sound thoughts
CONVERSATIONS ABOUT SOUND IN OUR LIVES

Why do some concert spaces excite you?

Just seconds after a concert begins, we know whether a performance will be exciting or not. Unfair as it seems, listeners respond intuitively to music long before the artful aspects of performance bring themselves to bear. Very often, that intuitive, emotional response is driven by the acoustical responsiveness of the concert hall. Even the world’s best artists or orchestras cannot overcome the deadening effect of a dull hall.

In music, as in life, first impressions rule. To a music lover, the first few seconds of a performance can conjure a blissful reaction or leave one cold. To most of us, appreciation of artistry or wonderment at virtuosity is not what we crave from music. Even those of us who might rank as connoisseurs are there to feel, not to think, and when we don’t experience what we hope for, we often leave blaming the artists or the art form when it is the hall that failed us.

Trust your own experience.
Have you even noticed while driving home that the music you hear from your car’s audio system sounds more exciting than the live concert you just left?
Such an experience might prompt you to think that you are no expert listener, but what you feel when you listen to music is genuine. You can distinguish when music sounds exciting to you and when it doesn’t, even though you cannot explain why. Take comfort in knowing that you are not alone.

The basics.
Among professionals, there is an understanding that while a musician plays an instrument, a concert artist plays a room. The room is the concert artist’s instrument. When a violin bow scrapes a string, causing it to vibrate, the string’s sound is colored and amplified by the violin which is, in turn, colored and delivered to the listener by the hall. Just as a factory-made violin is no match for a Stradivarius, some halls are no match when compared against others. The hall is that last critical link between the artist’s fingertips and your ears.
With concert hall acoustics, understanding your own emotional response can be traced back to distinct acoustical qualities of the room. The excitement, or lack thereof, can be traced back to the size, shape, and materials used to construct the room. Exciting concert halls are those which exhibit excellent running liveliness. Let’s explore the meaning of this concept – running liveliness – let’s start with the fundamentals.

Within a concert hall, the sound you hear results from of thousands of sound reflections bouncing off every surface within the room. Sound can reflect off dozens of surfaces, but those sound reflections which arrive first and loudest to your ears shape your emotional and physical response to the music.

It is not enough to hear a pin drop.
When the musician on-stage creates the music, the sound travels directly to your ears, but it also reflects from various surfaces all around you in the room: walls, ceiling, balcony fronts, pillars, etc. Sound reflects best off of hard, non-porous surfaces such as plaster and solid wood. Some surfaces, like people and fabrics, diminish or absorb sound reflections. Further, the angle and shape of a reflective surface determines whether sound is reflected to you in one bounce or many more. The proximity of the reflective surface to you determines just how quickly the reflected sound arrives to your ear. It also determines how loud that reflected sound will be when you hear it.

For you to hear subtlety and detail within music, you need strong and direct sound. You also need an abundance of loud sound reflections that arrive quickly – generally within 1/20th of a second. When the path of these sound reflections is sufficient, the concert hall exhibits clarity of sound. Clarity of sound (“you can hear a pin drop”) is important, but clarity, by itself, won’t excite you. The truth is that it is not enough to hear a pin drop. The sound has to be more than clear, it has to be beautiful.

What makes sound beautiful?
To most people, beautiful sound is reverberant. Reverberance is how long it takes sound to disappear after we first hear it. We call sound disappearance the “decay” of sound.
Reverberation time was a hot topic in the acoustics world. Most acousticians would agree that the ideal reverberation time ranges from 2 to 2.25 seconds for symphonic music. Think about this: You only hear the benefits of a 2+ second reverberation time when the music stops! So, reverberance heard when the music stops is desirable, but it is not what separates good concert halls from exciting concert halls.

So where does that excitement come from within great concert halls? It occurs when we can hear the reverberance of the concert hall as the music is ongoing!
For sound clarity, we need strong sound reflections arriving within 1/20th of a second following the creation of a music note while the music is ongoing. However, our ears are capable of gathering much more information from sound reflections that follow different paths within a concert hall from the stage to our ears. We also know that abundant sound is being held by the space because we can hear it linger when the music stops.
So, when the music is ongoing, what is the influence of sound we hear between the early sound reflections needed for clarity and those that we hear reverberate when the music stops?

