mkuller wrote:
You and the other objectivists have NEVER heard any of these things under
"controlled listening conditions" because the faulty use of DBTs
obliterates
them. DBTs with audio components have ONLY shown gross frequency response
and
loudness differences - therefore in your world those are the only
differences
that must exist.


Bruce Abrams wrote:
You continue to make this statement in spite of the Swedish DBT on CD
players (http://www.jrsaudio.se/dbtoncdplayers.htm) that I posted last week.
Your claim was that since one of the players was perceived to have been
brighter, the difference must have been due to "gross frequency response"
variation. In point of fact, no measurements were provided to either prove
or disprove your assumption. The fact of the matter is that you continue to
claim that the DBT process obliterates the "subtle differences", yet the
Swedish study does prove that 2 CD players were heard to be different under
DBT conditions. I suspect that you're merely quibbling over semantics and
drawing a fundamental conclusion based on the quibble.


First, Bruce, thanks for your thoughtful response. The term "bright" means to
me 'an overabundance of high frequencies'; i.e.a frequency response error.
That it was large enough to be successfully identified in a DBT indicates to me
that it is large enough to be termed "gross". The other unusual aspect of this
Swedish test which made it successful (showing a positive outcome rather than
a null result) was the inclusion of two (out of three) very experienced
listeners, rather than ordinary audiophiles.

If someone in the
Swedish test would have described a "slight hardness and grain in the
treble" of one of the players, rather than having described it as being
bright, would you have made the same comment vis a vis the test only
uncovering "gross differences"? I suspect not.
This is, I believe, the real crux of the rift between the objectivist and
subjectivist crowds. The subjectivists seem to believe that the "subtle
differences" are properties unto themselves, whereas the objectivist crowd
understands those "subtle" differences, when they exist, to simply be the
result of other basic properties (such as frequency and phase response
variations, ability to drive a reactive load, etc.)

While I am not an engineer, I am trained in science and have no doubt that all
of the properties we are able to hear in comparing audio components are related
to things that can be measured. The only thing mystical is what specifically
to measure to explain the particular audible phenomena in scientific terms.
John Atkinson in Stereophile has been trying to correlate the observational
listening results of his reviewers with measurements for some time now.
Sometimes he seems to be able to correlate the two and other times he isn't.
snip
Harry Pearson (and others) were forced to
develope the current "high-end language" to describe the sound of components
that couldn't be described by those published specs. That thinking morphed
into the false understanding that the language described properties that
weren't quantifiable and measurable at all. The language implied that there
was some magical, metaphysical property of a "liquid midrange." (I don't
believe it was HP or JGH's intent to ever create such a misconception, by
the way. They simply recognized that there were sonic properties that
weren't quantifiable and measurable using the popularly published
measurements of the day, and they created a language to describe those
properties.)

The language they developed was descriptive of what they heard with no attempt
to correlate that to measurements. As an audiophile, I don't care about
measurements, only the sound a component provides in reproducing music.

Fortunately, we now understand the fallacy of THD and RMS power ratings and
have a far more sophistiated understanding of the measurable properties that
make an amp transparent.


That still doesn't tell the whole story of amplifier sound or transparency.
Few amps are so transparent that they have no sound of their own. If we knew
everything about measurements and making amplifiers transparent, wouldn't they
all be built the same?

snip
Therefore, if such a characteristic is audible with bias controls in place,
it can be said to be audible. (Claiming that the bias controls destroy the
"subtle difference" is only a statement that can be made if one believes
that such "subtle differences" represent something other than the aggregate
of several other measurable properties.

Not necessarily. The observation that no DBTs have shown subtle audible
differences (dynamic contrasts, imaging, soundstage reproduction, tonal color,
timbral accuracy, etc.) between audio components using music as a source can
either be explained by 1.) those differences do not really exist in audio
components, or 2.) the test being used is flawed. From my extensive
observational listening as an audio equipment reviewer for over 15 years, I am
convinced those subtle differences exist. From my personal experiences with
DBTs, I am also convinced they are flawed. This is reinforced by the results
of Greenhill's famous speaker cable DBT in Stereo Review 20 years ago. With
music as the program, a level difference of 1.75db (gross) was not detectible,
but with pink noise it was. It should not be surprising that subtle
differences are obscured by DBTs, especially when music is used as the program
and average audiophiles' results are averaged to determine the result.

First these average audiophiles had to be able to identify subtle differences,
say in the reproduction of viloin strings, a trumpet or in the dynamics of a
performance. Then they would have to catalog that difference mentally,
remember it, hold on to it - while a dynamic, ever-changing music program is
playing - and then compare it to another sample of that same music. This would
be a daunting task for even the most experienced listeners with great audible
memory (if those individuals exist). It should be no surprise most of these
DBTs yeild null results.
Regards,
Mike

Pinkerton said
They (subtle audible differences) exist, but almost exclusively among
loudspeakers