Audio Engineering Society

Convention E-Brief 1
Presented at the 130th Convention 2011 May 1316 London, UK

Queen Marys Media & Arts Technology Studios Audio System Design
Martin J. Morrell1 , Christopher A. Harte1 , Joshua D. Reiss1
1

Queen Mary University of London, Centre for Digital Music, School of Electronic Engineering and Computer Science, Mile End Road, London, UK

Correspondence should be addressed to Martin J. Morrell ([email protected]) ABSTRACT The Media and Arts Technology Studios is a new multimedia facility at Queen Mary University of London comprising three spaces: an acoustically isolated Listening Room, a studio Control Room and a large Performance Lab. This engineering report discusses our design philosophy for our given brief to create a world-class audio recording/playback facility for a space that is a blank canvas for researchers. We detail considerations for making an audio system that is simple to use for standard recording and playback whilst at the same time having a tremendous amount of routing options between the connected spaces for users to create unique projects. The result is a 96kHz/24bit MADI based system using a multimode bre optic network and dedicated wordclock throughout. The system features two separate spatial audio reproduction systems and 5.1 monitoring.

1. THE STUDIOS Queen Marys Media & Arts Technology Studios are a new dedicated multimedia facility comprising three separate rooms in the Engineering building at the universitys Mile End campus. The Listening Room is an acoustically isolated space that originally housed an anechoic chamber when the building was rst built. The chamber fell into disuse in the ninties and was converted to a quiet room with a short reverberation time in 2007 for experiments and recordings. The Performance Lab and Control Room have recently been built next to the listening room in a space that used to be a large open plan computing lab area. The Performance Lab is a black box space for qMedia research groups. In the context of this brief, it can be used as a larger recording space and also and has a Multi-Channel Arbitrary Reproduction System (MARS). The Control Room provides facilities to record and monitor audio from the other rooms. All the cabling that connects the spaces is patched to a small plant room located in between the Control Room and Performance Lab (a oor plan of the new facility is shown in gure 1). This engineering brief deals only with the audio system connectivity and related uses of the studios.

Other uses include lighting, motion capture and other arts research but that is not the authors focus. 1.1. Brief & Design Considerations The brief given for the audio system design was to provide a world class audio recording/playback facility. Given this brief, a set of use cases were decided upon and oered for review to members of the Centre for Digital Music research group and other research groups of qMedia who will use the studios. The use cases related to audio are: 1. Multitrack Audio Recording of a Band Musicians perform in the Listening Room and are recorded by engineers in the Control Room. This will require high quality multichannel audio recording from Listening Room to Control Room and return channels from Control Room to Listening Room for musicians headphones and playback across monitors. Bidirectional MIDI connections may be required where electronic instruments are used. Twoway audio/video communication for talk-back purposes is also required. 2. Multitrack Recording of a Large Ensemble (e.g. Classical Chamber Group)

AES 130th Convention, London, UK, 2011 May 1316 1

Morrell et al.

Queen Marys Media and Arts Technology Studios Audio System Design

Foyer

Corridor
network/fiber

Control Room Listening Room Plant Room Performance Lab

network/fiber network/fiber

Fig. 1: Media & Arts Technology Studios Floor Plan. If the Listening Room is not big enough or the performers require a space that is slightly more live acoustically then audio recordings may be recorded from the Performance Lab. Technical requirements are similar to item 1. 3. Simultaneous Audio Recording from Performance Lab and Listening Room Where isolation of a particular performer or instrument is required (such as drums or very loud amplied electric instruments), audio from both the Listening Room and Performance Lab might be recorded in the Control Room simultaneously. 4. Live Performance of a Band with Simultaneous Video Projection A band might perform in the Performance Lab while audio processing and video projection are controlled either from within the space or from the Control Room. 5. Performance Rehearsals Groups or individuals may wish to use the spaces for rehearsing music or other types of performance. For rehearsals requiring music playback or band rehearsals requiring amplied instruments or vocals, the Listening Room and Performance Lab would need PA speakers and associated audio equipment (microphones, stands, cables, mixer etc). 6. Ambisonics Experiments The listening room contains a 16 speaker ambisonics system. This may be controlled from the Control Room audio equipment or from within the Listening Room if the user has a laptop with the correct audio connections. 7. Transducer Array Experiments Multichannel audio input/output and large arrays of microphones or speakers can be set up in the Listening Room or Performance Lab and measured with equipment in the Control Room (or in the same space if necessary). It is clear that the audio system and routing needs to be very exible for accomplishing many tasks. However, experience teaches us that systems also have to be simple and accessible to all without taking a long time to setup if they are to be truly useful. With this in mind, careful thought was given to the wiring of equipment; it was made a priority that all equipment be accessible via patchbays with half-normalled connections so that for standard use nothing needed to be patched in. One nal point was that a clear and concise manual with a guides section needed to be written. This hopefully goes some way to addressing the problem of only a handful of researchers knowing how to use the audio system. 2. SYSTEM CONNECTIVITY As discussed in Section 1, the studio spaces have been built in converted areas of the old Engineering

AES 130th Convention, London, UK, 2011 May 1316 Page 2 of 4

Morrell et al.