Our Brain on Music:
Decades ago, audiology (the science of hearing) experts learned that, at every moment, our hearing system collects information for 1/4th to 1/3rd of a second before it becomes overloaded. Our hearing system collects and integrates the sound it hears over this significant fraction of a second, then we use everything we hear to form our overall impression of the sound. Long-lasting sounds are often very pleasing. For example, the sound of an ocean wave is more pleasing than the abrupt honk of a horn.
Our incredible hearing system is capable of collecting information about a musical note even when an additional two or more notes have already sounded. Our hearing system exhibits what could be called “running integration,” using all overlapping sounds to create our experience of the music. Hence, we can simultaneously hear and process the distinctive nature of a single note plus those abundant sound reflections that occur for the first 1/4th of second or longer. The potential for hearing and feeling the excitement within a great concert hall resides in each and every normal listener.

Running Liveliness
This is why some concert halls give us goosebumps and some don’t. Good concert halls offer clarity, but great concert halls allow you to hear the reverberance as the music is ongoing while preserving clarity. In the parlance of acoustics, halls which offer the ability to hear the liveliness of the hall as the music is ongoing exhibit excellent “running liveliness.”

How is running liveliness achieved within a concert hall? Those of us who design these halls must create a room of proper size, shape, and materials to provide both clarity and running liveliness. We must design the space so that a multiplicity of sound reflections arrive at your ears while ensuring that the sound reflections arriving 1/4th of a second or longer after the sound’s creation are strong and audible. When we’re successful, the music will be very exciting to you.

Bookmark and Share

3 Responses

Neill Archer Roan Says:

This is, by far and away, the clearest explanation of a very complex topic that I’ve ever come across. Great work.

Michael Pastreich Says:

Your article is fascinating. It also leaves me a little confused.

I’m trying to understand what Running Liveliness is. I get the feeling it is something significant and is not a concept I have heard anyone else discuss. From what I read in your article, and what I learned at the League of American Orchestras Chicago conference, I get the feeling that Running Liveliness is a combination of two sets of sounds: 1) the sound from the stage and 2) the reverberation.

If I’m correct, a note sounded diminished in a more-or-less straight-line decline, sort of like this:

Then once the sound starts hitting the walls, ceiling, etc, and you get a new sound caller Reverberation, sort of like this:

And Running Liveliness is the combination of these two sounds:

And this sound is what brings the music to life.

Am I correct?

DAvid Griesinger Says:

Direct Sound, Engagement, and Running Reverberance
10/28/09
© David Griesinger

This note is inspired by a blog by Richard Talaske – http://www.talaske.com/?p=100 – where he describes a common acoustic experience:

“Just seconds after a concert begins, we know whether a performance will be exciting or not. Unfair as it seems, listeners respond intuitively to music long before the artful aspects of performance bring themselves to bear. Very often, that intuitive, emotional response is driven by the acoustical responsiveness of the concert hall. Even the world’s best artists or orchestras cannot overcome the deadening effect of a dull hall.”

Bravo! Talaske accurately describes the excitement and immediacy that some halls and nearly all recordings provide, and he is definitely correct in suggesting that this excitement is related to the combination of direct sound, early reflections, and late reverberation. He suggests that sufficient very early reflections – as early as 20ms after the direct sound – are vital.

But the devil is in the details: how many early reflections do we need, and what is their strength and time delay? How much late reverberation, and how strong should it be? What is it about the mix of direct and reflections that delivers both immediacy and running liveness in a recording, and why do so few halls do so over a majority of the seats? (All halls posess a few good seats, and the management steers VIPs and critics to them. Our goal is to get a great sound to at least 60% of the listeners. These halls exist, but are quite rare.)

Hall design is currently an art – not a science. Results from this art are similar to other forms of art; a few disasters, a great many partial successes, and very, very, few masterpieces. To put more science into this art we need accurate models of how human (and animal) neurology processes sound, and a wide body of data on occupied halls and stages. Neurological data is rapidly becoming avilable. Hall data is hard to come by. Hall managers prefer that their acoustics be mysterious and unmeasurable. Even with permission measuremements under occupied conditions are tedious for audiences and musicians. And occupied measurements are vital.

