Question to the AV Professional, How Much Does “Fidelity” Matter?

by Richard Honeycutt

As AV Professional, fidelity is a familiar and important term. “Will this addition or change really matter to the system as a whole?” If so, at what cost? Richard’s years of experience bring out some things to consider.

Many of us codgers (those aged above 50) in the sound business were first infected by a love of audio in connection with home music systems. We drooled over the Fisher, McIntosh, Harmon-Kardon, Heathkit, and even Knight-kit components in the Allied Radio and Lafayette Radio catalogs. We devoured the brochures from Klipsch, Electro-Voice, and perhaps Bozak and Acoustic Research. And we came away with the idea that a real sound system should have a frequency response flat from DC to light and THD well under 0.1%. IM Distortion (IMD) was not emphasized as much, and we hadn’t heard of Transient Intermodulation Distortion (TIM – also called Slewing Induced Distortion: SID).

Only Klipsch made a big deal of loudspeaker directional patterns for home applications, and although loudspeaker frequency responses were often printed in data sheets, these were often based more on marketing than on measurements. Speaker distortion was seldom mentioned on spec sheets; again, Klipsch was the exception.

When we migrated into pro audio as a career, many of us carried our previous learning/opinions with us with little or no critical thought. Naturally, a good pro sound system would just be a good home system writ large! Frequency response was naturally of primary importance, followed by distortion levels. But in the light of a half-century or more of work reported in journals of the ASA, AES, and other professional societies, how many of our early opinions still stand?

 

Hearing Loss with Age

Our adolescent idea was that since the range of human hearing was 20-20,000 Hz, a sound system should have a response at least that good. Although we were aware that some of our elders had lost some hearing sensitivity, we did not yet know that the loss was usually worse at high frequencies, or that our penchant for listening to loud rock music would damage our own high-frequency hearing response.

Figure 1: Hearing Loss vs Age (from http://www.roger-russell.com/hearing/hearing.htm)

Figure 1 shows that average hearing sensitivity for males  at age 65 is about 40 dB worse that at age 20; the loss is not as bad for females, but is still about 25 dB. Compared with our adolescent insistence that our music amplifiers be flat within at worst +3dB, and preferably within +1dB, these losses are huge. (Please note that these are averages, and people who have avoided exposure to loud sounds and have not been affected by diseases or ototoxic medicines that damage hearing may well have less HF loss in their old age.)

Frequency Response

So what are the implications for pro sound system design? If the average audience age is 35, a system HF response extending to 8 kHz is almost certainly good enough. Tests done by the BBC in the late 1960’s, using both trained and untrained listeners, indicated that very few people can tell the difference between a studio monitor speaker that has a flat response to 15 kHz and one that is low-pass-filtered to a 12.5-kHz cutoff. Probably a sound system in an auditorium or church sanctuary would be even less critical in terms of HF extension, because of background noise and reverberation considerations.

And as to LF extension, measurements indicate that most acoustic instrument contain little energy below about 40 Hz (4-string electric bass or string bass using standard tuning). Even these instruments have very little fundamental energy, so the second harmonic—80 Hz and above— becomes the useful LF limit. But then there are systems that must reproduce synthesizers and sound effects. I have a CD that contains a strong component at 20 Hz in the first few minutes, and movies with explosions often have lots of energy down into the subsonic range. So the necessary LF limit depends upon the use of the sound system.

Now how about our prized +1dB or +3dB tolerances for response flatness? Figure 2 shows the frequency response of one of the finest pro-sound speaker systems on the market today.

Figure 2: Frequency Response of an Excellent Loudspeaker

The response flatness is about +3.5, -5 dB from 50 Hz to 18 kHz. Note that all major deviations from flatness are pretty narrow-band. In my experience designing, building, and measuring  loudspeakers, I have found that response “wiggles” narrower than about 1/3 octave are not easily heard, and the narrower the wiggle is, the more extreme it can be and still remain inaudible. (It is true that sometimes an extreme response peak that is very narrow is caused by a resonance in the speaker that causes sound at that frequency to persist, and the persistent sound can be annoying, but that’s as much a result of poor damping at that frequency as of poor response flatness.)

Distortion

Now, how about that other bugaboo of our adolescent years: distortion? Researchers over the years have published perceptibility thresholds for harmonic and modulation distortion ranging  from 0.03% – 10% for harmonic distortion, and 0.1% to 1% for modulation (or intermodulation) distortion. The variation in results seems to come from lack of standardization in testing protocols. My own experience is that I only notice harmonic distortion if it is above about 3% and I’m listening for it. If my attention is on the flow of a piece of music, I may not notice harmonic distortion until it exceeds at least 5%. I am slightly more sensitive to IMD.

But I have been in many venues in which audience members praised the sound system, even though to me the distortion was extremely annoying. During the 1950’s and 1960’s, military training films were often produced using a clipping limiter that introduced about 5% to 10% distortion on peaks. Tests purportedly showed that the distortion actually increased intelligibility—presumably by increasing the HF energy a bit. Most speakers produce more than 5% THD at low frequencies, when operating at high levels. And the human ear also produces IMD when the sound level exceeds something like 90 dB-SPL.

At the end of the day, we codgers find that we have to admit that frequency response and distortion do not absolutely define sound-system quality. There are acoustical factors in any venue that are at least as important. But that’s a topic for another time. rh

Richard A. Honeycutt developed an interest in acoustics and electronics while in elementary school. He assisted with film projection, PA system operation, and audio recording throughout middle and high school. He has been an active holder of the First Class Commercial FCC Radiotelephone license since 1969, and graduated with a BS in Physics from Wake Forest University in 1970, after serving as Student Engineer and Student Station Manager at 50-kW WFDD-FM.  His career includes writing engineering and maintenance documents for the Bell Telephone System, operating a loudspeaker manufacture company, teaching Electronics Engineering Technology at the college level, designing and installing audio and video systems, and consulting in acoustics and audio/video design. He earned his Ph.D. in Electroacoustics from the Union Institute in 2004. He is known worldwide as a writer on electronics, acoustics, and philosophy. His two most recent books are Acoustics in Performance  and The State of Hollow-State Audio, both published by Elektor.