Ping: Arny Krueger

[Jason]

You posted a link to a paper a while back about how very small changes
in a listener's position can lead to very large discrepencies in
perceived sound. I've googled and cannot find the link. Could you re-
post it? Thanks.

Jason
--
reverse my name in email address

[Arny Krueger]

"Jason" > wrote in message

> You posted a link to a paper a while back about how very
> small changes in a listener's position can lead to very
> large discrepencies in perceived sound. I've googled and
> cannot find the link. Could you re- post it? Thanks.

I don't recall that, right now. I'll think about it and see if anything
comes to mind.

It's very well-nown among people who have a more detailed understanding of
acoustics that small changes in position, can strongly affect what you hear
or what you measure.

Every room is full of standing waves. SPL changes rapidly as you move your
sampling device, whether your ears or some test equipment, along any of
them. If you make steady-state measurements at any particular point,
whatever sound level you measure will change dramatically with small changes
in the position or orientation of the microphone. Thermal changes in the air
in the room will move the standing waves around, too.

These variations are one major reason why steady-state measurements at a
point in a room are pretty useless. Move your mic, and its all different.
The usual work-arounds are to use non-steady signals such as impulses,
noise or warble-tones for acoustical measruements.

If you do careful ABX-ing, you will notice that there are signficiant
changes in tonal balance as you move or re-orient your head, just an inch or
two. That people who do sighted evaluations have never mentioned this, tells
you something about the size of the changes that they can notice working the
relatively crude way they do.

[Anahata]

Arny Krueger wrote:
> "Jason" > wrote
>
>>You posted a link to a paper a while back about how very
>>small changes in a listener's position can lead to very
>>large discrepencies in perceived sound. I've googled and
>>cannot find the link. Could you re- post it? Thanks.
>
> I don't recall that, right now. I'll think about it

http://www.ethanwiner.com/believe.html

Might have been it. Not exactly a "paper" but quite detailed and relevant.

Anahata

[Ethan Winer]

Jason,

> You posted a link to a paper a while back about how very small changes in
> a listener's position can lead to very large discrepencies in perceived
> sound.

Anahata gave you what may be the link you meant, to my article.

--Ethan

[Jason]

In article >, "Ethan
Winer" <ethanw at ethanwiner dot com> says...
> Jason,
>
> > You posted a link to a paper a while back about how very small changes in
> > a listener's position can lead to very large discrepencies in perceived
> > sound.
>
> Anahata gave you what may be the link you meant, to my article.
>
> --Ethan
>
>
Thanks, Ethan - right you are! No wonder I couldn't find it with
google... : )

And thanks Anahata.

Jason


--
reverse my name in email address

[Mark]

On Dec 17, 9:34 pm, Jason > wrote:
> In article >, "Ethan
> Winer" <ethanw at ethanwiner dot com> says...> Jason,
>
> > > You posted a link to a paper a while back about how very small changes in
> > > a listener's position can lead to very large discrepencies in perceived
> > > sound.
>
> > Anahata gave you what may be the link you meant, to my article.
>
> > --Ethan
>
> Thanks, Ethan - right you are! No wonder I couldn't find it with
> google... : )
>
> And thanks Anahata.
>
> Jason
>
> --
> reverse my name in email address

get some white noise or pink noise source and listen to it via
speakers and move your head around a little

it will make you wonder why we even worry about little things..

Mark

[William Sommerwerck]

> You posted a link to a paper a while back about how
> very small changes in a listener's position can lead
> to very large discrepencies in perceived sound.

I've wanted to respond to this, and have been thinking about the best way to
do so.

I've been listening to "good" audio systems for over 35 years, and I do not
find this to be the case.

The "sound" of my system changes only a little as I move around the room.
Its basic character remains the same. The most-notable differences are in
(duh) imaging, and overall tonal balance.

To argue that small head shifts, slight changes in seating position, etc,
are a significant factor in biasing or invalidating subjective reviews is to
look in the wrong direction for explanations.

[Paul Stamler]

"William Sommerwerck" > wrote in message
. ..
> > You posted a link to a paper a while back about how
> > very small changes in a listener's position can lead
> > to very large discrepencies in perceived sound.
>
> I've wanted to respond to this, and have been thinking about the best way
to
> do so.
>
> I've been listening to "good" audio systems for over 35 years, and I do
not
> find this to be the case.
>
> The "sound" of my system changes only a little as I move around the room.
> Its basic character remains the same. The most-notable differences are in
> (duh) imaging, and overall tonal balance.
>
> To argue that small head shifts, slight changes in seating position, etc,
> are a significant factor in biasing or invalidating subjective reviews is
to
> look in the wrong direction for explanations.

I agree with William on this, but wouldn't dismiss the question. There's an
old story about a cub reporter sent to cover a society wedding; he returned
to the newspaper office and threw his notebook down on his desk. "What's the
matter?" asked his editor. "There's no story," the cub reporter replied.
"The groom didn't show up."

