Jump to music example
“Keith Richards said years ago that the Stones had a half-hour rule in the studio,” says Miami-based producer/engineer/mixer Doc Wiley. “If they couldn’t get a song poppin’ in 30 minutes, they moved on to the next song. To capture the best vibe of a band’s performance, I’ve adopted that same rule for engineering 30-hour albums.”
Thirty hours spread over a week to track, edit, mix, and master a CD’s worth of songs? Brian Wilson and the Beach Boys once spent a week chasing a kick-drum sound, and Fleetwood Mac spent two years and $1 million over-perfecting Tusk. But those days are over. Just ask Shufly, a new hard rock band Wiley recorded last month. Wiley’s 30-hour album process preserved the band’s true rock roots, overcoming their label’s misguided desire for Wiley to spend a month turning them into the next Dave Matthews Band.
“I took a page from the first Led Zeppelin and Black Sabbath albums, which were recorded in a day,” Wiley explains. “There’s that great Joni Mitchell quote, too, that no one would ever have asked Van Gogh to repaint Sunflowers, yet musicians are always expected to redo their parts. Nothing is more of a vibe-killer than overdoing each track. Once a performance communicates what a musician is trying to say—and it’s almost always in their first take—it’s time to move on.”
Ironically, Wiley is also a Digidesign-certified Pro Tools instructor who has polished tracks to perfection for 20 years. He has worked with U2, Prince, Ricky Martin, Christina Aguilera, and Whitney Houston, and often collaborates with producer Kevin “She’kspere” Briggs (Destiny’s Child, TLC, Pink, Backstreet Boys).
Shufly’s independent label may be surprised by the album, but how does the band feel after your 30 hours together?
You know you’ve done well when the band calls up the next day to say, “Dude! We love it!” The band wanted this record to be a sort of demo for them to shop around to the major labels, so they’re excited about how it turned out.
How did you approach those sessions?
We recorded four hours a day for a week. I had them play four or five takes of each song before moving on to the next one. They played to the same click track each time so I could go back in, using the playlist in Pro Tools, to assemble a better-than-average overall performance.
In a typical session it can take a day just to get the drums miked and sounding good. How did you do that so quickly?
I had drums ready to record in 20 minutes. I only used four mics: an old AKG D-12 on the kick, a Shure SM57 for the snare, a Neumann pencil mic for the left overhead, and a Shure SM87 vocal mic for the right.
How did you deal with the bleed between the drums, hi-hat, and cymbals?
I didn’t mic the hi-hat exclusively, so I leaned on the left overhead for my hi-hat sound and just went for a great overall stereo image of the drums. I rolled off the overheads and the low end in the mix to get a perfectly natural-sounding kit. We were all very comfortable with the sound of the kit—and especially with the hi-hat sound.
What about dynamics and signal processing? Didn’t that take a while to fine-tune?
I’m lucky to have worked with [Steely Dan engineer] Roger Nichols and, like him, I’m a big fan of using no EQ or compressors. I just run through the mic preamps into Pro Tools.
Also, I record the overhead mics very soft so that I have lots of transients to work with when I edit the songs before mixing. Because of recording like this, using the 30-hour concept, there’s no time to have the drummer punch in for the inevitable bad cymbal and drum hits. When I’m not manually editing drum tracks, or if there isn’t enough time, I use [Digidesign] SoundReplacer, and that plugin works best when there are a lot of transients for it to see in the audio file. For example, if the drummer’s best take for a verse includes a late tom hit or one that doesn’t sound good, I’ll either replace it with one of his hits I recorded while we were going for sounds during setup—or with a great tom sound from my sample library.
How does SoundReplacer work?
First I solo either the left or right overhead mic, whichever is closest to the tom I want to replace. Then SoundReplacer can isolate that hit by tapping the signal’s transients. Then I separate the regions and just drag the appropriate replacement tom hit from the Regions Bin, which I’ve already loaded in. If I hold down Control as I’m dragging the new tom sample into place, it’ll snap right to wherever I’ve just placed that separated region. The old hit is now replaced with a better one.
How do you apply your 30-hour concept to vocalists?
Once the first takes are recorded, it’s all about the alternate takes and the art of knowing how to ask a vocalist for more. Just like tracking the rhythm instruments, I’ll first record them singing several full passes through the song so they can get their whole idea completely out. A vocal recording engineer will not be able to move on until the vocalist gets out what they believe is their best take. You need to wait until that vision is recorded before you introduce the idea of alternate takes with them.
What do you tell vocalists to help that process?
I’ll say something like, “We’ve got your vision now, but can we get some safeties just in case?” I introduce the idea of alternate takes while acknowledging that they’ve already got it in the can so that they’ll relax more. There is a sort of muscle memory that lulls a vocalist into a repeated physical pattern once they’ve sung a song a few times.
Then, when I’m comping the vocals, I’ll give the producer the straight-ahead “here’s the comp you wanted” version of the lead vocal, and then I’ll do another one using parts of the alternate takes. [Comping, short for compositing, means constructing a new part out of the best sections of multiple recordings.]
What do you look for in alternate takes?
I might ask them to do one where they over-enunciate their words, and then one where they sing real softly for the entire song. I might say something like, “This is the ‘whispering into your lover’s ear’ take.” Very soft. Then I’ll ask them to address specific, short sections or phrases they’ve already done to try it this way or that way.
Can you site a specific example?
Recently there was a song on which the guy had everything together but wasn’t sure what to sing in the outro verses—he was lost there. So, for an alternate take, I had him scream the entire song as loud as he could! It was a very Radiohead-ish song, an “I don’t believe you broke my heart” kind of song. Just drums and bass and big chords.
