Steven Sullivan wrote:
bear wrote:
Steven Sullivan wrote:
http://www.paudio.com/Pages/presenta...ity/sld001.htm
Several problems with it.
It assumes some things not in evidence.
The presentation claims that the 0.4dB ~50% audibility group
is effectively chance. It may well be. But, do we know if
some members of the "test group" reliably were able to detect
this threshold? No. If they were, then the conclusions drawn
are simply wrong.
etc.
First, this is a slideshow summary, not a paper. But I know at least one reader has emailed
Moulton and gotten clarification on data behind one of the points made; I presume you can too.
Second, I think the author qualifies his claims about his results pretty well. He liberally
employs the words 'probably' and 'may be', for example. He addresses the limits of population
sampling. His conclusions (see below) *are* in line with the data.
PS. if we "design to" (as he puts it) the center of the bell
curve of all hearing in the general population we design for
Bose Wave Radio and iPod/MP3, yes??
The Bose Wave and the iPod and MP3s are not equivalent in terms of the sound quality they can
deliver. And does the author really *advocate* 'designing to the center of the bell curve'?
In fact he suggests the opposite (see slide 25) -- where he writes that te commonly accepted
definition for audibility threshold (75%) of loudness change for music (3 dB) 'might NOT be a
reasonable design standard'.
His conclusions seem rather uncontroversial to me. THey a
1) Audibility is a range, not a point (shich woudl seem to address you complaint about
the 0.4dB group, above)
2) Audibility is probabilistic, not absolute (ditto)
3) Audibility is a psychological as much or more than phsyical phenomenon.
and further that
1) there is no perfect data
2) there are no perfect experiments
3) there are no entirely valid conclusions (I would argue this with him, but he may be
referring to psychouacoustic experiments especially, or pe employing a very strict definition
of 'entirely valid')
I don't see that a careful reader of such would come away with the conclusions you fear.
Audiophiles, of course, will blanch at this deduction (even thoughit is again sprinkled with
appropriate qualifiers):
"And as a result, we have now created,a t considerable extra cost, siganl resolutions that
probably unnecessarily exceed by a significant amount any reasonably defined audible limits of
our hearing'
A skeptic must be ready to answer this: Where's the body of counterevidence about
audibility , that supports 'audiophile' claims?
Steven,
There is no need to rehash the past years of discussion on this topic.
The key conclusion that you restate above contains the keyword "unnecessarily":
"signal resolutions that probably unnecessarily exceed...limits...". Clearly,
this would lead the careful reader to the conclusion that I suggest regarding
MP3 and Bose Wave radios, as they are certainly designed based on extensive
statistically valid scientific testing about what people hear or do not, right??
Eg. so as not to unnescessarily exceed said "limits"? This seems to be the true
aim and intent of the "presentation" (as you call it).
Your apparently endless desire to have people respond about some nebulous
"audiophile claim(s)" and supply a body of "counter evidence" appears to be a
troll for a confrontational result which I will not participate in. There is
only ONE "body of evidence." The question is what conclusions can be drawn from
the one single body of evidence extant. About that we apparently disagree.
_-_-bear
___
-S
"As human beings, we understand the world through simile, analogy,
metaphor, narrative and, sometimes, claymation." - B. Mason