But Talaske points out that recordings – even heard on a car radio after a concert – can seem both more immediately engaging and more reverberant than the experience in a hall. The acoustic properties of recordings can be measured – if a bit of time is taken during the mixing process. And the properties of the electronic acoustic devices used to make them can be known precisely. (The author knows a great deal about such devices.)

If we look closely at differences between the acoustics of sound reproduction and the experience in a hall, we find the major difference is the strength of the direct sound compared to the total strength of all the reflections, both early and late. Recordings engineers start with the direct sound alone, and carefully add reflections and reverberation until the ideal balance is found. The ideal balance of direct and reflected sound and the delay of major reflections can be measured during the recording process by simply recording hand claps at the positions of different musicians, and analyzing the energy balance that ends up in the mix.

That balance invariably has more energy in the direct sound (the component of the mix from the closest microphone) than the sum of all the other components in the mix. Tonmeisters and sound engineers are convinced otherwise, as few bother to make the measurement. But when they do, the direct sound from the closest microphone dominates. In spite of (or more likely because of) the strong direct sound the recording is perceived as both exciting and reverberant. But by necessity in halls the direct sound is nearly always far weaker than the sum of the energy in the reflections. Otherwise the sound would be far too soft in seats distant from the orchestra. When the direct sound is weaker than the reflections it can become quite difficult to generate the excitement Talaske mentions – and surprisingly it is also more difficult to perceive the reverbeance. But it can be done – if the level and time delay of the reflections is favorable.

Even though the direct sound is much stronger than in halls recordings have something to teach us. In the era of the recordings Talaske mentions it was standard practice to move the orchestra onto a platform in the middle of the hall, in order to minimze the strength of early reflections at the small number of microphones employed. In fact, in these recordings reflections that arrive before about 30ms are quite rare. The majority of the reflected energy comes much later, and is relatively low in total energy.

The result is counterintuitive. If we want “running liveness” to be audible, why is it not better to have more reverberation? The answer is that the perception of “running liveness”, or running reverberation, depends both on the audibility of the direct sound and on reflected energy that arrives at least 100ms after the direct sound. This can be demonstrated easily with a dry recording and an electronic reverbeation device (preferrably by Lexicon.) Reflections that arrive earlier are desireable – but when there are too many they mask both the direct sound and the desirable late reverberation.

The time delay of reverbeartion is important – but how can the brain peceive this time delay? Clearly there must be some clear event that marks the start of a sound – be it a note or a phoneme. This event is the arrival of the direct sound at the ear – the “first wavefront”. But when strong reflected sound follows closely after the direct sound the brain is unable to determine just when the sound started. The direct sound simply becomes an inaudible part of the whole envelope of sound. The time of the true beginning of the sound is indeterminite, and it is also impossible to localize the sound through the interaural time delay or amplitude difference.

Ideally the brain separates the direct sound and the running reverberation into separate sound streams; one (or more) for the foreground, and one for the background. The foreground streams contain the notes or phonemes that carry information, the background stream carries the room noise and reverberation. The background stream has some special properties. It is perceived as continuous even though it is frequently masked by the foreground. It is also usually perceived as unusually spacious. It is the background stream that gives the perception of “running liveness”. The separation process is started by the arrival of the first wavefront – the first 20ms or so of sound that arrives before reflections overcome it. When there is sufficient time for direct sound to be analyzed, information is provided about the location and distance of the sound source – even if the reverberant energy which follows is much stronger.

The audibilty of what Talaske calls “running liveness” depends on the success of this separation. Where the direct sound cannot be separately perceived, the forground and background blend into one stream. The direction of which is perceived as frontal and not enveloping. This is why a recording – where the direct sound is stronger than all the reflections combined – can have more running liveness than a rear seat in a concert hall. In a recording the reverberation is easily separated from the foreground stream, and can be separately perceived. The rear seat is reverberant – but not enveloping. And the excitement, or engagement, of the performance is missing.

Perceiving and processing the first wavefront takes time. If strong reflections come too soon, the localization and distance cues are wiped out, and the brain treats the direct sound, early reflections, and reverberation as a single stream. This author and several collaborators are currently gathering neurological data on the time needed for this separation as a function of the direct to reverberant ratio and the initial time delay. The data show that with syllabic sources such as speech most people can accurately localize – and thus separate the direct sound – even when the total energy in reverberation is more than ten times stronger than the direct sound. But for this to be possible there must be at least a 20ms delay before significant reflected energy arrives. The shorter this delay, the stronger the direct sound must be if it is to be separately heard.

