In rec.audio.tech Peter Wieck wrote:
On Sep 11, 9:12 am, "Arny Krueger" wrote:
"Eeyore" wrote in message

...



Peter Wieck wrote:

As to the capacitor story, I had two identical amplifiers (AR - USA),
one with new-and-tested electrolytic replacements on the driver board
as equivalents to the OEM caps, one with 250V film caps also tested in
place of the OEM electrolytics (also replaced on the tone board, but
at 'flat' settings they are not relevant). From the OEM-like unit, the
bass was tubbier... softer if you will, and the treble a little fuzzy
as compared to the film-cap unit. The film-cap unit sounded much
closer to my Citation 16 amp. My wife could also tell the difference,
although her description of it would perhaps use different words. She
preferred the film unit.

Two words: sighted evaluation.

Sighted evaluations are the largest single stimulus to the fabrication and
justification of audio mythologies that have ever existed.

Inclusion of a "even my wife heard the difference" anecdote always gives me
a chuckle.

What utter crap. My wife would have no clue which was which,
furthermore her actual interest in audio other than as a moderate user
is about the functional equivalent of my interest in knitting. So she
would have no brief either way.

Well, that's what you think. But in fact people pick up cues unconsciously.It's why
the most rigorous comparisons are double-blind, for example.



___
-S
"As human beings, we understand the world through simile, analogy,
metaphor, narrative and, sometimes, claymation." - B. Mason

"Steven Sullivan" wrote in message
...
In rec.audio.tech Peter Wieck wrote:
On Sep 11, 9:12 am, "Arny Krueger" wrote:
"Eeyore" wrote in message

...



Peter Wieck wrote:

As to the capacitor story, I had two identical amplifiers (AR -
USA),
one with new-and-tested electrolytic replacements on the driver
board
as equivalents to the OEM caps, one with 250V film caps also tested
in
place of the OEM electrolytics (also replaced on the tone board, but
at 'flat' settings they are not relevant). From the OEM-like unit,
the
bass was tubbier... softer if you will, and the treble a little
fuzzy
as compared to the film-cap unit. The film-cap unit sounded much
closer to my Citation 16 amp. My wife could also tell the
difference,
although her description of it would perhaps use different words.
She
preferred the film unit.

Two words: sighted evaluation.

Sighted evaluations are the largest single stimulus to the fabrication
and
justification of audio mythologies that have ever existed.

Inclusion of a "even my wife heard the difference" anecdote always
gives me
a chuckle.

What utter crap. My wife would have no clue which was which,
furthermore her actual interest in audio other than as a moderate user
is about the functional equivalent of my interest in knitting. So she
would have no brief either way.

Well, that's what you think. But in fact people pick up cues
unconsciously.It's why
the most rigorous comparisons are double-blind, for example.

It's quite clear that Peter is still fighting a battle against the very
concept of double blind.

"flipper" wrote in message
...
On Wed, 12 Sep 2007 08:10:42 -0700, Peter Wieck wrote:

On Sep 12, 10:55 am, "Arny Krueger" wrote:

Nicely said.

'Cept for the fact that you are requiring a 'false positive'.

Not at all. As I mentioned, my analysis of the experiment said nothing
about whether there 'really is' a difference or not.

Agreed. Peter seems to have science backwards. It's not up to anybody to
prove anything before the experiment. However, it is a well-known fact that
the expected outcome of Peter's experiment would be a negative. One key to
any experiment is that it could possibly have a negative outcome. So many
critical subjective elements were uncontrolled in Peter's so-called
experiment that we just don't know much from it.

That a false positive is possible does not make it necessarily so.

Quite right. But that means "might be, might not be" so the experiment
doesn't resolve the question.

If there's a reasonble chance of either a false positive or a false
negative, why go forward with an experiment, especially one composed of one
trial.

That is the fallacy of your reasoning (Circular). I am simply stating
that my wife heard a difference.

The point is that the nature of the experiment precludes coming to
that conclusion because you can not know whether you and/or she
unconsciously influenced the outcome.

Exactly. The probability of a false positive (or a false negative) was about
0.5.

It is up to you to prove the contrary
vs. merely suggesting the possibility of the contrary.

Your logic. here, is backwards. The whole point of the experiment was
to 'establish' something and it is up to the experiment to do so, not
someone else to 'prove' the reverse.

