I guess #3 and #4 refer to time framed or free space response,
so nothing impossible there.
As a goal #9 is justified IMO.
There are different ways to handle floor bouce e.g.
Cardioid Speakers e.g. are able to excite room modes from local velocity and
pressure maxima as well and are rather universal in placement.
Of course it will be impossible to design a speaker with flat inroom response
below Schroeder frequency independent from room and placement. But
"smooth" as a goal just means "as smooth as possible".
The design of the speaker matters in how average deviation will be,
with respect to room and placement. Thus #9 in a valid point and
IMO an important one.
Last edited:
It seems to me that this is a failure of Ambisonics not a failure of our brain to localize sounds in a non-reflective environment.bolds are mine
the reality of an anechoic chamber is just as ra7 describes it - a disaster, sound in the head
it is not just hypothesis, it was rather painfully tested, what a disappointment...
Ambisonics is a type of multi-speaker matrix encoded surround sound system, and the failure of it to work in the anechoic environment suggests that it fails to simulate some key element of our spatial perception mechanism.
I don't see the relevance of this to perception of the phantom centre channel image in a 2 loudspeaker equilateral triangle playback situation, or how it proves your point ? Two completely different scenarios.
Show me a study testing standard un-encoded 2 channel stereo playback in an anechoic chamber that claims the phantom channel image was unable to form between the speakers and instead appeared inside the head and you might have a point.
Interesting, thanks for posting that. Fits in well with what I've experimentally observed and made mention of in the other thread that spawned this one. Emphasis of high treble can tilt up the apparent source location's angle quite considerably, 20 degrees does seem consistent with what I've observed. Of course the apparent tilt is only there on sounds which have significant high frequency content.The illustration below from Spatial Hearing by Jens Blauert demonstrates the principle:
The picture tells us the ear has trouble distinguishing what spectral information is coming from the source and what is caused by the Pinna and ear canal.
Dan
Application of pulsed tones to test this is interesting, because the HRTF would rely on estimation of the frequency response curve, but with only a single frequency present at a time there can be no "curve". The brain couldn't know whether an individual tone was within a peak or a dip of the curve that represented a given sound direction.
The fact that certain frequencies on their own stimulate a sense of vertical angle for the sound suggests the possibility that perception of vertical height by HRTF is some sorted of weighted mechanism - where different frequencies represent different angles (as shown in the diagram) and in the presence of a broad spectrum sound it's emphasis at particular frequencies within this range that predominately localize the vertical angle of the sound.
Last edited:
"slight" that is...? "from time to time"?
How these quantities relate the time-frame of localization process? We are talking about <0.001 s here - not about from time to time, and this <0.001 s corresponds to <34.4 cm - quite slight for a head movement
are drilling woodpeckers or are we humans?
I was referring to the anechoic room example introduced by ra7,
having a stereo setup presenting a centered phantom source.
ra7 claimed that in this case "in the head localization" will occur.
My point is that even if there would be any ambiguity for the ear/brain
where to place said phantom source, any movement of the head
yields evidence to place it "outside" the head at some distance
(in front).
_______
You know those ambiguous optical illusions, allowing a picture
to be interpreted in two different ways ?
Once the chosen "alternative in perception" has established
it will remain stable (at least for a while).
If you choose to place a human speaker or a phantom source
e.g. about 3m in front of you after some (blind ?) listening, that localization
will remain rather stable if the setting does not change.
No need for permanent "head shaking".
Btw. i wish you would interpret my statements in their given
context, that would make things easier.
We were referring to stereo , even ra7 did refer to stereo in his example.
Changing or deliberately altering the context in a discussion (even for postings of the past)
is rather exhausting IMO.
Last edited:
Maybe we can compare hearing to the way our eyes work. Because there is stereo photography.
Stereoscopy - Wikipedia, the free encyclopedia
You do not need reflections to see depth in the picture with the stereo photograph. The brain can build a 3D image out of two 2D pictures.
It will be the same with stereo sound using two loudspeakers to make a 3D image.
