mic pres because I wanted punch on these tracks, but not SSL punch. Also, this specific record was a bit weird, because I asked my assistants to dump all drums tracks to two‑inch tape, on a Studer A827, using Quantegy 499 tape (I have a certain amount of rolls left; it's very difficult to get good tape these days). Dumping stuff to two‑inch doesn't always work, but when it does it rounds out the sound beautifully. The analogue tape gives it a colour that I can't get any other way. Colour is very important. Sometimes I run a track through a piece of gear, like a Fairchild, without touching the compressor, just to get it to add some colour to the sound.
"With regard to the specific tracks, the 808 already sounded incredible and had a roundness that I loved, so I didn't do much to it, just some EQ on the board, and it needed a little contouring, for which I used the 1084. The one drawback of the 1073 is that the EQ does not give me enough parameters, so I will put the 1084 on anything that I want to do some detailed Neve EQ'ing on, and use the 1073 for tracks for which I'm only going to use a pre. The two kick tracks went through a signal chain that was made famous by Bob Powers and that was passed on to me by Dave [Pensado], which is a Dbx 160x compressor going into a Pultec EQP1A EQ. This runs on a parallel channel and I add it in with the original signal. It gives you absolutely amazing low end. It's unstoppable. If you listen to Powers' work with Common or A Tribe Called Quest, he's a pioneer and the father of hip‑hop low end. The snare has the 160x going into an API 550A, again parallel. If you don't use parallel compression you risk things sounding thin. You begin with the original track that the producer loves, and if you manipulate that you may deviate too much from it, whereas if you blend a treatment in, it will be more pleasing to the ear.
"I also used a number of plug‑ins on the drums, mostly the McDSP Filterbank EQ and the Digidesign Trim Adjuster. If there's one plug‑in I couldn't live without, it's the Filterbank, and here I used the P2 two‑band parametric EQ. With regard to the Trim Adjuster, if one side of a stereo track doesn't have the same level as the other, instead of pulling the pan to one side I'll use the Trim Adjuster to increase the gain on one side and to make sure I'm dealing with a perfectly balanced stereo track. There's also a Medium Delay on the Crash, because it was originally a mono track, and I wanted it in stereo. So I set the delay to 21ms and mixed the original 100 percent to the left and the delay to the right. Why 21ms? It's a Digi preset on the Medium Delay, 21.19ms to be exact, and when I heard it, it sounded right. With 21ms you get enough separation between left and right and it's a bit dramatic and not so phasey. I didn't try to set the delay in rhythm. You usually already receive a perfectly timed Session, and to get a little feel it's good to have some things that are not exactly in time. The sound becomes like a drummer hitting two cymbals, left and right, at the same time.
"'Bass 2' is a synth bass, which also was treated with the A827 with the rest of the drums, and I then sent it through the Empirical Labs Fatso and the Moog EQ on an insert on the board. There's a plug‑in version of the Fatso, but I find that the original is untouchable. Putting the Fatso on the bass is like night and day, especially if you're looking for that lower mid‑range roundness that will give you a fuller sound. The Moog EQ three‑band parametric is an absolute go‑to box for me on the bass, even though I'll occasionally use the Focusrite 110. But the Moog is incredible. The mid‑range goes down to 62Hz and to me is better than the low‑end parameter. I don't know why, it just sounds richer.”
Keyboards & guitars: Cranesong Phoenix, Aphex Aural Exciter, McDSP Filterbank, Digidesign Extra Long Delay & Tel‑Ray, 'Motown' EQ, BASE Spatializer, Waves GTR.
"The main instrumental track is the piano track that we retitled 'Synth Keys'. It sounded a little thin to my ears, so I used quite a few plug‑ins on that.The main instrumental element of the song was a track originally named 'Piano', but labelled 'Synth Keys 1' in Joshua's comments field. This was quite heavily processed using Cranesong's Phoenix Dark Essence, the Aphex Aural Exciter plug‑in, McDSP's Filterbank EQ and Digidesign's stock delay plug‑in. The Cranesong Phoenix Dark Essence is a great plug‑in, it emulates tape saturation and it makes the sound a bit darker. The Aural Exciter adds a little top end. A lot of people don't like the fact that it adds distortion to your sound, but that high distortion gave exactly what I was looking for. The Filterbank beefed up the low end, and there's also a Digi Extra Long Delay. 'Js Keys Delay' is an effects track with the Digi Tel‑Ray delay that was blended in with the piano below. The Tel‑Ray is a great‑sounding delay that I love on piano. In general I don't like using reverb too much because it can cloud up your tracks and make them less clean. So it's fun to use delays instead and the Tel‑Ray is very distinct, it kind of mimics the sound of the hammers hitting the piano strings.