Agreed. If an experiment can't shed any light, why bother doing it?

Or, put another way, there's little point to doing an 'experiment' if
you're going disregarding the validity of it, presume the result you
desire is 'true', and demand proof to the contrary. (see, below,
purpose of an experiment)

Agreed, that is essentially what Peter did. Furhtermore he seems to be
unaware of the fact that the probable outcome would be no differences. In
fact he's been pretty forcably arguing that a positive outcome was probable
because of the underlying technology. He's obvously been deceived by the
Marsh-Jung weirdness, and its ongoing negative impacts on general audiophile
wisdom.

Yes, you may
claim all sorts of problems with the 'test' and the means-and-methods,
but it is still entirely possible that the results are still valid.

No, if a test is critically flawed, then it can't have valid results. It
might have results that agree with the actual truth, but that would be mere
luck.

It may seem a fine point but the answer to that, as stated, is no, the
'results' of a flawed experiment cannot be 'valid' (with 'valid'
defined as something you can use as an established premise from which
rational conclusions can be drawn).

Agreed.

The confusion comes from confounding whether the thing you wished were
'true' is, in fact, 'true' (the 'real world') with the experimental
'results' of the test (a 'constructed' world). So, while it may be the
case that 'there is a difference', the experiment is, by it's nature,
incapable of producing a 'valid' result with which to resolve the
question.

There's a way to double-check an experiment like this. Measure the
performance of the gear before and after the change. If there are no
significant differences based on a comprehensive battery of tests, then the
original hypothesis is seriously in doubt. This is not mission impossible
for amateurs any more, because an adequate battery of tests has been
packaged up and delivered as the Audio Rightmark (RMAA) which is freeware.
With a little cleverness and a reasonably modern Windows computer, you can
insert any piece of audio gear with line level electrical inputs and any
output as long as there is at least unity gain, and get at least an estimate
of how it shapes up.

It is truely amazing what you can change in a piece of gear, with no
measurable affect on its performance, even in sensitive tests. One such
thing would be those op-amp *upgrades* that so many swear by. Ditto for
those passive component *upgrades*, that again many have so many favorable
things to say about.

Another thing that is truely amazing is the size of performance changes and
improvements that you can make by more valid means, and still not have an
audible difference, let alone an improvement.

I.E. The 'result' might be right or it might be wrong, and that makes
it invalid (or maybe we should say "inconclusive").

Note that 'invalid', here, does not mean 'backwards' or that the
reverse is 'true'. It simply means you cannot draw conclusions from
it.

You may not accept them, of course, but you also may not deny that
they are possibly valid.

The purpose of a properly constructed experiment is to preclude
'choice' in 'accepting' the results, which gets back to 'valid'. If
the experiment is valid then the results are valid and one must accept
the data, although there may still be room for disagreement on
conclusions derived from it.

But that one must accept (in science, anyway) the data of a valid
experiment is why there is so much emphasis placed on the testing
methodology. In fact, in the scientific community, that is always the
first bone of contention: was the experiment properly constructed?
Because, if not, then the results are invalid.

Agreed.

With respect, that is the problem with Fanatics... the baby is always
sacrificed whether the bathwater is clean or dirty.

IMO, a 'fanatic', in this context, would be someone who holds to a
particular belief, or set of beliefs, despite seemingly
incontrovertible and overwhelming evidence to the contrary. But the
evidence would need to be valid, wouldn't you agree?

Good point. Case in point would be the Moslem extremists who demand a return
to traditional Moslem law. We know where that sort of thing leads because
it held the middle east in a vice grip for at least half a millenium.
Frankly, at the beginning of the past millenium, it might have been
anybody's guess whether the prevailing law in Europe was any more effective
than traditional Moslem law. However, Europe went through a lot of dramatic
changes while the Moslem world stagnated. Eventually, European culture
became more effective and dominated. Now the Moslem fundamentalists want to
turn the clock backwards for the whole world. Their major accomplishment to
date in Western society has been some messy urban renewal in lower
Manhattan. :-(

Even a fanatic can be 'right', though, because one of the amusing
aspects to science is it's propensity for controverting the seemingly
incontrovertible.

That's why all results of scientific investigation are provisional until we
do something better. However, many of those provisional things have been
remarkably durable and effective, such as Newton's calculus and laws of
motion.