[Arny Krueger]

http://www.npr.org/euonline/pub/iboc...der_report.pdf

Thus, in general it appears that listeners made decisions about quality
fairly early in the listening experience (within the first 10-15 seconds)
and did not change their opinions after they listened for extended periods.

[Peter Wieck]

On Sep 4, 11:42 am, "Arny Krueger" wrote:
http://www.npr.org/euonline/pub/iboc...der_report.pdf

Thus, in general it appears that listeners made decisions about quality
fairly early in the listening experience (within the first 10-15 seconds)
and did not change their opinions after they listened for extended periods.

That would entirely depend on what you thought were significant
changes. I saw about what I expected. The most difficult sources
showed the most degredation between long and short-term listening,
with the least changes showing in the 'worst' perceived providers.
But, is 2-6% significant or not?

Keeping in mind that the largest demand for these services will be
voice and voice-over (most unsighted/poorly sighted people of my
acquaintance are quite capable and very adept at operating their own
music systems but cannot read the printed word, of course), and so
including classical and jazz as source materials is something of a red
herring even if interesting. Given the very limited dynamic range of
the human voice *while speaking*, nothing in that study is surprising.

It has been my historical experience that short-term listening (such
as in sales rooms and at first-impression at home) are barely
indicative of long-term results. At best, it serves as the crudest
screening of equipment that might be worth a longer listen. I do find
that a considerable amount of commonly-perceived "good" equipment is
unlistenable to me as for any of several reasons I cannot rest
comfortably in the same room with it when playing. And sometimes quite
mediocre equipment is very listenable without causing any level of
discomfort. These (good)impressions take hours to form in any accurate
way as they must be repeated in order to remove "mood" and "momentary
conditions" from the mix. Bad impressions may take from minutes (in
which case the equipment should never have followed you home after
first-hearing) to days to form as the niggles and discomforts can be
quite subtle.

Peter Wieck
Wyncote, PA

[Arny Krueger]

"Peter Wieck" wrote in message
...
On Sep 4, 11:42 am, "Arny Krueger" wrote:
http://www.npr.org/euonline/pub/iboc...der_report.pdf

Thus, in general it appears that listeners made decisions about quality
fairly early in the listening experience (within the first 10-15 seconds)
and did not change their opinions after they listened for extended
periods.

That would entirely depend on what you thought were significant
changes. I saw about what I expected. The most difficult sources
showed the most degredation between long and short-term listening,
with the least changes showing in the 'worst' perceived providers.
But, is 2-6% significant or not?

The authors gave their well-informed opinion.

Keeping in mind that the largest demand for these services will be
voice and voice-over (most unsighted/poorly sighted people of my
acquaintance are quite capable and very adept at operating their own
music systems but cannot read the printed word, of course), and so
including classical and jazz as source materials is something of a red
herring even if interesting.

I don't think so.

Given the very limited dynamic range of
the human voice *while speaking*, nothing in that study is surprising.

Nothing is a big word.

It has been my historical experience that short-term listening (such
as in sales rooms and at first-impression at home) are barely
indicative of long-term results.

Can you report any tests that you have done with care equal to those in the
cited paper?

Unless you do, you're not exactly comparing apples to apples.

[Peter Wieck]

Arny:

This is a moderated forum, so I will use more gentle language than is
my wont in these cases.

First, I am not disputing the study. I am disputing the conclusions
you draw from it. Put another way, I am disputing the inference you
make from the (to me) rather obvious findings of the study.

The study focused on a service with no direct claims of high-fidelity,
not necessarily even of FM-Broadcast Standards. So, leaping to the
conclusion that this study would be indicative of listeners in an
audio high-fidelity context is just a *bit* of a stretch. Keep in mind
that the participants in this study - users of the transcription
services with established expectations, the sight-impaired, would have
pretty much a two-level response: Good enough/Not good enough.

Again, and absolutely anecdotally, my experience with sight-impaired
and blind individuals is that not only are they quite adept at using
their audio systems, but that their level of sensitivity and
expectations of it are acute. Often to the point of having beloved
items repaired at costs well beyond replacement to maintain the sound
that they desire.

So, once again, I am not in the least bit arguing with the study or
its conclusions.

Peter Wieck
Wyncote, PA

[Lee Derbenwick]

In article ,
Arny Krueger wrote:
http://www.npr.org/euonline/pub/iboc...der_report.pdf

Thus, in general it appears that listeners made decisions about quality
fairly early in the listening experience (within the first 10-15 seconds)
and did not change their opinions after they listened for extended periods.

According to the report, "long samples" meant 35 to 75 seconds.

Unless you're talking about serious ADD, that's not "extended periods"
or "long-term". More accurate terminology would have been to call it
snap judgments versus short-term listening.

--
***** I'm sure I parked my opinions somewhere around here. *****
Lee Derbenwick
Alcatel-Lucent, Westford, MA, USA