MEASUREMENT

Sound systems must be listened to. We can make quantitative measurements of the system's performance but ultimately it is the subjective result that people are willing to pay for. However, as the old cliché goes, "ask any two people about what they hear and you will get three opinions." For this reason it is necessary to quantify system performance in a meaningful manner. Of course, this quickly leads to questions of which measurements are meaningful and which are simply academic.

Measuring a sound system is pretty straightforward while the signal remains in the electronic realm, but it becomes much more complex when the audio becomes sound in a venue. This is the natural habitat for sound systems, not the laboratory anechoic chamber, but separating the sound system from the acoustical environment can be tricky. Unless we can make the right choice in dividing these realms we will forever be measuring problems that cannot be solved from within the sound system component chain.

To measure meaningful acoustical performance it is necessary to be able to select both the frequency and timing of the measurement. The human auditory system makes many of the quality judgements based on the relative arrival time of the sound and can distinguish between reflected sounds and those that arrive directly from the source. Reflections that arrive within the first five to ten milliseconds after the first (and in most cases - direct) arrival from the source will alter the perceived tonal quality of the sound. Reflected sounds that arrive between 10 and 50 to 70 milliseconds will be ignored unless they are significantly louder than the first arrival. Reflections that are later than this begin to sound like echoes or alternate sources of sound, at least until they are diffuse enough to become reverberation. These reflected sounds can also be created by other loudspeakers but the perceived effects are similar. For these reasons, it is particularly important to be able to measure the sound system's performance in the time domain.

While the RTA (real-time analyzer) is ubiquitous, it's inability to discriminate between the direct arrival of sound from the source and the later arrivals (often lumping together all of the energy that arrives in the first ~300 milliseconds) makes the RTA a rather clumsy tool for this delicate task. It was the blindness (in the time domain) that these instruments have that caused many people to believe that a "flat" sound system sounded bad. In fact, if you measure the direct sound and include the reflections in that critical first 10 milliseconds after the direct sound, you have a very good idea of the sound quality of the system. You also are able to see those very early reflections and can judge whether physical changes (reorientation of the loudspeaker or nearby surfaces) are required to achieve the subjective sound qualities required from the system.

There are a number of good measurement systems available that can provide this time and frequency data. I analyze sound systems with my good old TEF-20 (recently acquired by Gold Line), which allows tremendous control over the signal measured and the manner the data is presented for analysis, and the very quick and easy to use JBL-Smaart Pro software. These measurement tools allow me to determine how much of the frequency range and time window I wish to view in the analysis. It becomes much easier to determine which problems can be fixed with tweaks of the system components and which problems must be solved with acoustical treatment or physical placement of the devices.

The flat frequency response at the listening position (all of them) may be just a starting point, if you wish to "voice" the system to suit the style or preference of the performer. However, unless you can achieve a baseline performance that is free from anomalies throughout the audience, the voicing process may be frustrated by problems in the basic setup of the system. Measurement is really a tool to assist by visualizing the acoustical performance. These tools help the experienced user to identify and quantify the problems that were often audible before the measurements began.

Once you become familiar with using your measurement tools, you begin to develop a correlation between what you perceive when listening to a sound system and the measured results. You will no longer have to always correct the measured response, by ear, to achieve a good sounding system. However, this does not preclude listening to the system before concluding that the setup is complete. No measurement system can match the sensitivity of a trained listener.

United Entertainment Media Inc.
460 Park Avenue South, 9th Floor
New York, NY, 10016
Ph. 212-378-0400
FAX 212-378-2160

page construction