The Ongoing Debate about Amplifier "Sound"
Several years ago, I wrote an article in which I attempted
to define the term "audiophile" as it has evolved in the years I have been
observing the hi-fi scene. I concluded that, whatever the word may have meant originally,
it now had come to refer to members of a group that "tended to ascribe almost magical
audio qualities to things that the rest of the world had either abandoned or didn't feel
were of much importance to begin with or were just plain crazy . . . . The notion became
widespread in this camp that formal measurements were far too coarse to be meaningful, so
that two pieces of equipment that measured identically often sounded very different to a
critical ear."
From early days, one constant point of disagreement between
these listeners and more measurement-oriented stereo fans has been the matter of audible
characteristics of power amplifiers. The test-bench bunch have always maintained that two
components that measured the same would -- and do -- sound the same (assuming, of course,
that they have enough power to do the job and are set up properly). The audiophile fringe
has always attributed very specific audible characteristics to amplifiers: some are harsh
or fatiguing, others are sweet and "musical" and so forth. You can always raise
a fine argument among sound people by bringing up this topic.
Maybe a definition is in order here. As it snakes its way
through a stereo system, an audio signal is very tiny, but in the end, it has to be
boosted to a high enough level that it can drive a pair of loudspeakers -- essentially
electric motors designed to push air. The last link in the audio chain magnifies the low
audio signal to a much higher level for feeding to the speakers. That's the power
amplifier and that's its only job. Any control features that might be included, on the
same chassis or elsewhere, are actually functions of what is called a
"preamplifier" because it comes ahead of the amplifier in the chain.
Ideally, a power amplifier should have no effect on an
audio signal except magnitude; otherwise, it should be a totally neutral device. But no
audio component is perfect, and there are always tiny measurable differences from one
device to the next. The question is: can we reliably hear these? And if not, who cares?
One side maintains that, beyond a certain level, gains in
measured performance may suggest more careful design or meticulous manufacture, but they
don't result in audible improvement. Others believe that if the variations exist, really
careful listening will reveal them.
To shed some light on this, several years ago Stereo
Review magazine [now Sound & Vision] commissioned an extensive series of
listening tests to determine whether those listeners who claimed to be able to discern
tiny nuances from one amp to the next really could. The listeners, about two dozen in all,
were equally divided between those who swore there were differences
("believers") and those who doubted it ("skeptics").
Six amplifiers were auditioned, ranging from a $200 cheapo
receiver to a $12,000 pair of mono tube amplifiers. They were teamed up in pairs for
comparison, and carefully matched as to levels and other parameters, so that any
differences heard really were inherent in the different amplifiers. In any given
comparison, the listeners were unaware of which they were listening to at any moment, and
as they switched back and forth, they might be toggling between the two components or
simply switching the same amp in repeatedly. Then a statistical analysis was done to see,
when they claimed to hear differences, how often they were listening to one amplifier in
both positions.
It was a long drawn-out process, and in my story on it at
the time, I concluded that ". . . all interpretations of [the results] lead to the
conclusion that correct choices were made totally by chance -- there were no audible
differences to be heard. . . . The evidence would seem to suggest that distinctive
amplifier sounds, if they exist at all, are so minute that they form a poor basis for
choosing one amplifier over another."
The result didn't surprise me very much because, ten years
earlier, I had taken part in a similar exercise mounted by the then-leading Canadian hi-fi
mag, AudioScene Canada. In that case, we started out assuming there would be
differences; we wanted to analyze and quantify them, but instead reported that we couldn't
hear any.
I wondered in that report: "So where are all these
phantom differences coming from? First, from the imaginations of people who think there should
be differences, and so conjure them up in the absence of proper facilities for proving
they don't exist -- and egged on by the purple prose of the little super-audiophile
magazines . . . who evidently must prove the superiority of their golden ears over
others' by hearing faults that don't exist and condemning perfectly good products on their
strength."
After that issue hit the stands, I had an interesting
footnote to it. I visited one of the companies whose amplifier had been included in the
test, and found that the president (who was also an engineer and had in fact designed the
unit in question) was unperturbed by our findings. He acknowledged that an amplifier doing
its job should have no effect on the sound. His marketing manager was less certain,
feeling we may have missed something. The product manager was even more adamant. And so
forth.
In fact, the closer you got to the sales floor, the more
outraged the employees of this company became at our assertions. The notion seemed to be
that if there weren't audible differences, then they had nothing to sell -- buyers
should simply pick the cheapest model and leave it at that.
There are lots of reasons not to do that: reliability,
warranty, brand reputation, and on and on. But so far no one has proved conclusively -- to
me, anyway -- that sound quality should be a factor in buying a power amplifier.
...Ian G. Masters
ian@mastersonaudio.com
|
