An Audio Horror Story: Testing and Tuning a Soundspace

Ron Pellegrino, 2/19/98


Horror seems to be an awfully strong word to attach to an audio report, but personally I find any experience that negatively affects my digestive system deserves the adjective horror. So this is an essay based on a gut-wrenching experience I had recently with a soundspace that had been tuned mechanically by an audio technician applying a formula he maintained was an industry standard. He insisted that the objective tuning procedure he applied to the soundspace was the only correct way. Of course, in the end, he had to go along with my subjective approach since I was the guest artist and his job was to supply me with technical services. Given that I use my own compositions as some of the test signals, I know exactly what I want to hear in a soundspace; nevertheless, the technician believed the soundspace should have a flat response according to his test instruments and procedures. In a nutshell, the conflict was between experience/irrefutable knowledge versus textbook/classroom mechanical procedures; further reduced, it amounts to art versus technique.

The strange soundspace

As I began my work in the context of a guest artist performance-multimedia residency in a university setting, I found myself working for the first few days in an astoundingly strange soundspace. The room was a "black box", a rectangular space (with a high ceiling and gently scalloped side walls) located on the inside of a large university center for the arts. In the 1990s black boxes are commonly found in institutional settings such as museums, cultural centers, and universities. They are popular because theoretically they are completely configurable performance spaces for music, theater, dance, multimedia, etc. Though certainly not my favorite audio environment, I´d worked many black boxes before this one and invariably found them easy to tune.

During the first two days of this residency, as I was meeting music classes in theory and composition, setting up and testing my performance-multimedia digital and analog instrumental systems, and auditioning acoustic performers for possible inclusion in a public performance-multimedia event on the final evening of my residency, I was repeatedly stunned by how strange, distorted, and unbalanced everything coming through the electronic audio path sounded in the space. This was a space that sounded simultaneously thin, boomy, harsh, strident, and unbalanced - truly puzzling. Over the course of my professional career of three decades and over 500 public performance events, I´d never heard anything like it.

Audio tests and experiments

The strangeness of the soundspace was surprising because I´d easily worked similar spaces in the past, the stereo audio system was new, and the system passed my first test with flying colors. That first test was to sweep a sine tone at various amplitude levels over the full audio frequency range of each channel and JBL speaker stack. That test confirmed that the audio paths, which I thought just included stereo speaker stacks, mixing board, and stereo power amplifier, were completely clean.

During the auditions and classes of the first two days I was puzzling aloud and focusing the ears of students and faculty on the twisted soundspace. All the time I kept walking the space and adjusting variables in the system trying to sculpt sound that was at least acceptable to my ears. As always I emphasized to students and faculty my position that true fundamentals of music theory should apply to all music from any place and any time. Based on that encompassing principle, the study of music theory fundamentals should focus on 1) the nature of sound, 2) how we hear, 3) how acoustic and electronic instruments work, and 4) the nature of music performance spaces. Clearly this wacky soundspace provided the perfect opportunity to test and demonstrate the validity of my position on music theory fundamentals.

During the first day and half in that space I tested the space with audio coming from equipment using the sound material that would be used in the performance-multimedia event on the final evening of my residency. The sound material included my own compositions, the acoustic instruments and voices that would be programmed on the performance-multimedia event, and other music I use in my demonstrations. The equipment included an S-VHS Hi-Fi video deck, a DAT machine, a lavaliere microphone, an analog synthesizer, a PowerMac, a cassette deck, and a guitar amplifier with a mix of straight guitar and voice through a microphone. In this particular soundspace everything sounded awful.

As I routed sound from various sources through the system into the air I walked all over the space while periodically testing various levels and combinations of the six equalization bands on the mixer. I was able to make small improvements in the sound but it took unusually large equalization offsets. I began to feel locked into a truly mysterious soundspace that mangled the sound of my own synthesizer compositions so badly that my digestive system was being effected.

Discovering the culprit

Finally during the evening of the second day I had to start setting up the laser animation system that I use for my visualizing music demonstrations and performances. That setup requires darkening the space as much as possible so the laser images appear to float off the projection screen into 3-D space. As I was accounting for and reducing the intensity of various light sources throughout the space, I noticed a rack with a lot of LEDs off in a far corner. When I asked the theater technician about the rack, I was informed it contained stereo equalizers in banks of 1/3-octave bands. He informed me that the technicians had "tuned"the equalizers using a pink noise generator patched through the board and speakers to create a "flat" space. What I saw when I walked over to the rack to examine the equalizer settings stunned me. The settings of the equalizer sliders were all over the map. I gulped, stared in amazement, and realized that the equalizers were the source of all the audio evil I´d been trying to exorcise for two days.

Finding the natural soundspace

I strongly suggested to the technician that we hit the defeat switch on the equalizers to allow the audio signals to by-pass the equalizers completely; that enabled us to AB test the soundspace with and without the equalizers in circuit. That was it; it was perfectly clear that the settings of the 1/3-octave band equalizers were the source of my gut-wrenching. With the equalizers out of the circuit the sound world was back to normal and I could return to my normal procedure of tweaking the equalizer bands on the mixing board for all required fine tunings of the soundspace.

