What microphones do you use on guitars?
I use Shure 545s on guitars. I use a lot of those dynamic microphones. They kind of have a built-in filter. They have a very limited frequency response compared to a condenser microphone, they don’t go out as far on the top end or down as far on the low end, but they get the timber of the sound, and they leave some space for you. You don’t get a lot of garbage and noise. The low end of a guitar is primarily noise. It’s a Helmholtz resonator and it really rings at one frequency mostly. Although the strings change pitch, the resonator itself is designed to ring at a number of different frequencies, but it will have a predominantly strong sound in the low registers somewhere. It becomes like a hum, it just goes on and on right through the music. I’m talking about really low.
And of course on the top end of a guitar, if you use a condenser microphone, and mike it up close, you just end up with finger squeaks and all kinds of ugly picking noises. Condenser microphones are nice if you’re recording (Andrés) Segovia in a big hall and you want to get away from the instrument, so you get a little sound of the hall. In that case, a condenser mike is really nice. The frequency characteristic of a condenser mike doesn’t change as you get away from the sound source. A ribbon mike like a 77, as you get closer to the source, you pick up more bass, which is groovy because you can work very tight on a voice with that mike and get a very warm sound. You can put something in the voice that just isn’t there. Once again, you’re trying to create something that’s pleasing to listen to.
How do you do horns and strings overdubbing sessions?
On “Aquarius” I did the horns and strings at the same time. Fifteen strings and eight horns and two French horns and two piccolos. You physically have to have the space to spread out. And their spreading out the musicians is not as much of a problem because everyone is listening to the rhythm section on earphones, and actually in those kinds of overdubbing sessions, you spread the people out so they can hear themselves. If the string players are close to the brass, they can’t hear themselves, because the brass overpowers their own ears, so they play out of tune.
So if you can get people to spread out so they can hear themselves, or if you can record them separately, they’ll play much better. There are times, if you are making a really funky R&B kind of thing, that the horns among themselves are not in tune. Many times if you have the standard R&B horn section, which is a couple of trumpets and a trombone and two saxophones, the saxes are always out of tune with the brass, and it gives it a kind of an edge.
Do you splice voices?
Only lead voices. With background voices, I punch in, because usually with lead voices if you have a lead singer who can really get it on, it isn’t necessary. I usually start with 8 track and do my track editing on 8, and then I usually transfer when it’s finished to 16, in its edited form, so that the 16 has no splices, and then I’ll add the voices by sel-sync-ing the voices in onto the 16. If there’s a lead voice that has to be edited, I’ll take the track and go 8 to 8 and add the lead voice, and then splice the lead vocal pieces together until that’s perfect, and then I’ll put that across to 16, but the background voices I always sel-sync in. If we have problems I’ll punch in different segments. I’ve gotten very good at punching in. On The Mamas and the Papas, we used to punch in 4-bar phrases. The Association was worse. Then the horns and strings or whatever else, the rhythm overdubs, are all sel-synced pieces. So sometimes you actually take it down an extra generation, but it’s worth it if you get a performance, because there again, the generation is not the important thing. The performance is the important thing. The technical thing is just there for the sake of the music.