Flo Rida and T–Pain scored a massive worldwide hit with 'Low' — but only after it received the attentions of Miami's star mixer, Fabian Marasciullo.
At 29, Fabian Marasciullo is perhaps the fastest rising star of the US mixing firmament. Originally from New York, Marasciullo is one of several top mixers — others include Phil Tan, Demo Castellon, and Marcella Araica — who came through the Full Sail school in Orlando, Florida. After completing his engineering course, Marasciullo went to work at the Hit Factory in New York, where he worked his way up from intern to engineer — a period he reckons was more educational than Full Sail.
While at the New York Hit Factory, Marasciullo was head–hunted by Rodney Jerkins, and during a three–year stint with the famous producer/songwriter, he helped him record the likes of Michael Jackson, Destiny's Child, Britney Spears, Toni Braxton, and so on. He also assisted star mixers such as Jean–Marie Horvat, Tony Maserati and Dexter Simmons, and eventually became a full–time mixer himself.
Marasciullo went independent in 2001, and moved back to Miami to focus on mixing the urban music he was a fan of. "At the time it was open season down here," recalls the mixer. "There had always been the Miami booty bass music, but there were very few real engineers. But Miami was about to become a booming success in the record industry, and I went in there with the New York attitude, very aggressive. Miami is a thriving place now for recording. Everybody comes down here, and people can go to the beach in the daytime, the studio in the evening and the nightclubs at night."
Since 2001 Marasciullo has mostly been based in Studio C at the Miami Hit Factory, and his rise to the top has been nothing if not meteoric. His breakthrough came in 2004, when he mixed Lil' Wayne's hit single 'Go DJ'. Barring the odd scheduling conflict, Marasciullo now mixes all releases of Cash Money Records/Young Money Entertainment, Terror Squad, The Runners and T–Pain, and in an age where most hip–hop and R&B albums feature a great variety of mixers, Marasciullo often mixes whole albums, such as Lil' Wayne's Tha Carter II (2005), T–Pain's Epiphany (2007) and Birdman's 5*S (2008). He mixed a staggering seven number one US hit singles during the last year alone, including 'Lollipop' by Lil Wayne, 'Kiss Kiss' by Chris Brown and 'Low' by Flo Rida featuring T–Pain. "Everybody thought I was nuts to want to get into the urban market, and out of the pop market, but look at the charts today: there will be five or six rap records in the top 10. Rap is pop. The line has become so blurred that it has become the same thing."
One would have imagined that Marasciullo ended up mixing Flo Rida's 'Low' through his connection with T–Pain, but apparently this was not the case. Instead, explained Marasciullo, he was called in because the original mix wasn't happening in the charts. "Mike Caren, the A&R guy, thought that the record could be a big hit, so he wanted me to have a shot at it. But the producer, DJ Montay, who is from Atlanta, hadn't even heard of me. Getting the files was a big ordeal, because the other engineer was reluctant to give me the right stuff. There were also different versions. They weren't sure that T–Pain would be cleared, so they had two different versions of the hook, one with T–Pain and one with another singer. Then they decided that they wanted a new hook, and so on. They continued tracking while I was mixing, and I actually mixed the song six times. These were not recall mixes, these were mixes that I began from scratch. By the time I had the correct files, I had been in the studio for six days!"
When asked whether he found it hard to maintain his objectivity in relation to the song after six days of mixing and using several different versions, Marasciullo has a surprising answer: "People make fun of me when I say this, but my job as a mixer is technical. It's about frequencies and pockets and spacing. I am very scientific when I mix, and I don't even listen to the song. I break things down to ones and zeros. I know people who are all about vibe and they like to mix with the lights dimmed and all that. I'll happily get down to work with the neon lights on. Maybe that's not the passionate approach people want to hear about, but it's the way I do it."
"I work very tightly with producers all the time," he explains. "Most of my clients are good friends of mine, and I respect what they do. I allow them to do their job, which is to produce. My job is to mix. The emotional, performance aspect of the tracks is more for the producer. Of course, it's not a cut and dried separation. We kind of feed off each other. But I respect the producer's space. When people ask me what I think of the song, I will often answer that I haven't even heard the song. I will have been listening to how the drums fit with the guitar and the vocal. I'll finally be listening to the song itself when I play my mix to the artist and the producer."
