ScottW wrote:
wrote in message
oups.com...
ScottW wrote:
wrote in message
ups.com...
Reply to ScottW. PART 2
ABX was proposed as a "test" to facilitate differentiating audio
components.
1)I did claim that ABX proposed as a "test" for uncovering subtle
differemces between COMPONENTS was not used by anyone,( including Sean
Olive), who got positve results and whose paper was accepted and
published in a reputable, peer-reviewed, professional journal.
POSITIVE report:i would be "Yes we heard a difference" None exist..
"Other applications" are of no interest.
2) I did claim that it points to the fact that as of now 40 years from
its inception it remains an unvalidated proposal for COMPONENT
COMPARISON . I have no opinion and no interest about its uses in
psychometric research etv. Do you?
3) Yes,that is a FLAW for the subject in question
3) The wording and reasoning behind the NONUSE is immaterial. As of
now it is an unvalidated, unusable project. If you want to dwell on the
interpretation of various reasons given for that nonuse be my guest.
And here is my slant on it. A proposal for a "test" which asks people
to listen to A then B then X and next asks :"Is X piece of music you
just heard more like B that you heard before or more like A that you
had had heard first is doomed to result in the "it all sound the same"
by the majority.
4) In the past you voiced other reasons why I "misinterpret" am "not
to be trusted" and "lie" about Sean Olive. I'm not very clear about
your objections. If you want to, please repeat them and we'll go from
there..
My response (can't understand why the process to mark
replied to text sometimes fails in OE).
No objections. I am not going claim ABX is proven any more
than you are claiming that it is disproven.
I would ask... if ABX is unsuitable for establishing differences,
what would you suggest?
ScottW
------------------------------------------------------
You say:
I would ask... if ABX is unsuitable for establishing differences,
what would you suggest?
My answer is :
) Measurements, if of well- established value, such as frequency
charting
used by Olive
are obviously important. Atkinson does some other measurement that I'm
not
competent to pronounce on.
2) Beyond that everyone is on his own. Just as our DNA's differ so do
our
perceptions. A car music fan may consider virtuoso violin a squeak and
I may
cover my ears when at a stop sign I'm blasted by boom boom from the
other lane.
3) It is a lucky event when mesurements and preferences of the majority
coincide
as in Olive's loudspeaker tests
But even there there were up to 27/1 differences in "consistency"
between
various groups of listeners.
What I'm driving at is that essentially there ain't no test.
Opinions, preferences, likes and dislikes yes. A"test" for them is a
pipe dream
Science at this stage
of its progress (and maybe never) an summarise how your,
mine and Jenn's grey matter process the information.
Even our decisions whose opinion to trust depend on instinctual inborn
or acquired neuronal affinities. Eg. in the newsgroups I'll trust Harry
Lavo's or
Jenn's preferences because they are more likely to be same as mine.
I won't say whose choices I wouldn't trust-,I'm trying hard to be
non-controversial
But I'll say that bestsellers are no guides to me. In lterature, in
music nd so on.
In brief my answer is: When it comes o preferences ie. aesthetic
choices
Any " test" tests the guy/girl performing it.
Ludovic Mirabel.
You might find this interesting.
http://www.aes.org/e-lib/browse.cfm?elib=9248
and Bern Muller will discuss its current status here.
Message id:
I'd be interested if his presentation becomes available
on the web.
ScottW
-----------------------------------------------------------------------------
D.L. Clark is or was Arnie's collaborator on the ABX research. I think
he's
serious and has serious things to say. It is interesting that he
appears to be
going beyond ABX. But the summary does not extend a promise of a
change revolutionary enough to fork out twenty bucks. I'll wait till I
can get it
next time I visit Vancouver Public Library.
I'll take this opportunity to raise one more point. You said that ABX
has not been
proved or disproved. That is not the way things work in research.
Imagine a guy saying to someone with Aids:" I discovered a cure for
Aids.
Please start my pills".
Wouldn't the sick man say: "Did you present your evidence to my
doctors."?
"No, I asked them to disprove my cure and they could not"
ABX is a test for showing differences between components. So far, in
four decades, it did not
amass enough convincing evidence to get it published in
a professional journal for acceptance or rejection by other
experimenters.
In medical research such a proposal would have been dead and buried
at birth. It is worse than putting a cart before the horse.
You can't "disprove" something that does not yet exist
Ludovic Mirabel
P.S. I hope we're setting an example to fellows RAOers. I for one
prefer to
disagree, if I have to, while respecting my opponent..
..
Ludovic Mirabel