By Bryan W Cole / FlatPanel Audio 

Why are there no EASE models for FlatPanel Audio DML Products?

This is an excellent question, and it is one that we receive fairly frequently at FlatPanel Audio. It is an answer that does require some explanation, which we will endeavor to do here. You may find it surprising, both in the "why," and also in the “what” it can mean to you, the audio system designer. Please know that a DML is a very different type of loudspeaker, generating sound in very different ways than we have grown accustomed to over the last 100 years of point-source loudspeaker history.

Important Note: this is a simplified explanation of this topic - a summary, if you will. To substantiate this topic in full would result in a lengthy scientific paper, and would involve a lot of math! We feel that the concepts presented are true and can be substantiated experientially, but this document is not intended to support a deep dive into the science.

 

What is EASE?

EASE is a very powerful design tool that is used extensively in sound system
engineering. EASE was created for designers to have a way to model point source
loudspeaker direct field coverage, as well as interactions with room boundaries (walls and surfaces.) EASE does this by using very accurate loudspeaker models based on phase coherent point-source loudspeaker polar measurements. EASE then models the effects of the given loudspeaker into the subject space, helping the designer to determine the best loudspeaker placement, room treatments, and the resulting audio capability, including intelligibility, RTA time, etc.

 

How does EASE do what it does?

EASE uses complex math to calculate the effects of introducing sound from a point-source loudspeaker in a space - both the direct sound, and reflections as a result of boundary (surfaces) interactions in the space. These extensive math formulas are based on a fundamental understanding that point-source loudspeakers are “correlated,” “phase-coherent” radiators. In other words, there is a predictable, modelable relationship between the sounds radiating from a point-source loudspeaker.

 

How is EASE Used?

Due to the correlated nature of sound radiating from a point-source loudspeaker, the boundary interactions within a space can be modeled with EASE, revealing the summation and/or cancellation of the original sound source that results from those interactions. Managing those boundary interactions (and their negative artifacts) is one of the biggest challenges of using point-source loudspeakers in rooms. Because these radiations are correlated in nature, and correlated sound sources interact negatively with boundaries, we can conclude that room interactions from point-source loudspeakers are often “destructive” to the original intended sound. Peter Mapp states it in this way: “It is well documented that specular phase-coherent radiators are subjected to destructive interference with their boundaries. Adverse boundary interaction is probably the single weakest attribute in the traditional types of loudspeakers in real-world environments.”*

 

Why are there no DML EASE loudspeaker models?

Remember that sound radiating from a point-source loudspeaker is correlated in nature. In other words - sound emanating from that point-source is predictable, and because it can be predicted, it can be modeled with mathematical formulas. Unlike point-source loudspeakers, DMLs are uncorrelated sound radiators2 - in other words, there is no predictable relationship between various sounds radiating from a DML. Uncorrelated sound is said to be diffuse in nature, or, using a broad generalization, the radiations are random, and not related to each other with regard to phase. Since EASE requires mathematics, and thus far there are no appropriate formulas to approximate diffuse sound sources, EASE can not model DML radiators. This would require new math, and modifications to EASE itself - confirmed by Stefan Fiestel at AMFG (makers of EASE) when he states, “transfer function math cannot predict uncorrelated sources.” Think of it this way: to model DMLs, EASE and the associated math would need to know how to model “randomness.” Unfortunately, using EASE to model room loudspeaker room interactions, at this point produces data that is of no real value. One could make an argument that such modeling is not necessary, due to the very characteristics of a DML.

 

So... Why do DMLs work so well, even in highly reverberant spaces?

The diffuse nature of sounds from a DML, as they interact with boundaries in a space, are non destructive in nature to the original source. This means that when the uncorrelated radiation from a DML meets a boundary, there is no significant summation and cancellation as would be experienced with a point-source. This results in the audio content being much truer to the original source. Also, due to the lack of summing and cancellation, DMLs experience less “hot spots'' within rooms. This is all in addition to many other DML benefits - such as 8 octaves of coverage without a crossover, nearly even dispersion throughout the band, true stereo direct soundfield, and intrinsic ability to scale to larger rooms.

 

What are we to take away from this?

Many will agree that a goal with most point source loudspeaker installations is to keep the direct soundfield pointed as much to the listener as possible, allowing as little of the direct sound to reach the boundaries, minimizing destructive interaction. EASE is an excellent tool to help system designers visualize room interaction, helping with precise placement and aiming of point-source loudspeakers to overcome their limitations.. You should also now be aware that a DML is not a phase coherent audio radiator, avoiding many of the shortcomings of point-source loudspeakers in the presence of boundaries. It is clear that one of the main reasons EASE exists as a tool is to help designers minimize destructive room interactions. Since DMLs interaction with boundaries are mostly non-destructive in nature, could we then conclude that, at least when it comes to measuring boundary interaction, such analysis is not necessary with DMLs? “ Science, as well as your own ears will help confirm this assertion..

 

Does this mean designers don’t need EASE anymore?


It is important that we do not draw this conclusion. EASE remains a very useful tool, especially in rooms where the standards are high for system design. While a DML can provide an above-average audio solution to a very difficult space, the better the space (from an acoustic perspective) the better the result will be from a DML. We understand you will have needs for low cost, quick solutions where a DML can be a problem solver and you may also have higher-end installations where the venue would like full analysis and treatment of the room to tame a long reverb time. While the DML will function better than most in an acoustically marginal room, the same DML will bring an even better result in a room that has had some attention paid to acoustics. And to do this kind of work, EASE is a critically important tool to assist. While DMLs will overcome many issues in tough rooms, DMLs will bring some of the best sonic results possible to a well-behaved room.

Audio System Design - Simplified!

Audio system design has now become much more simple. Drastically reduced negative interactions with room boundaries, and surfaces, with fantastic intelligibility without spending large sums on acoustic treatments. And if you have a well behaved room, be prepared to experience what many have called "studio monitors at live-sound SPL." No complex software is required for designing DML systems - only a basic understanding of DML coverage (currently 165 degrees conical on DML500) and SPL requirements for the venue. Bad rooms can still have fantastic clarity (even with long reverberation tails) - and you can setup a FlatPanel Audio Demo System in the venue, and deliver an extremely true demonstration of what the final product will sound like to your customer.

 

If you have not auditioned the DML500, please contact us. We would love for you to experience our technology - it will ground the many assertions we are making above - that can seem rather far-fetched when considered from a point-source traditional understanding. Visit please fill out our Demo Request Form here.

______________

*Azima and Mapp, Diffuse Field Distributed-Mode Mode Radiators and Their Associated Early Reflections, page 1

**Phase-coherence is not required for sound intelligibility. It is required for localization - sensing where in space a sound is coming from. The initial impulse of a DML sound is in fact “pistonic” (like a piston, or traditional loudspeaker.) This initial pistonic behavior helps the brain, through the ‘precedence effect’ identify the source of the sound (even though the “tail” of the DML’s sound is not phase coherent.)