"A lot of people are down on the Digidesign plug‑ins because they don't look special, but I reckon that Digi are the ones who invented the game we're playing right now, so why would their plug‑ins not be good? I also like the DVerb, for instance, which was used by the client to make things more roomy and bigger, and I left it. With regards to outboard, I had the Motown EQ on the main 'Synth Keys' part, but no compression, because I wanted to retain an open feel and for the track to have dynamics, and I had enough compression in the rhythm section. I also had the old BASE Spatializer on the main pad in the hooks. I'm big on space, but we only have two speakers to work with, and it's a challenge to fit everything. I love the way someone like Dr Dre, for example, doesn't fill up his records with too many instruments. On this record, the Buzz Synth had a lot of low‑mid and sounded very full, so to separate that from the pad synth I put the pad synth outside of the speakers with the Spatializer.
"The guitar sounds were pretty good, so not much needed to be done. I used the Waves GTR Stomp 2 on the electric. It's one of Waves' better plug‑ins. I love the phaser on it, which is really lush, similar to the MXR phaser. The only thing with the Waves SSL bundle plug‑ins is that they overdid the emulation of vintage analogue gear, and made the hiss too loud. So I will always gate these plug‑ins, like the VEQ or the API. I could just switch the Analogue button off but I'm superstitious! I'm big on my tracks being as clean as possible. If there's hiss, it's there because I want it to be there.”
Vocals: Celemony Melodyne, Digidesign TL Space & Extra Long Delay, Massenburg EQ, Cranesong Phoenix, Waves Center, Renaissance Compressor & Metaflanger, Sound Toys Pitchblender. This part of the Session includes most of Justin Bieber's (red) and Ludacris's vocal tracks. The lead vocal was split to several tracks with different EQ settings for different parts of the song."Justin is in red, Ludacris in light blue. The main vocal is at the top, 'Lead All'. In fact, all the tracks in red are Justin's lead, but I vary the effects compression and EQ on his vocals during the song, so I split the main vocal track up and every one of the tracks below is dedicated to a different EQ for a different section of the song. When he changes his tone, the EQ has to change.
"The vocals were tuned by us. Tuning is one aspect of the game that you have to have. Consumers' and clients' ears have become accustomed to hearing vocals perfectly in tune, to the degree that if you now hear a record on which a person sings natural and is not tuned, you feel that something isn't right. I prefer to use Melodyne. I am not a huge fan of Auto‑Tune, unless it is the sound that you are looking for. For me it's an effect, not really a tuner. If you want to tune your vocals and for them to sound natural, you can do stuff with Melodyne that seems almost impossible. I also retain the original vocal in the Playlist, so if I want to retune or go back to its natural note, I can dial that up.
"I did the vocal effects 100 percent in the box on this record, and they're all sends to effects tracks that are marked in the darker green. (I didn't use the original vocal effects in the lighter green.) There are so many effects because, like with the EQ, I'll treat his vocals differently for different sections of the song. I'm using two convolution reverbs, one being the Digidesign TL Space, especially the 'Silky Gold Plate', 'QRS', and 'Big Pan Hall' reverb settings. The last is a phasey panning reverb.The TL Space convolution reverb was widely used, in this case to emulate a Quantec Room Simulator patch as one of the vocal reverbs. Other plug‑ins are the Massenburg EQ, which is very precise, with very tight Q, so you can notch out very small bandwidth frequencies. It has something that every EQ should have, which is that you can solo and sweep the frequency you are affecting, so you can find the frequency that you hate and notch it out.
"I also used the Cranesong Phoenix again, which is so important in rock music. I know a huge producer who uses the Phoenix Dark Essence with the 'Gold' preset on every track before he even starts. On this track, because Justin is so young and is a little harsh in his mid‑range, I went for the Luster to add fullness and some warmth.Cranesong's Phoenix was used again to thicken Justin Bieber's lead vocal, which was also compressed using Waves' Renaissance Compressor. Finally, on Justin's backing vocals, 'JBVrsBg1', which are the 'yeah‑yeahs' at the end, I put the Waves Center plug‑in, which is great because it lets you place exactly where you want your low or your high or your punch in the stereo spectrum. There's a great preset called 'Less Vocal' that takes the vocal away from the centre and expands it, so it's not so in your face. This gives your lead more space in the middle. Often it's good to play with the spatial aspects of your backing vocals.
