Andrew Scheps has worked on mega-hit albums for a who’s-who of superstar artists such as Red Hot Chili Peppers, Metallica, U2, Justin Timberlake, Jay-Z, the Rolling Stones, Linkin Park, Jewel, Neil Diamond, and Adele. We had an excellent chat that was included in the latest edition of my Mixing Engineer’s Handbook, and here’s part of that interview where Andrew talks about the inner workings of a mix.
Where do you build your mix from?
It depends. I’d love to say that I always build it from the vocal, but usually what I’ll do is deal with the drums to get them to act like one fader’s worth of stuff instead of 20 or whatever it is. Once I’ve gone through that process that I just described, everything seems to come up at once. I’ll have listened to vocal and the background vocals and know exactly where they are, but I’ll get the band to work without the vocals first, which I know a lot of people don’t think is a good idea.
I think it’s the same thing when you’re working on a particular instrument in solo. After 20 years, my brain sometimes unconsciously knows what an instrument will sound like soloed, so I’ll tend to get the tone on things separately, and then it’s all about the balance. I almost never have to go back and change things once I get the vocals in. My brain seems to know what that balance is going be when the vocals are inserted.
What do you use parallel processing on?
On this mix right now there’s a parallel compressor on the kick and snare, then there’s another just on the snare. There’s a stereo one on the toms and overheads, a mono one on just the dirty bass (this song has three basses), a stereo one on the guitar and vocals, and then a couple of different ones just for the lead vocal—one that’s sort of spitty and grainy and one that’s sort of fat. That changes from mix to mix. In fact, it changes a lot.
Are you tucking the parallel processed channel just underneath the unaffected one?
Yeah, although sometimes the parallel one ends up being pretty loud, in which case it’s almost like using an insert compressor, but it’s across a few things. Sometimes it’s just tucking it in to add power or weight.
Do you EQ the parallel processing as well?
Not much, because everything is post-fader, so it’s only the EQ’d stuff that gets in there. Sometimes on the drums that will start to really bring out a badly ringing cymbal, so I’ll go in and do something surgical to fix the problem, but most of the time it’s the same tone as the uncompressed.
Are you using buss compression as well?
Yeah. I used to never compress the mix because I could never find one that I liked that didn’t take away from everything else that was going on. Then a couple of years ago I started using the 2264s [the onboard Neve compressor modules] that are in the console. I also learned the lesson that if you’re going to compress the mix, you begin your mix with the compressor already on. You don’t get your mix and then put your compressor on, because that doesn’t work. You have to mix to it.
I don’t use heavy compression, though. I don’t think I ever add more than 3 or 4dB, so I’m not really smashing it. It is a pretty aggressive setting, though, like a super-fast release. I’ve tried printing uncompressed mixes and bringing those to mastering, but you can never re-create the sound, so I always mix to it now.
How long does it take you to do a mix?
It really depends upon the material. If it’s well recorded and I’ve already done a song for the album, then it can come together in as little as three or four hours. The first one on an album usually takes longer because there’s a lot to sort out, so it will usually take a full day. I’m usually ready to play something for the band by late afternoon at the latest.
How loud do you monitor?
When I’m getting the mix together, I monitor pretty loud and for longer than I probably should. Once I get the balance, then I mix really quietly, but still occasionally check things loud every once in a while. Mixing is such an emotional thing. You’re trying to get it to seem exciting, especially on the rock stuff, so you have to hear it loud to know that the kick and the snare and vocal are hitting you in the chest.
There are things that you can’t judge when it’s loud, though. You can’t judge the vocal level properly because the vocal will sink into the mix more when it’s loud, but in terms of impact and emotion, you’ve got to crank it.
Do you ride the rhythm section for fills or is your mix fairly static?
I still do ride things because the compressors are sucking out some of the natural dynamics of how the instruments were played, especially on some of the louder rock stuff. I’m only adding it back in, rather than creating something out of nothing. I don’t turn the whole mix up at every chorus, for instance. Some people can do that with success, but I always hear it when I do it. I do definitely push the drums for the downbeat of the chorus and really try to accentuate anything that might be cool in a guitar performance, as well as some of the idiosyncrasies. The rides aren’t drastic, but most things are moving.
You can read more from The Mixing Engineer’s Handbook and my other books on the excerpt section of bobbyowsinski.com.