Previous posts in this series:

What do young artists and ensembles need to know about recording?

What musicians should know about recording, part 1: Where to record

The Recording Process, Part 2: The Recording Team

Before I get to the body of this post, a quick news item: album downloads are now available for sale on the Cedille Records web site.

Before any music can be recorded in earnest at a recording session, the recording team needs to get a sound balance everyone likes. This often involves the positioning of players as well as of microphones, along with the "mixing" of those microphones. The way this is done is dependent on the size and type of the ensemble being recorded and the room in which the players are playing.

In a smaller room, where sound bounces back off the walls creating lots of "early reflections" it is usually necessary to "mic" (i.e., put a microphone on) each instrument separately. In a larger hall that isn't overly reverberant, it may be possible to record an entire ensemble of different instruments (e.g., strings with piano) with a single pair of microphones because there is a sweet spot in the hall where the sound blend is absolutely perfect with no loss of detail.

This can also have to do with the philosophy of the recording label and its engineers and producers. The audiophile label Dorian used to record everything it could in the Troy Savings Bank Music Hall in Albany, New York, with a single pair of microphones. However, according to artists who worked with them, it could take a long time for the engineers to find the aforementioned "sweet spot," since it differed depending on the instruments and players involved.

Another label renowned for its sound, Reference Recordings, takes the exact opposite approach: they put a separate microphone on each instrument and rely on their mixing ability to get the perfect amount of blend and detail. And yet their recordings sound as lifelike and "natural" as Dorian's.

At Cedille, we have no overarching philosophy. We just try to get the best sound possible given the circumstances we are working in. This week, when we record the William Ferris Chorale in a church, we will need only a single pair of microphones to capture their sound and the ambiance of the space. Recording solo piano in the WFMT performance studio (an excellent Steinway "D" concert grand) finds us using four microphones, however: because of the nature of the room, we use two mics in the middle of the room for our "main" sound but also employ a pair of close-in mics to pick up extra detail that gets lost amid the early reflections off the studio walls.

The more tricky sessions for getting the sound involve multiple players on different instruments that have to be balanced against each other. I will use this kind of session as my example of how the process works.

At the start of the session, the musicians play while microphones are placed. I recommend that the musicians use this time to acclimate themselves to the room (which may be very different from where they've rehearsed or performed the piece in the past) and work on passages that can benefit from extra rehearsal. We often find it necessary to reposition players and frequently have to come up with compromises between the musicians' needs to hear each other and communicate visually, and our need to have them where "their" microphones (in cases where we have to "spot" individual players) can actually pick up their sound without too much of their neighbors' sound leaking in.

Once we get close to the right player and microphone positions, I will ask the musicians to play passages that demonstrate different balances well. When balancing instruments against each other, it is important to ensure that the each player has equal "presence" in the recording: i.e., that each one's sound feels about as close to the listener as the others' (which, depending on the players, instruments, and microphones could mean positioning the microphones very differently with respect to the various instruments). Usually, I want to hear both passages where there are distinct solo lines (fugues are always nice) and those where the players should be perfectly blended (e.g., playing chords together). (Of course, with large ensembles such as orchestras, some instruments (e.g., brasses) actually should sound more distant from the listener than others....)

Some of the things that we are listening for as we position the microphones (and players) and set the (volume) "levels" on the mixing board include: presence and loudness (not the same thing: presence has to do with proximity to the microphone, loudness is how much of that microphone gets mixed into the recording), balance (of instruments against each other) and blend (of the instruments together), imaging (do the instruments sound like they are coming from where they should be on the stage), life-likeness of the sound, range of dynamics (when the players play soft does it sound soft, is there enough (or too much) difference between soft and loud, etc.), and equalization (do some instruments sound overly thin/nasal, or tubby/fat?).

Once all the microphones and players have been properly positioned and the recording engineer has "mixed" the sound to his and my initial satisfaction, we ask the players to play a representative passage for a "sound check" which they will come back into the control room (or at least the area in which the recording team and equipment happen to be) and listen to when they are done playing. Their comments on the sound - how it meets their expectations, what they would like to hear more or less of, etc. - are extremely valuable and often lead to a few rounds of repositioning and/or remixing. Actual recording of "takes" does not begin until everyone is happy and agrees that the sound meets their needs and expectations.

It is extremely important that this be a fully open and collaborative process. Musicians tend to listen for their own instruments in the recording, for obvious reasons. If they don't listen together and come to a mutual agreement you get the story of the producer who got comments on the sound separately from each of the players in a group and then turned to his engineer and said "make everyone louder than everyone else!"

Once we have a basic sound everyone is happy with, we often ask the musicians to give us a "level" - that is, play the loudest passage to be recorded so we can set a proper overall "master" level where the sound doesn't go over a "zero" peak (which can lead to clipping and/or distortion) but where it gets close enough to zero that we are getting the largest possible dynamic range in our recording so even the softest sounds are audible.

Then we get to actually record some music. How we go about that will be the subject of my next post.