Naturally, Marasciullo's unorthodox view of his mixing work is reflected in his actual mixing approach. He relates that, once his assistant has conducted all the usual preparations — cleaning up the session and arranging it according to Marasciullo's preferences — he starts a mix by working on the vocals in relation to what he calls the "two–track". The latter is not to be confused with the rough mix, but is instead "the rough instrumental mixdown of the track that the artist sings to. The producer will often send it to the artist as an MP3."
Marasciullo explains: "Sometimes I feel like wearing a T–shirt saying 'If you like the rough mix so much, use that!' But the song will usually have been recorded months before it comes to me for the final mix and the artist will have been listening to the rough mix for all that time, and regardless of whether it's shitty or not, for them it will sound right. So I take my cue from it. My job is to mix what the producer and artist originally created, not to give my own interpretation of it. The only thing I don't pay attention to on the two–track are the drums, because I will make them a lot harder and in your face. Of course, if the producer and artist tell me to go crazy, I will do so, but that's rare.
"I know right from the bat that the levels of the instruments will remain at more or less the same volume levels as on the rough and the two–track. So I'll begin most mixes with working on the vocals relative to the two–track. I will do my own cleaning, getting rid of frequencies that bug me, and so on. When I'm done, I'll start on the drums, and then the bass, and then the rest of the instruments. It depends on the record, of course, but in hip–hop it's all about the drums. You want that 808 to be screaming. I will often restart mixes three or four times. Put everything back to zero and start again, re–blend and EQ everything and put the vocals back in."
Despite being too young to have witnessed the golden age of analogue equipment, Marasciullo is a fan of analogue desks and tape. "I run all my mixes through the Neve 8068 at my room here at Criteria. Saturday Night Fever and Hotel California were recorded on it. I sometimes move around Criteria depending on what record I'm working on and what room is free, but nine times out of 10 I'll be in Studio C with the Neve. I prefer it to the SSL, I find that I get more 'oomph' out of the Neve. I did a lot on the board for this song. If you look at the [Pro Tools] Mix window for 'Low' you'll see that I didn't use so many plug–ins on this track. On the vocals, yes, but the rest of the track was pretty much all on the board. My stuff will always touch outboard gear. If I had to mix without my Distressor, I might have to shoot myself — I don't think I could do it. And I will still almost always throw something, often a drum blend, on half–inch analogue and put it back into Pro Tools and then slide and nudge it until the feel is right again. It's not only to do with submixing to analogue, there's also something that happens to the sound when you go down to two tracks... I guess you could call me old school in that sense.
"I have to admit, though, that I'm working on doing things in the box now, mainly because budgets have become so much smaller. Budgets plummeted after 9/11, and have never been the same again. With Rodney we had a $1000–a–day food budget, stayed in five–star hotels and rented Ferraris and Porsches for the engineer, producer, and so on. We were superstars. After 9/11 we had to argue over plane tickets, equipment rental and so on. I'm at the moment considering setting up my own room, and doing everything in the box. I'll probably just get a small fader pack. Buying a large controller seems a waste of money to me.
"On Lil' Wayne's Tha Carter III I've moved into the box. I've been learning to have the same quality while spending less time and money in the studio. I still treat Pro Tools as a console, by laying my sessions out in it like on an in–line console. The technology has become better, yes, but it's also been a matter of me finding the right tools, the right plug–ins that give me the sounds I want. I've definitely been finding that more and more recently, with favourites being the McDSP plug–ins, the SSL plug–ins, Sound Toys EchoBoy. They're all great. "
Vocals: Waves Renaissance Vox, De–esser, Metaflanger, Renaissance Channel and S1 Stereo Imager, Digidesign Revibe, Line 6 Echo Farm, McDSP Analog Channel, Neve 8068 desk EQ, Neve 33609, Empirical Labs Distressor, Dolby 740
Marasciullo: "There were lots of vocals and I chose not to bounce them down, because I prefer to process every individual bit separately. It uses more DSP on the computer, but you have more flexibility in the way you can treat sections, and hence more depth in the overall vocal sounds this way. The 'Lead' and 'Dbl' tracks are the main verse vocals by Flo Rida, he's doubling himself like many vocalists will do. Flo has a keen melodic sense when he's rapping. The A, B and C auxes separate the three different parts of the verse. The 'Whine 1, 2, 3' backing vocals only occur in the third verse. Underneath 'Aux 4' are T–Pain's backing vocals. 'Main Hk1,2,3,4' are all T–Pain doing a unison of one note.