In addition to by-passing the equalizers, we moved the JBL speaker stacks to "compromise" positions. When I first walked into the space the JBLs were set along opposite walls very close to the first row of audience seats, about 1/4 of the way into the space from the rear; far too close for audience comfort and safety. We moved the speakers to the corners of the back wall but that was too far away from the audience and did not provide enough presence in the sound. The compromise position was about 1/6 of the way into the space from the rear. The point here is that in designing a soundspace it´s crucial to experiment with speaker placement and to factor in audience comfort and safety as well as the quality of sound from all locations in the audience space, not just the "sweet spot." For tuning a soundspace there is no substitute for walking around the performance space while listening and tweaking the sound. It´s also a good idea to test all the sound sources at slightly above the levels you plan to use in performance so as to allow for a bit of headroom. The best soundspace tuning instrument is a personal experiential knowledge base built up over the years by using your ears and by fully comprehending what you´re hearing!

An appropriately tuned soundspace versus an "ideally"flat soundspace

Of course, the technician was insulted that I defeated all the work they had done to provide me with an "ideally" flat soundspace. The following morning I asked the technician about the process used to "tune" the space. Outlined below is the procedure they used. They told me that what they did is what all the professionals are doing today in inside soundspaces. It was news to me. And heaven help the unwary who place their musical trust in the hands of audio professionals who rely solely on test equipment and mechanical formulas to tune a soundspace.

The audio technician´s procedure for tuning a flat soundspace (or setting up a musician´s test-your-mettle/audio-hell-on-earth):

  1. Choose an arbitrary location for the speakers without regard for audience comfort and safety.
  2. Patch a pink-weighted random noise generator at the beginning of the audio train and patch a 1/3-octave band equalizer in series before the power amplifier.
  3. Find a matching 1/3-octave band spectrum analyzer and go to the audience point of an equilateral triangle formed by the placement of the stereo speakers and that audience point (the so-called "sweet spot" where nobody will sit anyway because it´s in the middle aisle).
  4. Crank up the pink noise generator and adjust all the 1/3-octave equalizer levels so the readings of all bands on the 1/3-octave band spectrum analyzer are equal and flat.

From an artistic perspective the only value of the above procedure is that it provides plenty of food for thought. It´s a mechanical, technical exercise that is worthless musically. To understand why that´s the case, let´s walk through the procedure and examine each step.

  1. Anyone tuning a soundspace should always consider the comfort and safety of the audience . They are our guests. They are our reasons for being there. Never place speaker stacks immediately in front of audience seats. People want to enjoy a musical experience, not be assaulted by excessive audio levels.
  2. Pink-weighted noise is noise that is weighted so that the average power in each octave band or 1/3-octave band is the same. The fact is that any noise generator is a valuable tool for audio processing, experimentation, and ear training but it should not substitute for or be confused with the music and speech that populate the performance world. As test signals for tuning a soundspace there is nothing better than using the music and speech that will be part of the public event. And, most importantly, have someone present who knows exactly how the test signals should sound.
  3. When testing audio in a soundspace, walk the entire space while you´re listening to and testing all the types of audio material you plan to use in performance. Chances are slim that you´ll be doing a pink-noise concert and even slimmer that the entire audience will be sitting in the sweet spot.
  4. Tuning a soundspace is an art that requires experienced, knowledgeable, musically well-trained ears in the style and requirements of the audio being presented. It is not the sort of problem solved by a non-creative mechanical exercise based on formula solutions. The ideal tuning of a soundspace requires a judicious balance of functionally excellent equipment, speaker placement, individual level settings, combined level settings for a mix of sources, and finely tuned equalizer settings to compensate for the inherent inadequacies of various pieces of gear as they relate to the soundspace.

Placing responsibility

In retrospect what I find most disturbing about this soundspace experience is that it further confirms my misgivings about the quality of instruction given to audio technicians and audio engineers. The educational institutions seem to be providing students mainly with technical information and formulas based on equipment. In doing so the educational institutions are cultivating mechanics and technicians rather than artists. The difference between the mechanical approach and the artistic approach is one of attitude. Mechanics approach the problem with lists, specifications, formulas, and standard procedures in their search for an objective solution. Artists approach the same problem with an experimental, intuitive, relative, flexible, and experiential attitude in full confidence that they´ll find the correct subjective solution. Unfortunately institutions find it far easier to teach and measure the mechanical approach than to teach and measure the artistic approach. It´s clear that institutional laziness and low standards threaten to undermine and diminish the musical experience in our culture. But hope springs eternal; it´s never too late for a turnaround.

Site Navigation Links

Booking information and comments.

©1996-2004 Ron Pellegrino and Electronic Arts Productions. All rights reserved.