Objectivism vs subjectivism - not always in conflict
skeptic67 -- Mon, 09/08/2008 - 22:19
After reading RH's AES paper "The role of Critical Listening....." and his more recent "From the Editor" column in October's TAS, I have come to the opinion he is trying to paint too black and white a picture of critical listening and the role measurements play in assessing audio equipment. It seems there is no middle ground in which objective measurement can never be used in concert with critical listening. I disagree.
In his "Audio Fundamentalism" column he clearly states he "never found a single correlation between how a piece of electronics measured and how it sounded" after measuring several hundred audio products. Does this opinion exclude passive audio (i.e. loudspeakers)? I'm not sure. However, in the area of loudspeakers, I submit some headway has been made in correlating subjective listening tests with scientific measurements in this vital end point of the audio chain. I refer the reader here to Dr. Floyd Toole of Harman Int'l Industries and his paper "Audio - Science in the Service of Art".
http://www.harman.com/wp/pdf/AudioScience.pdf
RH's time frame of reference for his statement above covers the years 1989 to 1997. Eleven years have passed and the science of audio measurements has evolved significantly since that time period as evidenced by Dr. Toole's paper. Sure, it's biased toward JBL products but the text speaks for itself.
I believe there are numerious loudspeaker engineers/designers today who use both measurements and critical listening in the development of their products. They would be crazy not to.
I agree the science of audio measurement has a very long way to go and that the evaluation by human ears will never be displaced completely by it. But to say that measurements have no relavence, as RH implies, is folly in today's world of audio technology.
All content, design, and layout are Copyright © 1999 - 2011 NextScreen. All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
Loudspeaker measurements can provide insight into a product's performance. Notice that in my editorial I said that I had not found a correlation between how electronics measured and sounded. I intentionally didn't include loudspeakers. There's no question that today's sophisticated loudspeaker measurement tools are partly responsible for the overall improvement in speaker performance over the past ten years. Still, measurements are no substitute for a skilled pair of ears. I discuss the roles of listening and measurement with David Wilson in my interview with him that will appear in the December issue.
I also believe that certain designers have developed sophisticated measurements that can be correlated to audible phenomena. These measurements go far beyond the traditional distortion, noise, channel separation, and other measurements commonly used. A good example is Keith Johnson's cluster tones for assessing digital-to-analog converter performance. Such examples are, however, quite rare.
I think there also tends to be a confusion between measurements being useful for designers and measurements being useful for consumers. The former (likely) doesn't imply the latter. The usage context for measurements matters.
CEO and Editorial Director, Nextscreen LLC
I'm Skeptic 67 who started a new post the other day in your Evaluation, testing, measurement and Perception area. This morning I tried to submit a new post related to Objectivism vs Subjectivism, etc. and could not. The submit button only led me to the preview window.
Here is my reply: perhaps your webmaster could post it for me? I don't know what else to do.
Thanks RH for now differentiating loudspeakers from electronic audio equipment. I only wish you had been more explicit about this apparent distinction in the "Role of critical listening...." paper. I was completely led to believe you felt ANY measurements were essentially useless.
In light of your reply, perhaps you might consider combining some measurements with your next review of loudspeakers. Stereophile has started to add 1/6 octave response measurements (taken at a loudspeaker reviewer's listening location) to their already hi-tech laboratory set of measurements. In a recent review, JA noted some correlation he found in measurements he took at M. Fremer's listening room during MF's review of ProAc Response D28 (Aug. 2008 issue). In this case the 1/6 octave response trace was found to correlate with MF's description of the D28's balance.
I only suspect at this point your Wilson X-2 review is complete. It would have been nice to see a 1/6 octave response plot of those speakers in your listening room along with your review.
[I am an admin, posting on behalf of Skeptic 67, whose message is below.]
Thanks RH for now differentiating loudspeakers from electronic audio equipment. I only wish you had been more explicit about this apparent distinction in the "Role of critical listening...." paper. I was completely led to believe you felt ANY measurements were essentially useless.
In light of your reply, perhaps you might consider combining some measurements with your next review of loudspeakers. Stereophile has started to add 1/6 octave response measurements (taken at a loudspeaker reviewer's listening location) to their already hi-tech laboratory set of measurements. In a recent review, JA noted some correlation he found in measurements he took at M. Fremer's listening room during MF's review of ProAc Response D28 (Aug. 2008 issue). In this case the 1/6 octave response trace was found to correlate with MF's description of the D28's balance.
I only suspect at this point your Wilson X-2 review is complete. It would have been nice to see a 1/6 octave response plot of those speakers in your listening room along with your review.