Stereoscopy - Wikipedia, the free encyclopedia
You do not need reflections to see depth in the picture with the stereo photograph. The brain can build a 3D image out of two 2D pictures.
It will be the same with stereo sound using two loudspeakers to make a 3D image.
I've read stuff along these lines before too.
An example is that there is some ambiguity between a sound source directly in front of us and one directly behind us. In this case there is no ITD, so no difference in phase or amplitude, all the brain has to go on is the HRTF. (And maybe non-audio clues, like seeing or not seeing something that looks likely to be the sound source)
High frequencies are greatly diminished for sounds behind us due to the shielding of the outer ear and don't have nearly the same sharp variations in response with angular change that the forward hemisphere has.
Even with an actual real sound source outdoors whether it's directly ahead or behind can be a little bit ambiguous, so the automatic subconscious effect is to turn our head slightly to help localize the sound with the aid of ITD.
Much like the optical illusions you mention (like the candle stick and faces one, or the inside out rotating mask) once the brain decides to interpret an ambiguous input one way or the other, it locks onto that and maintains it until some other contradictory cue comes along.
So in the case of the sound directly behind, you turn your head slightly to one side which lets the brain work out for sure that the sound is behind you, (since left and right is far less ambiguous) and now as far as your brain is concerned this sound is still behind you even after you turn back to your original position, despite returning to an ambiguous input - a reasonable assumption to make because a physical sound emitting object can't just jump from directly behind you to directly in front of you.
You may not even be aware it's happening sometimes, just as you're not always aware of when you blink or take a breath. Any time your head is locked in place in an auditory scene (for example listening to a binaural recording through in-ear earphones, where head movement does nothing to the sound) some aspects of the localization like sounds directly behind can become a bit more vague than in real life where your head has freedom to move, even if the movements are small and subconscious.
Last edited:
yes, I understand Your point and I disagree - not "any movement" in the first place, and secondly how do You think such tests look alike? The listener is seated, and then the test signals start, what makes You think that the listener is then starts to wriggle and turns his head? I believe that He is rather asked not to wriggle but to keep His head still and focus.
Therefore there is really no room for any choice of "alternative in perception"
really no room for any "head shaking"
what makes You think that stereo setup and ambisonics setup are significantly different under such circumstances and in that regard specifically?
yes, I agree but who does it?
I'm pretty sure the head turning is constantly used to identify fore/aft direction. Furthermore, once you know the direction of a source your brain fixes it there, even if things change.
I heard a demo once with two speakers. The operator asked "which speaker is playing" and then you heard a steady sine wave tone turn on. We all pointed to the left speaker so he went over and disconnected the left speaker. Sound still came from that direction.
The trick was that the test tone started in the left speaker and then smoothly panned to the right. With sine wave in a live room you have a hard time telling direction but the turn on transient is easy to detect. Once your brain hears the turn on in the left speaker it ignores the smooth pan to the right.
I would have bet my house that the left speaker was always on.
I think the only way to remove the head moving direction determination would require clamping your head in a vice. It is a subconscious action.
David S.
I heard a demo once with two speakers. The operator asked "which speaker is playing" and then you heard a steady sine wave tone turn on. We all pointed to the left speaker so he went over and disconnected the left speaker. Sound still came from that direction.
The trick was that the test tone started in the left speaker and then smoothly panned to the right. With sine wave in a live room you have a hard time telling direction but the turn on transient is easy to detect. Once your brain hears the turn on in the left speaker it ignores the smooth pan to the right.
I would have bet my house that the left speaker was always on.
I think the only way to remove the head moving direction determination would require clamping your head in a vice. It is a subconscious action.
David S.
Hi David
The transient nature of music is key to good imaging. This is why the first arrivals are so important because it is the transient nature of these first arrivals that are crucial to our assesment of direction. Mess up those very early arrivals (with reflectrions and diffractions) and you have an ambiguous transient direction and imaging will degrade.
