Historical Analysis and System Reconstruction

The project began with the identification and analysis of the individual elements that formed the Wall of Sound.

Photographs and technical descriptions were used to determine driver types, enclosure dimensions, cluster arrangements, and system layouts. Where dimensions were unavailable, image-based reconstruction techniques were applied using known component sizes as reference points.

The resulting data provided the foundation for recreating the physical structure of the system and defining the parameters required for acoustic modelling.

Creating Acoustic Datasets with EASE SpeakerLab

One of the most demanding stages of the project involved generating the acoustic data required for simulation.

Because measured loudspeaker datasets were unavailable, a comprehensive Excel-based calculation matrix was developed to derive the required parameters from historical information and engineering calculations.

The calculation workflow included:

• Driver specifications and Thiele-Small parameters
• Enclosure dimensions and effective cabinet volumes
• Driver coordinates within each enclosure
• Sound pressure level calculations
• Array summation behaviour
• Directivity balloon calculations
• Horn directivity modelling
• Export parameters for SpeakerLab

These calculations produced the acoustic foundation for the digital loudspeaker models.

The resulting datasets were then transferred into EASE SpeakerLab, where the individual systems were assembled using source definitions, filter groups, crossover structures, enclosure information, and driver configurations.

Particular attention was given to recreating realistic directivity behaviour. Since no measured balloon data existed, the radiation characteristics had to be derived from driver dimensions, horn geometries, crossover frequencies, and physical driver placement within the enclosures. SpeakerLab provided the framework required to combine these calculated datasets into complete loudspeaker models that approximately represented the interaction of multiple drivers within a single system.

The completed models were exported as custom Generic Loudspeaker Library (GLL) files.

The generated GLL files became the foundation for all subsequent simulations.

Beyond enabling the reconstruction itself, they also created reusable digital representations of loudspeaker systems that previously existed only through historical documentation. Future researchers and educators can use these datasets without repeating the entire modelling process.

The project demonstrates how GLLs can function not only as a manufacturer data format but also as a practical tool for preserving acoustic knowledge.

Reconstructing the Wall of Sound in EASE 5

After creating the custom loudspeaker datasets, the complete Wall of Sound was reconstructed in EASE 5.

An open-air simulation environment based on the Hollywood Bowl deployment was created, including a realistic audience area to account for frequency-dependent absorption and ground reflections.

The model incorporated the individual vocal, piano, bass, guitar, drum, and kick drum systems using the custom GLL files generated during the previous stages.

This virtual reconstruction enabled detailed analysis of:

• Coverage behaviour
• Sound pressure level distribution
• Directivity
• Loudspeaker interaction
• Long-distance performance
• Speech intelligibility

In addition to conventional acoustic analysis, EASE 5 was used to generate binaural impulse responses (BIRs) for selected audience positions. A BIR describes how sound from a specific location reaches a listener's ears, including the acoustic effects of the environment and the acoustic filtering caused by the listener's anatomy. For this study, the BIRs were calculated using the head-related transfer functions (HRTFs) of a KEMAR dummy head, enabling realistic binaural reproduction of the simulated sound field through headphones.

From Simulation to Listening Experience

Unlike conventional PA systems, the Wall of Sound was built around independent loudspeaker systems for individual instruments and vocals. Preserving this concept required a different approach than traditional single-source auralization.

Historical recordings from the Grateful Dead were therefore separated into instrument-specific stems corresponding to the original Wall of Sound subsystems. Individual signal groups were created for vocals, bass, guitars, piano, drums, and kick drum.

Within EASE 5, separate BRIRs were generated for the individual loudspeaker subsystems and selected listening positions. Instead of processing a complete mix through a single impulse response, each instrument group was assigned to the corresponding subsystem response.

The BRIRs were exported and imported into Steinberg's Cubase 13 Pro, where convolution processing was performed using the REVerence convolution reverb engine. Each stem was processed with its associated subsystem response before all processed signals were recombined into a final binaural mix.

This approach preserved the original signal-routing philosophy of the Wall of Sound and enabled the reconstruction to reflect the system's defining characteristic: the acoustic separation of individual musical sources through dedicated loudspeaker systems.

Evaluating a Historic Sound System

The completed reconstruction enabled objective analysis of a sound reinforcement system that had previously been understood primarily through historical accounts.

The simulations confirmed several characteristics frequently associated with the Wall of Sound:

• Strong source separation between musical elements
• Effective long-distance coverage
• High directivity resulting from extensive loudspeaker arrays
• Consistent performance across large audience areas
• Good speech intelligibility within the dedicated vocal system

The results support many of the claims made by engineers and audience members who experienced the original system.

Listening to the Wall of Sound

The binaural renderings provided additional insight into the listening experience.

Rather than behaving like a conventional stereo mix, the reconstructed audio often appeared as a collection of overlapping but individually identifiable sound sources. Depending on listening position, the balance and localisation of individual instruments changed in a way that reflected the physical arrangement of the loudspeaker subsystems.

This behaviour is consistent with the original design philosophy of the Wall of Sound and illustrates how strongly the listening experience was influenced by the placement of independent source systems.

The listening tests also revealed that positional differences were most noticeable at closer audience locations, while the overall sound became more uniform with increasing distance from the stage. Vocals remained intelligible across the listening positions evaluated, while bass energy played a dominant role in the overall sonic impression.

Preserving Acoustic Heritage

Beyond the technical findings, the project demonstrates a new application for acoustic simulation technology.

By combining historical research, engineering calculations, EASE SpeakerLab, custom GLL development, EASE 5 simulation, and binaural auralization, a sound reinforcement system that no longer exists in physical form was transformed into a reusable digital acoustic archive.

The methodology developed during the project can potentially be applied to other historically significant sound systems, helping preserve engineering knowledge, design concepts, and listening experiences that would otherwise be lost over time.

In doing so, the project highlights how modern simulation tools can contribute not only to the design of future sound systems, but also to the preservation of professional audio history.