September / October 2019
Editor's Lead In
Objective Versus Subjective
Editorial By Greg Borrowman

  One disagrees with J. Gordon Holt at one's peril. Although he's no longer here to defend himself (which he could do quite frighteningly), having left these realms for one in which there are no distortions and the music from the harps is always gloriously in tune, there are many who will troll you even after all the bridges have fallen.

For those who weren't around when JGH was running Stereophile magazine, he was the person who introduced the word 'subjective' to the sphere of high-end audio. 'While it is true that a loudspeaker which measures poorly will almost invariably sound poor, it is equally true that speakers that measure exceedingly well often sound almost as lousy, in a purely subjective listening test, as ones that are objectively poor. This is why, of all components, the loudspeaker is the one that must be selected mainly on the basis of how it sounds,' he wrote (his italics).

The passing of the years has meant that JGH's words are now used to justify why measurements are meaningless, which completely ignores the fact that JGH was actually a great fan of measurements for all components other than loudspeakers. (My italics.) 'Objective tests have become far more sophisticated, and meaningful, than they were even five years ago, and have reached the point where, at least for some components, measurements do tell just about the whole story,' he once wrote. And when did he write this? Back in 1966, when test instrumentation was in its infancy, and it was as rare to find an audio oscillator that could produce signals with less than 0.01% distortion as it was to find an audio analyzer that could detect such levels.

This means that somehow, over the years, audio subjectivists have managed to twist JGH's original words to imply that he meant that all measurements are meaningless and, in this internet world this idea is gaining traction. In the last month we have had two different distributors refuse our offer of a review in Australian Hi-Fi magazine because the manufacturers of the equipment they are distributing 'did not want their components tested.' As it happens, one of these components was a loudspeaker and the other an amplifier. I am prepared to bet serious money that both manufacturers not only own and use test equipment, but also test their products before they leave their respective factories, so they obviously believe in 'testing'... so why would they not want their products tested? The only reason I can think of is that they know that such tests would reveal that their products do not meet the claims they're making for them.

However, the reason given to me by the local distributors was that 'the manufacturers don't believe that measurements will help consumers assess the sound quality of the components.' Whether or not this is true actually begs the question. I wholeheartedly agree that testing the power output of an amplifier, for example, will tell me absolutely nothing about how that amplifier will sound. However it will tell me whether the amplifier is as powerful as the manufacturer claims it is, and it will tell me the maximum sound pressure level I will be able to obtain in my room when I connect a pair of loudspeakers whose sensitivity has been measured as 90dBSPL. As a consumer, I would like to know both these things. Other measurements can certainly predict how an amplifier may sound. For example, an amplifier with high levels of low-order even-order harmonic distortion will sound far 'richer' and more 'musical' than an amplifier with very low overall distortion. How would I know which amplifier sounded the best (to me)? By listening to them, of course! But at least with suitable objective tests in my hand I would not only know why I was listening, but also what I should be listening for.

— Greg Borrowman

Subscribe To Australian Hi-Fi Magazine
Get this issue of Australian Hi-Fi Magazine by clicking here.