"with reflectrions and diffractions" <1 ms? that is before precedence effect comes in?
and >1 ms only if the effect does not work because direct and reflected wavefronts are not sufficiently similar?
For arrivals from reflections and diffractions below the
time delay needed for precedence effect to occur you
get summing localization.
That is phantom sources getting smeared in location and
tend to appear larger and less defined.
Every reflecing or diffracting object adds a certain
pattern to the response and will affect phantom sources
differently according to their specific spectral fingerprint.
Stable and consistent imageing with phantom sources keeping
their places (independently from pitch or kind of sung or spoken
sounds e.g.) calls for speakers and positioning having as little as
possible disturbances from edges or boundaries near the (real)
sound sources.
That is the reason for using baffles without protruding edes,
baffles with large roundoffs, or baffles with absorbing blankets.
It is the reason for integrating tweeters into the baffle without
edges and it is also the reason in some designs
for mounting drivers asymmetrically on the baffle to disperse
diffraction patterns.
time delay needed for precedence effect to occur you
get summing localization.
That is phantom sources getting smeared in location and
tend to appear larger and less defined.
Every reflecing or diffracting object adds a certain
pattern to the response and will affect phantom sources
differently according to their specific spectral fingerprint.
Stable and consistent imageing with phantom sources keeping
their places (independently from pitch or kind of sung or spoken
sounds e.g.) calls for speakers and positioning having as little as
possible disturbances from edges or boundaries near the (real)
sound sources.
That is the reason for using baffles without protruding edes,
baffles with large roundoffs, or baffles with absorbing blankets.
It is the reason for integrating tweeters into the baffle without
edges and it is also the reason in some designs
for mounting drivers asymmetrically on the baffle to disperse
diffraction patterns.
Last edited:
Quite true, but no wide directivity source will do this. Wide horizontal directivities will have large summing localization reflections from the nearby walls, etc. Only a very narrow directivity (< 90 degrees) meets the criteria that you have stated above - unless one uses a lot of damping. But damping at HFs kills the spaciousness so you have a tradeoff between spaciousness OR imaging - you can't have both - you need to compromise. With a narrow directivity you can have both. Leave the room live and avoid the walls with directivity.
Diffusivity of the walls near the speakers seems often more attractive
than damping IMO, given the speaker has sufficiently smooth power response.
For wide radiating speakers also larger distance to the walls is needed ...
Last edited:
As I've commented the narrow directivity approach has worked extremely well for me in a room that would otherwise be pretty problematic. I've used JBL 2397 Smith horns (140 degrees) in the same room and system (no other changes, same 2440 drivers) and can state that they did not image well at all - probably due to early reflections from the nearby walls.
I limit my comments to small rooms which are always "problematic" and one seldom has the option of a "larger distance to the walls".
How large should the distance from the walls be to avoid "summing localization reflections"?
Last edited:
I thought it does no harm if i underline the detrimental effect of early
reflection and diffraction.
Especially since a white paper discussed by yourself in a different thread
reports edge diffraction like beeing an effect which was recently discovered
to be detrimental due to imaging:
Versions which may arrive from elsewhere simply add to the perceived loudness but do not
change the perceived location of the source unless they arrive within the inter-aural delay of about
700 microseconds when the precedence effect breaks down and the perceived direction can be
pulled away from that of the first arriving source by an increase in level. This area is known as the
time-intensity trading region. Once the maximum inter-aural delay is exceeded, the hearing
mechanism knows that the time difference must be due to reverberation and the trading ceases to
change with level.
Unfortunately reflections with delays of the order of 700 microseconds are exactly what are
provided by the legacy rectangular loudspeaker with sharp corners. These reflections are due to
acoustic impedance changes and if we could see sound we would double up with mirth at how
ineptly the sound is being radiated. Effectively the spatial information in the audio signals is being
convolved with the spatial footprint of the speaker. This has the effect of defocusing the image.
Now the effect can be measured.
Taken from: John Watkinson, Putting the Science Back into Loudspeakers
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.