"As is usual for me, I began the mix with the vocals, set against the rough music bed. You can see the latter in the left pane track window in the Edit Window, where it's called 'T_Pain_5 LO'. Of course I later muted that track. After having cleaned up and EQ'd the vocals I worked on the drums, and then the synths. I then brought the vocals back in to get their levels in relation to the music and do some automation.
"I used the Waves Renaissance Channel on the A, B and C auxes. I use the RChannel on all my vocals. It's basically a combination of the RCompressor, the REQ, and the RVox, all in one, and it's great. For this record I used a pretty cool preset called 'CloseWarmVox', and I tweaked from there. You can see that it rolls off at the bottom and it has a little touch at the top. I almost used it as a filter on this particular song to get clarity in the bottom, add some high and have a bit of compression. I also used it on T–Pain's vocals.
"In addition, I used the Waves De–esser and Renaissance Vox on both Flo's and T–Pain's vocals. With the De–esser on Flo I cut around 5424Hz and T–Pain around 4500Hz. Some people put a de–esser on an aux, but I find that this doesn't really grab the frequency enough. So I de–ess twice. I will first grab a mid–frequency with a plug–in directly on the channel, and I will then put a de–esser on an aux. You will see my second de–esser on the Mix Window, on the 'Verse' auxiliary track.
"The RVox got rid of extreme peaks on both Flo and T–Pain's vocals. I used it to just clean things up if there were big level differences. But it's not hitting the vocals heavy, it only works when the vocals hit the threshold. It also has a great gating feature, and so it cleans up little blips here and there. The settings on all plug–ins were very similar for both Flo and T–Pain, because they have very similar voices. I had the McDSP Analog Channel only on T–Pain, though, basically to take off some edge. I use that plug–in when vocals are a little too brittle or bright. It really simulates hitting a tape recorder, and it also provides a little bit of analogue tape compression. It's a pretty good plug–in and you can select different tape recorders, like Tascam, Ampeg, Studer and so on. For this record I used the Studer preset.
"For space around the vocals I used the Digidesign Revibe, and the Echo Farm delay and Waves S1 Imager on the Aux tracks called '1/4', '1/8', 'Tape1/4' and 'Tape 1/8'. Oh, and there was a Metaflanger spread on the 'Spread' aux track, affecting the hook. The '1/4' denotes quarter–note delay and '1/8' eighth–note delay, using the Echo Farm. I use that a lot. It's cool because it simulates tape delay and it gives this kind of old Beatle–esque delay. Basically, '1/4' and '1/8' are the clean–sounding delays, and 'Tape 1/4' and 'Tape 1/8' are dirty–sounding delays. I put the S1 Stereo Imager straight after the Echo Farm, to get the delay just a little bit out of phase, so you're feeling it more than hearing it. I do this pretty much on every record that I do.
"The rest was done outside of the box. I used the Neve desk for EQ, and on the hook, ie. on T–Pain, I had a Neve 33609. On Flo I had a matched stereo pair of Distressors. I would have a hundred Distressors if I could. I use them on everything, vocals, drums, bass, guitars. They are just magical. Finally, I sent all the vocals through a Dolby 740. It's an old Dolby unit that was made for film music. It acts like a compressor, but it's an EQ. You can set it so that when the guy sings harder, he sounds a bit brighter. It's a really interesting device, and it gives me a glistening on the voice that no other EQ can give me."Drums: Fairchild 660, Waves Renaissance Bass, Digidesign Revibe, API 560, Neve 8068 EQ, Empirical Labs Distressor, Drawmer MX40
"The Fairchild 660 plug–in is on the 'Boom' track, which is an 808 running through the whole track. It provides the bass for the song. The 660 has a nice smooth sound, and it really controls all the peaks. There are no compressors on my Neve board, so I had to use the 660. I added the RBass [bass enhancer] to the 'Hitz' track, because it was a high–pitched orchestra stab, and I added an octave below. The main reverb was the Revibe, routed via bus 55–56, just to give some space to the 'Hitz' sound. I had that automated so it would pop up in certain parts. I also used an API 560 on the clap.