When we got to the outro section of the song in mixing, I simply took the best of those screams from the first verse and pasted those into the outro. Between his softer verse lines of “No, no, no,” I inserted his screamed “You lied to me!” and “Know this truth” lines, and it resulted in this great, grandiose outro that never existed before. It all came from those alternate takes after the singer got the idea of what he should sound like out of his head in the first couple of takes.
Which vocal version does the producer typically end up using?
The first takes are great, but often it’s the more relaxed alt takes that are chosen. Now, with Whitney’s voice, it’s more like, “Wow, which great take do we throw out in this verse?” Even her outtakes are completely usable because nothing is a throwaway track with her. Even her laughs between verses or chorus takes are just right.
You graduated from the Recording Workshop in 1980 and engineered for a few years before taking a ten-year break from engineering. Why?
Back in the ’80s there were still a lot of engineers who didn’t empower the artists. I actually heard an engineer once refer to musicians as “track killers.” And it was those types of engineers who embraced the early Pro Tools systems as a way to rely less on artists and more on technology to create their idea of perfect songs. That’s when I decided to drop out of engineering and play bass in bands, which I did until 1995 when I joined the staff at South Beach Studios.
Would you say you’re a bridge between those tech-happy engineers and today’s modern creative recordists?
That’s the plan. My dad [well-known jazz bassist Cornell Wiley] knew Miles Davis, and when the first multitrack recorders came out Miles told him, “Wait until we, the creative artists, get ahold of this technology.” That comment has always stuck with me because it also applies so much to when ADATs [inexpensive multitrack digital tape recorders] and the first DAWs [digital audio workstations] came out. The artist-empowerment part of the recording experience in the early ’90s is why I got back into engineering. Suddenly you had Jagged Little Pill recorded on an ADAT and a Mackie [mixer]—very empowering.
What’s another example of artist empowerment you’ve encountered as an engineer?
I did a project for the Japanese market with Sly [Dunbar] and Robbie [Shakespeare] and after it they said, “It’s great when the technology and engineer get out of the way of God.” If we do our jobs right, there’s no impedance to creativity. It’s our job to be their advocate, to clear that path for them so that they can do their thing and find their “lightning in a bottle.” The job of an engineer, I think, is to coax out and manage their beautiful accidents.
“Depending on how meticulous the producer is, there are a couple of ways you can set up a vocal session before the vocalist arrives at the studio,” says Doc Wiley. “Assuming you have the lyric sheet, the first thing to do in your DAW for every session is to put up a set of song markers: ‘Verse 1,’ ‘Verse 2,’ ‘Chorus,’ ‘Bridge,’ etcetera. Let’s say this first set of markers goes from the beginning of the song to marker 10 at the end of the song.”
If the producer is detail-oriented, Wiley then creates and names a second series of song markers based on the first word in each song section. To do this, he first creates a blank MIDI track directly above the song’s guide vocal track. He then puts the MIDI track in record mode and lets it run through the entire length of the song.
“Place that blank MIDI track playlist in Region, click inside the track, separate the Region, and then type in the first line of lyric corresponding to where the words appear in the scratch vocal track through the rest of the song,” he explains. “For instance, if the second verse begins at marker 4 with ‘My life is,’ type those words right into the MIDI track.”
Wiley says that when a vocalist or producer wants to jump back and forth to any place in the song, he simply uses Pro Tools’ Grabber tool and selects the region where they want to record a new take. When he then hits the audio track’s Record button, Pro Tools automatically records during just that particular region.
“Now when they say, ‘Go back to that line, “My life is,”’ you can just go to your memory location, click on it, hit Record, and you’re recording without having to set up a punch-in,” Wiley says. “Now you’ve set up both your song markers and can use the words typed into the blank MIDI track as a guide to make things go faster. It makes you work faster and more professionally, and most of all it keeps the vocalist and producer from waiting for you to find all the places they need to jump around to.”
Although his favorite dynamics processing plugin is McDSP’s Channel G—a collection of Neve, SSL, and other high-end mixer channel-strip emulations—Doc Wiley’s following tip will work well for any signal path, especially when recording in a variety of environments. [A channel strip is the signal processing built into each channel of a mixing console—typically EQ, compression, and gating.]
“I used to have my own little kludged collection of channel-strip plugins for routing my everyday vocals and instruments into Pro Tools,” Wiley recalls. “Now I just use the Channel G and Waves Renaissance DeEsser for everything. I have lots of Channel G presets now that I’ve created—‘Baby Bass,’ ‘Rock Bass,’ ‘Lead Female Vocal,’ etcetera—to speed up the process of hooking up a vocalist or instrumentalist. Those are just jumping-off points which I then tailor for the particular song.”
Wiley burns hundreds of his Channel G presets to a DVD-R disc, which he takes to the various commercial and home studios he works in. “This way, if you’re a freelance engineer like me, you can carry all of your presets with you no matter where you go and can get a bass, or vocal, or overhead mic mix going instantly,” Wiley explains. “Once you’ve addressed the individual issues of each voice, guitar, or instrument you might encounter, you have solved most of what you’ll be doing throughout a given record.”
The following MP3 will open in a new window. Wiley notes that the song, by the rock group Shufly, is “lathered in SoundToys EchoBoy plug-in processing.”
Randy Alberts is an author, musician, and photographer who lives on Lummi Island, Washington.
Return to digitalmedia.oreilly.com
Copyright © 2009 O'Reilly Media, Inc.