"I also have a lot of plug‑ins on Ludacris's vocal track. It seems more than it is, because in fact his engineer always gives me great vocal recordings, so it was a matter of notching things up or down a little bit here and there. The 'PingPong' effect is the Extra Long Delay that is ping‑ponging his lead left and right. It's a panning delay that's EQ'ed, put through a reverb, then phased out a little through the Waves Enigma. The 'RapH3K' is my emulation of an Yamaha SPX90, and the 'RapVerb' isn't really reverb, because everyone knows that reverb is the kiss of death on rap vocals. Reverb and rap don't mix. Right at the bottom are The‑Dream's ad libs. On his BGs [background vocals], I used the RCompressor, the Choruser, the PitchBlender and Metaflanger.Sound Toys' PitchBlender was used to treat The‑Dream's vocal ad libs at the end of the song. Plus I added the Center plug‑in to place him outside of the speakers. Dream sets the pace creatively in terms of effects and trying to entertain people sonically and he's always trying new stuff.”
Submixes: SSL G384, GML 8200, Avalon 2055.
"As I mentioned above, after having gone through the SSL, the tracks will come up on the desk in three stereo pairs: drums, music and vocals. Using the inserts on the drums submix I had the SSL G384 compressor and the GML 8200 EQ. I left the music untouched, and on the vocals I had the Avalon 2055 EQ for some air, to add some sparkle. I'm huge on dynamics, and in the subgroup mix I may lift an entire section a dB for the chorus or an outro. Dynamics are key in pop music, and subgroup mixing gives me some flexibility and control over the dynamics. Finally, the Session was in 48kHz/24‑bit and I mixed back into Pro Tools, using my modified Lavry Gold A‑D converter. A lot of people don't realise how important your converters are. I monitor my mix after the converter which is important, and this converter has a specific sound that I love.” .
In & Out Of The Box
An 80‑channel SSL XL 9000K desk with a custom suede trim dominates Jaycen Joshua's room at Larrabee. When queried, Joshua proudly declares himself to be beyond the analogue versus digital and in‑the‑box versus out‑of‑the‑box debates. "I am new school: whatever works, works. I owe 99 percent of my know‑how to Dave [Pensado] and he's a big fan of the SSL, and during my A&R career every mix that I witnessed was done on an SSL. So I'm conditioned to like the SSL sound and to working with it. At the same time, people like Serban [Ghenea] and Phil Tan [see SOS February 2007 issue] are doing incredible mixes in the box. A lot of people have opinions about in the box versus out of the box, but both have their advantages. There are things that you can do in the box that you can't do outside. So with regards to the out of the box and in the box arguments, I am a proponent of both. When I go into battle, I want all my guns ready to go. I want the best of outboard and the best of in the box.
"For example, I really love the 9000 K, because it gives me some of the characteristics of the G, and the SSL computer is amazingly fast. However, I'm now doing all my automation moves inside of the box, and my SSL is in essence just a big summing box with multiple faders. To my ears, summing in analogue still sounds better than summing in the box, though I don't think there are any consumers out there saying: 'Man, this record sounds so great because it's been summed in an SSL.' I don't think the consumer cares. The issue is nonetheless important, because while we're not trying to please the consumer, who may be listening to MP3 and very compressed files anyway, as a mixer you are trying to please the clients, who are the artists and the A&R managers and record companies. Mariah Carey, for instance, is very hands‑on with her records and she hears the differences. She'll hear whether her vocal is pristine and clear and whether all the dynamics are there and whether the reverb tails right to the very end. You can lose clients if you don't consistently blow them away. You want your client to go: 'Wow, I've never heard anything like this!'
"The advantage of mixing in the box is, of course, that you have instant recall, and A&R people and artists are now so used to being able to request unlimited changes that the guys working outside of the box are getting killed by it. The way I handle that is that, first of all, I try to get pretty close with my first mix. I tell people that I work with that if they are totally not happy with my mix, we have a different vision on the record, and I may recommend them another mixer to use. Who am I to tell them that their vision is wrong? This is rare, but it happens to all of us. However, for cases where they want minor tweaks I now pretty much have to stem everything out. I'll do multiple stems of every instrument, because you also have your stereo bus compression, and if you change the stem of, say, the kick, it will affect the stereo bus compression. So I'll give myself stems from every angle possible, every instrument, the whole rhythm section, the vocals, so that it's easy to accommodate minor requests for changes later on. It's a lot of work for my two assistant engineers, but I had to do it.”
No comments:
Post a Comment