BBCVerb · Maida Vale 1

How an orchestra is recorded, and what each microphone in Maida Vale Studio 1 is for. You can hear most of it for yourself on The plugin.

Recording an orchestra

An orchestral recording is usually built from layers of microphones, which the engineer balances.

The main pickup. A tree or pair of microphones above and behind the conductor hears the whole orchestra at once, as the players and the hall balance it. On most orchestral recordings it gives most of the sound.
Outriggers. A wider pair out to the sides. They widen the orchestra and add the room's spaciousness at the sides.
Spot microphones. Close to a section or a soloist, mixed in quietly under the main pickup to add definition. Maida Vale 1 has one close on the double basses.
Room microphones. Placed to hear the hall's reverberation, with little direct sound. In Maida Vale 1 a pair of Neumann TLM 170s faces the ceiling.

The further a microphone is from a player, the less of the player's direct sound it hears, and the more the room takes over. That balance, direct sound against reverberation, is much of what makes a recording sound close or distant. What is an impulse response? →

How we hear distance and direction

Distance: the direct-to-reverberant ratio. A player's direct sound falls by about 6 dB each time the distance doubles, while the room's reverberation stays at much the same level everywhere. The further away the player, the more room you hear relative to the player, and your ear reads that ratio as distance.
Direction: time and level. A sound from your left reaches your left ear first, by up to about 0.7 ms, and louder, because your head shadows the right ear, most of all at high frequencies. Your brain combines the two, and takes the direction from the sound that arrives first, even when reflections follow.
In a recording, the microphones stand in for your ears. Spaced omnis like the Decca tree give direction mainly through time differences between the channels, cardioid pairs mainly through level, and ORTF pairs through both.

Why a DPA 4006TL Decca tree

A little history. The Decca tree was developed at Decca Records in the early 1950s, by engineers including Roy Wallace and Arthur Haddy, for the company's first stereo orchestral recordings. It originally carried three Neumann M50s: omnis with the capsule set in a small sphere, which lifts the top end for sound arriving from a distance, the same job the diffuse-field grid does on the 4006TL. Engineers such as Kenneth Wilkinson made it the sound of Decca's orchestral catalogue, and it is still the standard main pickup for orchestras and film scores.

Omnidirectional capsules. A pressure omni hears the whole orchestra with an even, full low end at any distance. It has no bass roll-off as players get further away, and no off-axis colouring of the players at the edges of the stage.
Wide response, low noise. DPA's omnis are flat from the deepest bass to the top of the audio band and have very low self-noise. That matters most for impulse responses: the last part of a reverb tail is very quiet, and it has to stay well above the microphone's own hiss to be captured cleanly.
Diffuse-field grids. At the tree, most of the sound from the back of the orchestra arrives as reflected sound from every direction. The diffuse-field grid lifts the high frequencies of exactly that sound, so the brass and timpani, 15 m away, stay bright and present instead of dull.
Wide spacing. Omnis hear the same level from every direction, so their stereo image comes from differences in arrival time rather than level. They must be spaced much wider than directional pairs (a Decca tree's outer mics are typically about two metres apart) to decorrelate the signals and make real width, especially in the low end, where closely spaced omnis collapse towards mono.
The centre mic. Set forward of the outer pair, it fills the middle and steadies the image. In these renders it sits 3 dB below the outer pair.

The other microphones in Maida Vale 1

Every chair's impulse response was recorded through all of these at once, at 96 kHz. This app renders the DPA tree; every array is in the BBCVerb VST plugin.

A Schoeps tree. MK 2H omnis, voiced for distance, as the outer pair, with an MK 21 wide cardioid in the centre: the Decca tree's idea, with a different sound.
ORTF pairs of Schoeps MK 4 and DPA 4011 cardioids: 17 cm apart with 110° between them, a layout named after the French broadcaster ORTF. Their stereo comes from both level and time differences.
A Schoeps MK 22 pair. Open cardioids, between an omni's openness and a cardioid's focus.
Neumann TLM 170 hypercardioids, facing the ceiling. A room pair: the hall's reverberation, with little direct sound.
A Schoeps ORTF-3D array and an Eigenmike 64, for surround and immersive sound. The Eigenmike's 64 capsules sit on a sphere and are processed into sixth-order Ambisonics, which describes sound from every direction and can be decoded for any loudspeaker layout or for headphones.

Hear it for yourself

On The plugin, each chair plays its own player in Beethoven's Eighth.

Dry and room. Click a chair, then switch between Dry · anechoic and In Maida Vale · BBCVerb: the same player, recorded with no room at all, then placed in Maida Vale 1.
Near and far. Click a first violin at the front, then Timpani 1 at the back: the timpani has less direct sound and more of the room.
Left and right. Click Violin I 1.1, then Double Bass 2.1. With omnis this far apart, much of the sense of side comes from which microphone the sound reaches first.
The whole orchestra. Press All 53 players: every part in its own chair, adding up in the room as the orchestra does on stage.

Click a chair to hear its player in Beethoven 8; click more to build the orchestra.

DPA 4006TL tree · L / C / R · centre −3 dB Other microphones shown for reference.

Why a Decca tree, and what the other microphones are for: learn about orchestral recording →

Anechoic recording: Beethoven, Symphony No. 8 in F major, Op. 93, I · Orchester Wiener Akademie, cond. Martin Haselböck · Böhm, Ackermann & Weinzierl (2018), TU Berlin, doi:10.14279/depositonce-6729 · CC BY 4.0 · an 8-bar excerpt, looped, each part rendered through BBCVerb.

Place your audio in Maida Vale 1

Choose dry audio and a chair, preview the room, then add the wet recording to your Audiotool session.

IR renderer

The renderer keeps the IR bank and processes your dry audio. Rendering needs your Audiotool account, connected below.

Renderer address

No IR bank connected yet.

Audio source

Project sample

Processes the complete source sample, before timeline edits or mixer effects. For an instrument track or an edited passage, export a dry stem from Audiotool and choose that file. Mono/stereo, up to 10 minutes and 96 MB for cloud renders; converted to 48 kHz when needed.

Chair Microphones

Connect the renderer to load its available chairs and microphone choices.

Connect your Audiotool account to read one of your sessions and place its audio in the room. You are sent to audiotool.com to approve it — no account here, no password shared, and you can revoke it at any time.

Maida Vale 1

Local recordings are offline

Place your audio in Maida Vale 1

Wet preview

Session

Device graph