"Otherwise there was lots of EQ on the board, and Distressor, Distressor, Distressor. On drums it gives me an attack that no other compressor can give me. I will have one drum just raw, exactly as it was given to me, with just a bit of EQ on it. I will then mult it into a Distressor and that will give it this extreme attack, and then I blend it in the way like it. I also use a Drawmer MX40 Punch Gate, which is great, because you can set the frequency where you want the gate to open. When you put it in punch mode it gives you just a little boost when the gate opens at that selected frequency. It is incredible on snare drums, when the gate opens it and pops to give that snare frequency a real extra boost. I may also use the punch gate on some really bland vocal performances; if you set the attack right it gives it a great feel with a little bit more excitement. I didn't use it for the vocals on this track, because Flo Rida's vocal is just so dynamic as it is."
- Synths: Digidesign D–verb, Roland Dimension D
"Some of the synth tracks had a little bit of hall reverb from the D–verb. 'E/G3' on the 'Teckn' tracks is a gate to filter out some buzz on this track. I also used a lot of Dimension D. I often do this with synthesizers. It's a chorus unit, but I tend to use it to give more space. It opens things up a little bit. I didn't add a lot of EQ on the desk, just some cleaning up at the top end. Synth sounds, especially soft synth sounds, are so good these days that it tends to be more a matter of balancing and making sure they fit with everything else."
"I don't use anything on the stereo bus, but we mixed down from the Neve to an Ampeg half–inch machine, and then back to Pro Tools at 24/96, via Cranesong Hedd A–D converters. The song was mastered from the 24/96 two–track mixdown. The project itself was at 24/44.1. I used 96k for the mix to capture the nuances that came from using the tape. As I said, all my stuff will touch analogue tape at some stage. In that sense I'm very old school."
With only 44 tracks, the 'Low' session is relatively simple in this day and age. Fifteen tracks are instrumental, divided half and half between drums and synths/samples, all played and programmed by DJ Montay. Flo Rida is spread out over 14 tracks and T–Pain over 15, with Flo Rida singing and rapping mostly in the verses and T–Pain doing mostly the main vocal hook. All tracks are neatly grouped in the Pro Tools Edit Window, with the tracks in the track list on the left corresponding to the names in the Edit Window proper. 'Boom' until 'Hitz' are drum tracks; 'synth' until 'oomp' are synth and sample tracks; 'Breakdown lead 1' until 'Whine 6' is Flo Rida, and 'Main Hk1' until 'Hk Hey 4' is T–Pain.
Aux tracks, including 'A', 'B', 'C', and 'Verses', tend to relate to the group of tracks just above them. 'Boom' is an 808, and ':–)' a kick Marasciullo added to give more body to the bass drum. 'Synth' is the main synth line and 'RvEFX' is a reverse effect just before the downbeat at the beginning of the song. 'Hitz' is an orchestral stab. All tracks down from Aux 6 are auxiliaries.
Asking an engineer or mixer about his use of pitch correction can be a bit like mentioning an elephant in the room, but not with an artist like T–Pain, who has made a name out of featuring Auto–Tune on his voice. "T–Pain has been using Auto–Tune for a long time, even before we met," says Marasciullo. "Since we've been working together we've created a bit of a warmer way of applying it, a more vintage–like sound if you want, as opposed to a more digital, brittle sound. T–Pain is a very technical person and a great engineer and he'll already have applied Auto–Tune to his voice in about 50 percent of his tracks, and for the rest I'll add it. In the case of 'Low' he had already added Auto–Tune when I got the files.
"I get hundreds of emails every week to my MySpace page asking me about the AutoTune effect," he adds, and the Internet is chock–a–block with how–to–do–the–T–Pain–effect instructions. Marasciullo doesn't argue with these settings when asked whether they are correct, but stresses that "The reason the Auto–Tune sound works so special on him is because of his voice, because of the tremolo he sings with. That just seems to work magically with Auto–Tune. He also intentionally sings a little bit flat, to get Auto–Tune to grab his voice the way he wants. You still have to know what you're doing from a musical perspective to get the effect to work."