Strings:

Coincident or near-coincident stereo omnidirectional microphone arrangements augmented with spot microphones (cardioid) were used to record the violin, viola, cello, double bass, and the entire viol family. In general, we recorded the brighter timbres resulting from playing near the bridge and sound post on the top plate. The lowest fundamental of the double bass was more than an octave below the air resonance of the instrument. As a consequence, sounds in the lowest register of this instrument did not radiate well. Moreover, dominant low frequency radiation from the back plate created a darker timbral colour, except for the area directly around the sound post. The naturally weak projection of the double bass in this important range required additional spot microphones directed at the lower side of the instrument's back plate. In addition, both the cello and double bass were recorded with a close microphone near the bridge. The main stereo pair was centred one meter above the performer. For the cello, double bass, and bass viol, a second stereo pair spaced one meter from the instrument was placed a few centimeters above floor level.

Acoustic guitar, lutes, and harp:

Rather than using stereo microphone arrangements for these instruments, recordings were produced using three microphones all placed within 20 centimeters of the instrument. The guitar and lutes radiated lower frequencies hemispherically. Higher pitches were more prominent between the end of the fingerboard and port, and directly behind the bridge. Microphones were thus placed in these areas. On the other hand, the harp radiated in a more unidirectional manner. In addition, the lowest string was not acoustically supported by the sounding board, and was therefore extremely weak. Omnidirectional microphones were thus placed on either side of the sounding board, and a third microphone was positioned opposite the performer near the lowest string.

Flutes, trumpets, saxophones, and historical wind instruments:

The flute family was recorded using a near coincident pair of microphones. The microphones were situated 50 centimeters above the performer. Because of the bass flute's difficulty in projecting its lowest notes, the pair was moved closer. The trumpets and cornet were primarily recorded with a stereo pair of cardioid microphones. An additional cardioid was placed in front of the bell, while a second was placed directly behind at a distance of 50 centimeters. In its higher registers the trumpet radiated unidirectionally, and so was picked up primarily by the front microphone. Both the alto and tenor saxophones were recorded using a coincident stereo microphone pair placed two meters above the performers. Secondary microphones were placed (1) near the first holes close to the bell and (2) just below the mouthpiece, for recording lower and upper registers respectively. In general, single microphones spaced vertically were used to record most of the historical wind instruments. Distances from instruments varied greatly. Some instruments such as the classical oboe, soprano Baroque and Renaissance recorders, treble cornet and alto shawm had no trouble in projecting their sounds. In these cases microphones were placed at a distance of around 50 centimeters. The great bass, bassinett, and quart recorders were relatively quiet instruments, however, and required close miking (within 10 centimeters of the tone holes).

Pianos and harpsichord:

The pianos on Volume 3 were recorded using two stereo pairs of omnidirectional microphones. The main pair was in a coincident arrangement and was placed approximately three meters away from the mid-section of the piano. This pair captured some of the hall's natural ambience. A second pair quite close to the string crossing added clarity to the attack. For the sustained piano samples on Volume 9, only the close setup was employed. The harpsichord was recorded in the dry studio environment. Two stereo pairs of condenser microphones (cardioid pattern) in near coincident arrangements were used. One was placed close to the sound board near the tail. It picked up much low frequency information. A second pair was placed near the bridge, and was sensitive to the various attacks of the stops. The pairs were mixed so that overall levels for the nasal quality of the 8 foot stop and the muted buff stop were about the same.

Percussion, tympani, vibes, marimba, and accordion:

All of the acoustic snare drum samples were recorded by two cardioid microphones. One was placed 10 centimeters above the skin. Mixed with this main pickup was a second cardioid placed directly under the snares. The phase was electronically reversed then mixed with the top microphone. Cymbals were recorded by two spaced cardioid microphones 50 centimeters above them. In recording the tympani, we used a single stereo pair of omnidirectional microphones placed two meters above the performer. Both the vibraphone and marimba were recorded using three omnidirectional microphones placed equidistantly over the instrument, at a height of 50 centimeters. The treble and bass sections of the accordion were each recorded with a single cardioid microphone, placed at a distance of 10 centimeters from it.

Many of the microphone arrangements used to record the acoustical instruments throughout the MUMS library were based on theoretical and experimental studies conducted by Frank Opolko and discussed in his paper, "A combinative microphone technique using contact and air transducers" (presented at the Audio Engineering Society Convention, Anaheim, 1982 (A.E.S. preprint # 1948 B-3, October, 1982).