Darryl Miyaguchi wrote in message ...
On 1 Jul 2003 15:10:59 GMT, (ludovic mirabel)
wrote:
Apologies for rearranging your text for my convenience.
You say:
In my opinion, there are two topics which should not be mixed up:
1) The effectiveness of DBT's for determining whether an audible
difference exists
2) The practical usefulness of using DBT's for choosing one audio
producer (component or codec) over another.
Let me be quite brutal about this.
My ONLY concern and interest is in part 2. I'm reading and
participating in RAHE not because I'm interested in psychometric
research but exactly for help to "choose one audio producer over
another". And the "producer" that I'll use in my listening room is
not codec but a musical reproducing device.
I may just as well tackle the question of codec or any other artefact/
vs. a musical audio device.
Are they different? I don't know. You challenge me below to
demonstrate why they shouldn't behave identically. The shoe is on the
other foot- you have first to show that they would.
I'll have to explain something he Basically I come from a different
corner of science from yours. Mine is applied science not basic
research. You ask me why codex shouldn't act like amplifiers. Maybe
they do.
But till there is a convincing experiment to show that it is so, to me
it is just more inference, reasoning by analogy.
For millenia physicians reasoned, speculated and looked at analogies.
The diseased blood is full of noxious miasmas so let's bleed poor Lord
Byron to death.
Sometime in the XXth century things changed. The question to ask
became not: "Is it likely to work because Herr Professor thinks it
should or because there are "good reasons" why it should? or what not.
It became: "Can I design an experiment to show if it will or will not
WORK ?" Reasoning and speculation be damned in medical research-
coarse practicality culminating in Random controlled Double Blind
testing rules. Patient's answers, doctor's impressions are collected
for documentation- the outcome is decided by demonstrable physical
changes.
All that I said before, But now I come to Codex vs Amplifier. No
matter how close the analogy you make it means nothing in the applied
medical research. Add one hydrogen binding to a life saving drug and
it becomes a killer. Of course the Rhinoceros are being exterminated
because the upright horn is a cure for impotence in China. An unfair
(I confess) reductio ad absurdum of reasoning by analogy: " Why
shouldn't the horn work? It looks like IT doesn't it?"
I said:
I am not knowledgeable enough to decide on differences between
your and Greenhill's interpretation of the methods and results.
In my simplistic way I'd ask you to consider the following:
PINK NOISE signal: 10 out of 11 participants got the maximum possible
correct answers: 15 out of 15 ie. 100%. ONE was short of 100%t. He got
only 14 out of 15
But when MUSIC was used as a signal 1 (ONE) listener got 15 corrects, one got 14 and one 12. The others had results ranging from 7 and 8 through 10
to 11.(one)
My question is: was there are ANY significant difference between
those two sets of results? Is there a *possibility* that music
disagrees with ABX or ABX with music?
You answered:
Even between two samples of music (no pink noise involved), I can
certainly believe that a listening panel might have more or less
difficulty in determining if they hear an audible difference. It
doesn't follow that music in general is interfering with the ability
to discriminate differences when using a DBT.
Sorry, but the questiion is a simple one :"Did THIS panel perform
differently on pink noise and music or not?"
And the answer should be simple: Yes or No. Once you answer that you
can give your qualifiers, explanations and so on..
You said:
It's too easy to mix up the topic of the
sensitivity of DBT's as instruments for detecting audible differences
with the topic of the practicality of using DBT's to choose hifi
hardware. The latter is impractical for the average audiophile.
As I said before the only topic I'm interested in is exactly the
"practicality" of its use for comparing components. Were I interested
in DBT sensitivity for other *audible differences* I'd be reading JAES
or a "Journal of Psychometrics" (if there is such a thing) -not RAHE.
But your last sentence certainly rings true- that's why your getting
in hot water with the Old Believers is to be expected.
You said that good ABX component tests are possible (paraphrase). I
answered:
Who can argue with motherhood? The problem is that there are
NO ABX COMPONENT tests being published- neither good nor bad, NONE.
I heard of several audio societies considering them. No results.
Not from the objectivist citadels: Detroit and Boston. Why?. Did
they not pan out?
Your answer:
I can think of a couple of reasons:
1. It's expensive and time consuming to perfom this type of testing
2. The audible differences are, in actuality, too subtle to hear, ABX
or not. Why bother with such a test?
Then there is the possibility that you seem to be focussing on,
ignoring the above two:
3. DBT's in general may be decreasing the ability to hear subtle
differences.
Which of the the above reaons do you think are most likely?
If you ask me : the last one. Enough changed in audio since 1990 to
spur newer comparisons. (See my posting to Mr. Audio Guy). It is
expensive but not beyond the possibilities of such as Boston Audio
Socy. I saw the design of a Seattlee AUDIO Socy AbX test. Then
silence. No results. Was it your reason 3.? I don't know and I never ,
never speculate. Especially since the stock market expired. If it
works why don't people do it?
You asked:
Are you really telling me that you didn't understand the gist of the
group listening test I pointed you to?
For one thing, it says that although people have different individual
preferences about how they evaluate codec quality, as a group, they
can identify trends. This, despite the variety of training, hearing
acuity, audio equipment, and listening environment.
Another point is that it would be more difficult to identify trends if
such a study included the opinions of people who judge the hidden
reference to be worse than the revealed reference (simultaneously
judging the encoded signal to be the same as the revealed reference).
In other words, there are people whose listening opinions can't be
trusted, and the DBT is designed to identify them.
Re "trends and statistical averages:" How "likely" am I to hear those
musical differences under ABX that I heard without it? As likely as
the 72% of Greenhill's subjects who failed at a much simpler task when
being ABXed?.
Or do you have any other experimentally proven statistics?
Let me say something about statistics as applied to prognosis (
outcome forecasting) in medicine.
A patient has inoperable lung cancer. His family want to know how long
he'll live. If you're a heartless fool you say: "average survival with
this is 6 months."
If you're a clever and humane physician you say:" Whatever I'll say
you'll probably want to check in the Public Library anyway- you'll
find a 6 months AVERAGE survival rate . But you husband's name is Joe
Smith not Joe Average. Some die in a few weeks, some in a couple of
years and some-very, very few have an unexplainable, complete
disappearance of the growth. I can not tell exactly how long your
husband will live, but more likely months than years."
A "test" which depends on statistics can not be used as a universally
recommended method of differentiation. It may be O.K. for some and
worse than useless for the others. And the devil is that no one can
tell if he is being deceived by it when he gets a null result because
he may perform differently in 6 months time or when not bothered by
ABX. Or not. Do it at your own risk. Like choosing a book to read, a
movie to attend or wine to drink. Nobody pesters you with a" test"
there. Sorry, "trends" are good for public opinion polls, not as
guidance for inndividuals.
The last point is that I can see no reason why such procedures could
not (in theory, if perhaps not in practical terms) be applied to audio
components. Why don't you explain to me what the difference is (in
terms of sensitivity) between using DBT's for audio codecs and using
DBT's for audio components?
I will not repeat why I consider the above an example of reasoning by
similarity and analogy without experimental evidence. And THAT YOU
fail to supply. Once you do that I'll see if the experiments were well
designed, properly carried out, had good controls and so on.
Ludovic Mirabel