Here's why I think the head-position stuff is interesting: if you move your
head a few inches at your listening position, you will hear small changes in
frequency response. But if you put measuring mics in the position your ears
would occupy, measure the response, then move them around, you will find
HUGE changes in frequency response.

I think the question of interest is not whether differences heard in
subjective reviews are due to small head movements, which would be very hard
to prove one way or the other at this point. I think the real question is
this: how the hell does our ear-brain mechanism turn huge measured frequency
response changes into quite small perceived ones? My bet is that binaural
hearing has a lot to do with it; Scott's trick of plugging one ear to find a
good mic position might indicate that this is the case. Beyond that I don't
know, and I suspect no one has investigated the issue. It's also possible
(pure speculation) that by moving our head around as we listen, we're taking
an average of the sound in our vicinity, and our brain is cancelling out the
variations in the same way one might average an array of microphones.

Peace,
Paul

[Ethan Winer]

William,

> The "sound" of my system changes only a little as I move around the room.

Bass traps reduce the change over distance at low frequencies, and having
absorption at the first reflection points reduces the changes at higher
frequencies. But my graphs don't lie. Even in a treated room the response
changes quite a lot due to speaker beaming and comb filtering from different
arrival times for mono content. The reason the sound doesn't SEEM to change
as much as the graphs would imply is because we have two ears. Paul Stamler
nailed it (BTW, nice article in AudioXpress) with this comment :

My bet is that binaural hearing has a lot to do with it;
Scott's trick of plugging one ear to find a good mic
position might indicate that this is the case.

Nulls are more damaging than peaks, and frequencies missing in one ear are
usually present at the other given the large change in response even a few
inches away.

> To argue that small head shifts, slight changes in seating position, etc,
> are a significant factor in biasing or invalidating subjective reviews is
> to look in the wrong direction for explanations.

Agreed. The MAIN reason audiophile reviewers (and listeners) believe the
sound changed - always for the better, of course! - with a new power cord is
self-delusion. The real issue is therefore psychology, not audio science.

--Ethan

[William Sommerwerck]

"Ethan Winer" <ethanw at ethanwiner dot com> wrote in message
...
> William,

>> The "sound" of my system changes only a little as I move around the room.

> Bass traps reduce the change over distance at low frequencies, and having
> absorption at the first reflection points reduces the changes at higher
> frequencies. But my graphs don't lie. Even in a treated room the response
> changes quite a lot due to speaker beaming and comb filtering from
different
> arrival times for mono content. The reason the sound doesn't SEEM to
change
> as much as the graphs would imply is because we have two ears. Paul
Stamler
> nailed it (BTW, nice article in AudioXpress) with this comment :

> My bet is that binaural hearing has a lot to do with it;
> Scott's trick of plugging one ear to find a good mic
> position might indicate that this is the case.

> Nulls are more damaging than peaks, and frequencies missing in one ear are
> usually present at the other given the large change in response even a few
> inches away.

This is probably correct. There is also the fact that standing waves do not
build up the same way with (relatively short) musical notes as they do with
continuous test signals.


>> To argue that small head shifts, slight changes in seating position, etc,
>> are a significant factor in biasing or invalidating subjective reviews is
>> to look in the wrong direction for explanations.

> Agreed. The MAIN reason audiophile reviewers (and listeners) believe the
> sound changed -- always for the better, of course! -- with a new power
cord
> is self-delusion. The real issue is therefore psychology, not audio
science.

Perhaps. I think something more psychologically subtle and more-poorly
understood than "self-delusion" is at work, but "psychology" certainly
accounts for a goodly percentage of subjective differences.

[Arny Krueger]

"William Sommerwerck" > wrote in
message
>> You posted a link to a paper a while back about how
>> very small changes in a listener's position can lead
>> to very large discrepencies in perceived sound.
>
> I've wanted to respond to this, and have been thinking
> about the best way to do so.
>
> I've been listening to "good" audio systems for over 35
> years, and I do not find this to be the case.
>
> The "sound" of my system changes only a little as I move
> around the room. Its basic character remains the same.
> The most-notable differences are in (duh) imaging, and
> overall tonal balance.
>
> To argue that small head shifts, slight changes in
> seating position, etc, are a significant factor in
> biasing or invalidating subjective reviews is to look in
> the wrong direction for explanations.

That's not quite the point I was trying to make.

Compared to the differences between some audio products that I've found to
reliably sound different, audible differences due to head shifts can be of a
similar nature.

How does one reliably tell the difference between them? Well, you can clamp
your head in a vise! ;-)

In a DBT, the inherent rejection of false positives means that you have
reliable detection of false audible differences due to head shifts. They can
dilute your results, but they won't give you a false postive.

In ABX tests, not only are false positives due to head shifts rejected, but
audible differences due to head shifts can be reliably identified by the
listener during the test. This facilitates the listener disregarding them
during the test, thus increasing the sensitivity of the test to small
differences.

Thus, more sensitive results than can be more easily obtained in an ABX test
than in a sighted listening evaluation.