If we look at hall data, either the rare actual data from occupied halls and stages, or data from models, we find that this criterion is seldom achieved over a wide range of seats. For example, in a typical shoebox hall the direct sound is too weak in more than 50% of the seats to meet this criterion. This is in part because the direct sound becomes weaker as you move back in the hall, and in part because the delay of the first reflections becomes smaller. In small shoebox halls (under 1500 seats) the time delay is smaller than in large halls, and the reverberation is stronger. Fewer than 30% of the seats meet the criterion. In the remainder of the seats the sound is reverberant, muddy, and not enveloping. A performance can still be enjoyable, but the acoustic excitement mentioned by Talaske is missing.

Some halls do work well. In the Amsterdam Concertgebouw the square audience seating area and the central position of the orchestra brings the average listener closer to the orchestra, while at the same time increasing the time delay of the first reflections. As a result the direct sound is easily separated, and both clarity and running liveness are strong. In Boston Symphony Hall the time delays for the first reflections are shorter than in Amsterdam, but the expected reduction in clarity is mitigated by the effect of the nitches and coffers in the walls and ceiling. These features reduce the strength of the first reflections in the rear of the hall at frequencies around 1000Hz, where most of the excitement of a performance is perceived. A great small hall – such as Jordan Hall in Boston – succeeds by bringing the audience closer to the stage than a shoebox hall, while increasing the late reverberation by raising the ceiling. Musicians on the forestage in Jordan are heard with clarity and reverberation througout the hall. (Musicians deep in the stage house are not so lucky.)

It is current acoustic dogma that strong early lateral reflections are the key to a successful hall, but this is not the case when the reflections come too soon and are too strong. A few great halls – and small halls like Jordan Hall, contradict this dogma. Reflections arriving before 20ms are few and weak. But this is not the case in most halls, particularly smaller ones. Geometric acoustics shows that as a listener moves back into a shoebox hall the sidewall reflections become both stronger relative to the direct sound, and arrive with shorter delay. In London’s famous Wigmore Hall strong prompt lateral reflections in seats more than half way back both blur the sound and cause instruments to appear to come from the side wall. This effect is not typically noticable unless you close your eyes, but it is clearly there. We have found that in a small hall these reflections (and prompt reflections from the stage back wall) can be deadly, because they arrive very closely after the direct sound, and prevent the stream separation process.

In a series of recent experiments in a small hall, blocking these strong reflections by adding a few absorbing panels to the back of the stage and the side walls dramatically increased both the clarity of the music and the sense of the hall, in spite of the slightly reduced reverberation time. The musicians immediately heard the difference, both in the hall and on stage. One particulally experienced musician exclaimed that he had never thought about or heard the enormous difference in sound these modification could make. He was amazed at how what appeared to be such simple acoustic modifications could change the musical values of a performance so greatly.

A few recent findings: The ability to separate the direct sound from reverberation is at least 4dB better in the 1000Hz octave band than at other frequencies. So maximizing the D/R ratio in this band over a wide range of seats is an important goal. In a recent experiment with several subjects we found that direct sound band limited to the 1000Hz octave greatly increased the envelopment of reverberation in the 500Hz band, even when the direct sound energy was 14dB less than the energy in the reverbeartion. It appears that the increase in envelopment or running liveness occurs over all frequencies, even when the direct sound is only perceivable in the 1000Hz band.

The implications of this recent research are beyond the scope of this note. Readers interested in more information, and some hard data on the subject, are encouraged to visit my web page: http://www.davidgriesinger.com. The recent set of power points “The importance of the direct to reverberant ratio in the perception of distance, localization, clarity, and envelopment” contain some of the latest data on the separation process and propose a mathematical method for predicting its success from acoustic measurements and architectrual models. The author believes a thorough understanding of the effects of early reflections on excitement, engagement, and running liveness will both guide acousticians in improving existing halls, and greatly increase the success rate for new consert halls and opera houses.

Leave a Comment

Please note: Comment moderation is enabled and may delay your comment. There is no need to resubmit your comment.