|
|
|
|
Why DBTs in audio do not deliver
On Mon, 21 Jul 2003 22:29:04 GMT, (ludovic
mirabel) wrote:
2) In a previous posting I denied your repeated claim that JJ. quoted
mysterious nonpublishable industrial component tests. I said also that
an anecdote was ALL he had for evidence that ABX does produce
occasional positive results.
You requoted that anecdote in full and added it to other similar
anecdotes about Pinkerton and Krueger. And...? So what? Do you know
what a reference is?: Name of the mag, author, year, month, page?
Mirabel, you are perfectly well aware that I have posted positive ABX
results on several occasions, plus Arny's PCABX website is well-known
within this community, and is an excellent resource for those who wish
to know the *truth* about detecting sonic differences using
douyble-blind protocols, as opposed to your verbose denials.
3) The discussion about what "sounds" to use in ABXing goes like this:
a) we, consumers buy components to listen to music
b) we Old Believers have a test to help you out.
c) when we, O.Bs. do our test (seldom!) we find that the performance
improves using an artefact called "pink noise" instead of music.
Yes, the above is all true.
This leaves three possibilities: electronic components are designed
for pink noise, our brains are designed for pink noise, our test
stinks for assessing music reproduction by electronic components.
It also leaves the possibility that steady-state pink noise is quite
simply a more sensitive test signal for detecting small sonic
differences, which might go unnoticed with the much more varying sound
of music.
No way could our test stink. We'd, Lord forbid, would have to drop it
and stop boring the pants off everyone with "but it should be proved
by controlled etc, etc....". So.... let's drop music..
Typical non-sequitur argument from you.
4) You, like every other RAHE "scientist", failed to produce any
creditable, positive, published with the necessary statistical detail
etc ABX component comparison tests
even though 30 years have gone by to get one.
No, many such results have been quoted, but you conveniently dismiss
them all for assorted spurious reasons, while offering *zero* evidence
for your own position.
It is a basic research principle that a "test" with null outcomes,
only, is a non-test (Mr. Nousaine please note).
No, it isn't a 'non-test'. It may provide much useful information for
future test design, despite the null outcome in that particular case.
Perish the thought that all those ultra-expensive 'audiophile' cables
actually *do* simply sound the same as cheap 'zipcord'............
--
Stewart Pinkerton | Music is Art - Audio is Engineering
|