[Arny Krueger]

"Paul Stamler" > wrote in message


> Here's why I think the head-position stuff is
> interesting: if you move your head a few inches at your
> listening position, you will hear small changes in
> frequency response. But if you put measuring mics in the
> position your ears would occupy, measure the response,
> then move them around, you will find HUGE changes in
> frequency response.

Yeah, but if you weight those changes perceptually, they usually become
smaller.

> I think the question of interest is not whether
> differences heard in subjective reviews are due to small
> head movements, which would be very hard to prove one way
> or the other at this point.

Oh, I think a lot of that happens. Sighted evaluations have very few
safeguards against false positives.

> I think the real question is
> this: how the hell does our ear-brain mechanism turn huge
> measured frequency response changes into quite small
> perceived ones?

Well, that's a good question too! If human hearing was too susceptible to
false differences due to head shifts, our ancestors might have been more
frequently gobbled up by tigers, because they got tired of jumping due to
false positives due to head shifts. Sometimes not jumping is a survival
skill.

> My bet is that binaural hearing has a lot
> to do with it; Scott's trick of plugging one ear to find
> a good mic position might indicate that this is the case.

I agree that the difference between one eared hearing and two eared hearing
is a big explanation. But, I don't think that Scott orginated using
one-eared hearing during recording.

> Beyond that I don't know, and I suspect no one has
> investigated the issue.

I don't know about that. Some of the large measurable differences affect
narrow bands, and the sizing of critical bands can explain some of the ear's
reduction of their perception.

> It's also possible (pure
> speculation) that by moving our head around as we listen,
> we're taking an average of the sound in our vicinity, and
> our brain is cancelling out the variations in the same
> way one might average an array of microphones.

I long ago learned to move my head around to increase my sensitivity to
small differences in audio products during ABX tests involving listening
through loudspeakers.

[Paul Stamler]

"Arny Krueger" > wrote in message
...
>
> > My bet is that binaural hearing has a lot
> > to do with it; Scott's trick of plugging one ear to find
> > a good mic position might indicate that this is the case.
>
> I agree that the difference between one eared hearing and two eared
hearing
> is a big explanation. But, I don't think that Scott orginated using
> one-eared hearing during recording.

I know, but these days he's probably its best-known advocate. I first read
it in a column by Bert Whyte back in the 1960s.

Peace,
Paul

[Scott Dorsey]

Paul Stamler > wrote:
>"Arny Krueger" > wrote in message
...
>>
>> > My bet is that binaural hearing has a lot
>> > to do with it; Scott's trick of plugging one ear to find
>> > a good mic position might indicate that this is the case.
>>
>> I agree that the difference between one eared hearing and two eared
>hearing
>> is a big explanation. But, I don't think that Scott orginated using
>> one-eared hearing during recording.
>
>I know, but these days he's probably its best-known advocate. I first read
>it in a column by Bert Whyte back in the 1960s.

I learned about it from Gabe Weiner, and I think he deserves a lot of
credit for it. Bert Whyte already has enough things credited to him,
and Gabe didn't live quite so long... and he surely did more to popularize
it than I ever did.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Mark]

>
> Nulls are more damaging than peaks, and frequencies missing in one ear are
> usually present at the other given the large change in response even a few
> inches away.
>
>
Really, I have always read it was the other way round?
i.e. peaks are more damaging than nulls.

Mark

[Don Pearce]

On Tue, 18 Dec 2007 15:57:19 -0800 (PST), Mark >
wrote:

>
>>
>> Nulls are more damaging than peaks, and frequencies missing in one ear are
>> usually present at the other given the large change in response even a few
>> inches away.
>>
>>
>Really, I have always read it was the other way round?
>i.e. peaks are more damaging than nulls.
>
>Mark

No, peaks rarely go more than three dB or so. Nulls can easily be
-30dB. They really screw a mix.

d

--
Pearce Consulting
http://www.pearce.uk.com

[Ethan Winer]

Mark,

> Really, I have always read it was the other way round?
> i.e. peaks are more damaging than nulls.

Yes, that's the conventional wisdom, but it's wrong IMO. Or at least wrong
in most cases. As Don said, nulls are much more severe than peaks, and this
causes what's probably the most common newbie "not translating" complaint.
Most people report their mixes sound good in their room, but too boomy
elsewhere. Meaning they hear too little bass while mixing. Sometimes it's
the other way around, but not usually. In a typical bedroom studio, the
worst nulls are caused by quarter wavelength cancellation off the rear wall
behind you. This program on my company's site helps to distinguish modal
from non-modal peaks and nulls:

http://www.realtraps.com/sbirlbir.htm

--Ethan

[Laurence Payne]

On Tue, 18 Dec 2007 15:40:40 GMT, "Paul Stamler"
> wrote:

>Here's why I think the head-position stuff is interesting: if you move your
>head a few inches at your listening position, you will hear small changes in
>frequency response. But if you put measuring mics in the position your ears
>would occupy, measure the response, then move them around, you will find
>HUGE changes in frequency response.


I think the aural differences CAN be huge, but we're very practiced in
compensating for them.