Queen Marys Media and Arts Technology Studios Audio System Design

building. Due to the nature of the Listening Rooms room-within-a-room architecture inherited from its previous use as an anechoic chamber, the cable runs between that space and the Plant Room are very long and in many places quite narrow. This meant that it was unfeasible to use multicore analogue cabling to connect the spaces for recording audio. Instead we had to make a choice between ethernet and optical bre for our audio routing between the rooms. We eventually chose to use a MADI [1] audio system over multimode bre optic cable because the MARS system mentioned in Section 1 was already MADI compatible. MADI oers 64 input and output channels per duplex cable at 44.1/48kHz, but the authors chose to specify the system with all the features to remain at up to 96kHz, limited by the A/D D/A converters used. To give maximum exibility of routing between the spaces, an SSL MADI X8 unit is used which allows routing and duplication of single MADI channels so audio can be sent to multiple locations. Although the unit will be left with a setup default preset in the majority of cases, it oers a huge amount of exibility in the system that wouldnt be possible with an analogue wiring system. One downside to MADI is that external wordclock [2] is recommended due to the high jitter rate of the embedded wordclock. The studios therefore have coaxial wordclock cables throughout linked to a master clock in the Control Room. 2.1. Listening Room The Listening Room has a 24 channel SSL MADIAX converter which has 20 channels of preamps connected to the A/D. Sixteen of the preamp channels are from the same model of 8 channel Focusrite ISA828 preamp so that the preamps and converter sections are the same. When carrying out research recordings the transfer functions of the preamps should all be identical and thus become negligible within an experiment. The preamp choice was based on the wide and at frequency response oered by the units. The nal 4 channels of preamp is from a Focusrite Red 1. On the D/A side of the converters are PA speakers, Stereo Headphone Mix, 4 individual headphone sends and the 16-speaker spatial audio system. All this connectivity is accomplished over one bidirectional optical bre connection. There is also a small analogue mixer connected to the PA speakers for use as a rehearsal room. This

setup allows brief items 1, 5, 6 and contributes towards item 3. 2.2. Performance Lab The audio system is specied for using two bre connections as part of the MADI network in this room. This is for two 24-channel SSL MADI-AX converters that are part of the MARS rack unit. The inputs to the converters are made using a Yamaha 01V96 digital mixing console via optical ADAT outputs to provide 16 microphone preampliers for use as stated in brief item 2. The digital mixer is also connected to PA speakers for in-room monitoring matching brief item 4. The outputs of the converters are used to control up to a 40.1 speaker system via the MARS for spatial audio reproduction of ambisonics, vector base amplitude panning and wave eld synthesis to name the main spatial audio types matching brief item 2. 2.3. Control Room At the centre of the Control Room recording setup is a MacPro computer DAW which has three MADI connections from the Plant Room for connecting to the audio system. We use 16-channel Apogee A/D and D/A converters in this room for monitoring and recording. The master word clock unit is also present in this room to allow user control of the audio system sample rate. Monitoring is provided by a PMC AML2 and SB-100AP 5.1 surround system and alternate Yamaha HS50M reference speakers all connected to an Audient ASP510 surround monitor controller for level control and source/destination routing. All the devices may be used simultaneously by way of an aggregate audio driver on the computer. This room is used for all of the brief items listed in section 1.1. 3. CONTROL ROOM CALIBRATION Calibration of the Control Rooms main 5.1 system is essential for research and teaching purposes. The Audient ASP510 monitor control system allows calibration using the inbuilt pink noise generator. A REF button has a user-dened level which is set in conjunction with individual channel trims to make each speaker sound pressure level equal [3]. The use of this means that inconsistencies in listening experiments carried out over multiple sessions are eliminated due to a predened level being used. For teaching, students can be given a specication level

AES 130th Convention, London, UK, 2011 May 1316 Page 3 of 4

Morrell et al.

Queen Marys Media and Arts Technology Studios Audio System Design

that their work will be marked at, with this accessible by a simple button push to access the reference level. This means there will be no dierences between mixing and marking level as a result of listening curves. 4. MIDI CONNECTIVITY Part of the Media & Arts Technology Studios brief given in Section 1.1 included the use of MIDI instruments to connect to the Control Room. The problem arose again of connecting the rooms with MIDI with the routing channels available in the building. It is unfortunate that MADI does not oer MIDI within its specications and neither MIDI or USB cables are specied to the length needed to be used between the studio rooms. To address this problem a USB-to-ethernet converter system is used to provide a transparent connection between a USB MIDI interface and the studio computer. The MIDI interface is stored in a rack bag along with a hardware synth/sampler module for use in rehearsals or connecting to the audio system as described in brief item 1. 5. CULMINATION The authors have described how they successfully met their given and self-assigned briefs laid out in the engineering report and have designed a professional level recording and mixing environment at Queen Mary University of London as part of the Media & Arts Technology Studios. The resulting system is MADI-based, oering a high level of exibility for current and future research experiments and installations.We are especially proud of some of the features such as the professional level calibrated monitoring, multiple headphone mix abilities, high quality preamps and high quality documentation of the system design and usage. We believe that we have designed a system that not only meets our needs but is robust in its operation and is upgradable either via patchbays or through the other bre optic cables that are currently unused. We nally highlight again our approach to make it usable by all audio researchers and other researchers via some basic audio training whilst being a very exible, powerful and capable system if one wants to delve into the advanced setups oered by MADI routing. 6. ACKNOWLEDGEMENTS The authors wish to acknowledge the academics at

the School of Electronic Engineering and Computer Science at Queen Mary for their support, trust and nancing in the audio system described here. We also thank all the researchers who gave their time to help carry equipment. 7. REFERENCES

[1] AES, Aes recommended practice for digital audio engineering serial multichannel audio digital interface (madi), AES Standards, Tech. Rep. AES10-2008, 2008. [2] , Aes recommended practice for digital audio engineering - synchronization of digital audio equipment in studio operations. AES Technical Council, Tech. Rep. AES11-2009, 2009. [3] , Multichannel surround sound systems and operations, AES Technical Council, Tech. Rep. AESTD1001.1.01-10, 2001.

AES 130th Convention, London, UK, 2011 May 1316 Page 4 of 4